NFL + College FBS

Football

NFL + college FBS spreads and totals from a matchup-exploitation model: every unit is percentiled within its league, strong units are matched against weak opposing units (with an offensive-line layer and game-time weather), and the projection is anchored 65% to the de-vigged market. Picks are paper-only until they show positive closing-line value vs the sharp close. Records are always broken out by league and market — never blended.

Paper / simulatedLive O/U →

Paper records · by league and market

CFB totalsWins–losses–pushes on committed CFB total paper picks (model matchup_v1). Exactly on the number = push, stake returned. Flat ROI -13.2% · avg CLV -0.24pp on 11 tracked.2-2-0

Football board

Sat, Sep 5, 2026

AllNFLCollege

Coastal Carolina @ West Virginia · CFB · week 1

Under 54.5 (total) -105 (1.95)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile
EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+8.2%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+7.2%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.55.4%
MarketThe market's fair probability for this side — the price with the vig removed.48.8%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).36
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.00pp
Projected Coastal Carolina 20 — West Virginia 29 · margin +8.9 · total 48.1 · stake 0.8u
Why this pick · matchup breakdown

Net edge +49.4 (Weak O vs strong D)

  • West Virginia pass offense (20th pct) vs Coastal Carolina defense (35th pct): edge -15
  • West Virginia rush offense (26th pct) vs Coastal Carolina defense (14th pct): edge +13
  • Coastal Carolina pass offense (5th pct) vs West Virginia defense (52nd pct): edge -47
  • Coastal Carolina rush offense (23rd pct) vs West Virginia defense (74th pct): edge -51
  • West Virginia OL (-10.0 pts): run-block proxy 3.69 ypc
  • Coastal Carolina OL (-9.6 pts): run-block proxy 4.31 ypc
  • 73F, wind 6mph at kickoff
  • Line movement: sharp consensus 1.2pp against over vs draftkings (line 54.0 vs 54.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp against over vs draftkings (line 54.0 vs 54.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 54.5 (-115/-105) · sharp consensus 54.0 across 3 books · market mean 54.0 (model off by 5.9) · blend 35/65 model/market

Kent State @ South Carolina · CFB · week 1

Under 53.5 (total) -105 (1.95)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile
EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+7.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+6.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.54.9%
MarketThe market's fair probability for this side — the price with the vig removed.48.8%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).50
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+1.16pp
Projected Kent State 10 — South Carolina 38 · margin +28.2 · total 47.7 · stake 0.8u
Why this pick · matchup breakdown

Net edge +89.6 (Dual edge) — South Carolina holds both the pass and rush matchup, the highest-conviction spot.

  • South Carolina pass offense (45th pct) vs Kent State defense (25th pct): edge +20
  • South Carolina rush offense (31st pct) vs Kent State defense (20th pct): edge +10
  • Kent State pass offense (31st pct) vs South Carolina defense (60th pct): edge -29
  • Kent State rush offense (0th pct) vs South Carolina defense (91st pct): edge -91
  • South Carolina OL (-10.0 pts): run-block proxy 3.28 ypc
  • Kent State OL (-10.0 pts): run-block proxy 2.99 ypc
  • 96F, wind 5mph at kickoff
  • Line movement: sharp consensus 1.2pp against over vs draftkings (line 53.5 vs 53.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp against over vs draftkings (line 53.5 vs 53.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 53.5 (-115/-105) · sharp consensus 53.5 across 3 books · market mean 53.5 (model off by 5.8) · blend 35/65 model/market

Florida Atlantic @ Florida · CFB · week 1

Under 59.5 (total) -108 (1.93)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile
EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+6.0%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+5.0%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.55.1%
MarketThe market's fair probability for this side — the price with the vig removed.49.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).35
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+1.75pp
Projected Florida Atlantic 25 — Florida 29 · margin +4.4 · total 54.0 · stake 0.8u
Why this pick · matchup breakdown

Net edge +47.6 (Weak O vs strong D) Turnover pairing: Florida's high-takeaway defense meets a turnover-prone offense.

  • Florida pass offense (39th pct) vs Florida Atlantic defense (52nd pct): edge -13
  • Florida rush offense (48th pct) vs Florida Atlantic defense (14th pct): edge +34
  • Florida Atlantic pass offense (53rd pct) vs Florida defense (69th pct): edge -16
  • Florida Atlantic rush offense (12th pct) vs Florida defense (62nd pct): edge -50
  • Florida OL (-10.0 pts): run-block proxy 4.27 ypc
  • Florida Atlantic OL (-10.0 pts): run-block proxy 3.56 ypc
  • 88F, wind 6mph at kickoff
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 59.5 (-112/-108) · sharp consensus 59.5 across 3 books · market mean 59.5 (model off by 5.5) · blend 35/65 model/market

East Carolina @ Alabama · CFB · week 1

Under 52.5 (total) -105 (1.95)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile
EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+5.9%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+5.0%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.54.3%
MarketThe market's fair probability for this side — the price with the vig removed.48.8%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).20
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+1.16pp
Projected East Carolina 20 — Alabama 28 · margin +8.3 · total 47.4 · stake 0.8u
Why this pick · matchup breakdown

Net edge +4.4 (Weak O vs strong D)

  • Alabama pass offense (71st pct) vs East Carolina defense (49th pct): edge +22
  • Alabama rush offense (27th pct) vs East Carolina defense (77th pct): edge -50
  • East Carolina pass offense (55th pct) vs Alabama defense (62nd pct): edge -7
  • East Carolina rush offense (38th pct) vs Alabama defense (71st pct): edge -34
  • Alabama OL (-10.0 pts): run-block proxy 3.36 ypc
  • East Carolina OL (-10.0 pts): run-block proxy 4.10 ypc
  • 86F, wind 4mph at kickoff
  • Line movement: sharp consensus 1.2pp against over vs draftkings (line 52.5 vs 52.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp against over vs draftkings (line 52.5 vs 52.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 52.5 (-115/-105) · sharp consensus 52.5 across 3 books · market mean 52.5 (model off by 5.2) · blend 35/65 model/market

Tulane @ Duke · CFB · week 1

Tulane +7.5 (spread) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Strong O vs weak DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile
EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+5.8%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+4.8%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.55.4%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).19
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+1.16pp
Projected Tulane 29 — Duke 30 · margin +1.1 · total 59.3 · stake 0.8u
Why this pick · matchup breakdown

Net edge +2.5 (Strong O vs weak D)

  • Duke pass offense (71st pct) vs Tulane defense (50th pct): edge +21
  • Duke rush offense (39th pct) vs Tulane defense (48th pct): edge -8
  • Tulane pass offense (54th pct) vs Duke defense (34th pct): edge +20
  • Tulane rush offense (45th pct) vs Duke defense (57th pct): edge -12
  • Duke OL (-10.0 pts): run-block proxy 4.26 ypc
  • Tulane OL (-3.4 pts): run-block proxy 4.53 ypc
  • Line movement: sharp consensus 1.2pp against home vs draftkings (line -7.5 vs -7.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp against home vs draftkings (line -7.5 vs -7.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -7.5 (-110/-110) · sharp consensus -7.5 across 3 books · market mean 7.0 (model off by 5.9) · blend 35/65 model/market

North Texas @ Indiana · CFB · week 1

Over 55.5 (total) -115 (1.87)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile
EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+5.7%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+4.7%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.56.6%
MarketThe market's fair probability for this side — the price with the vig removed.51.2%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).49
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.16pp
Projected North Texas 27 — Indiana 35 · margin +8.4 · total 61.5 · stake 0.8u
Why this pick · matchup breakdown

Net edge +86.5 (Dual edge) — Indiana holds both the pass and rush matchup, the highest-conviction spot.

  • Indiana pass offense (100th pct) vs North Texas defense (38th pct): edge +62
  • Indiana rush offense (100th pct) vs North Texas defense (19th pct): edge +81
  • North Texas pass offense (85th pct) vs Indiana defense (72nd pct): edge +13
  • North Texas rush offense (80th pct) vs Indiana defense (96th pct): edge -16
  • Indiana OL (+10.0 pts): run-block proxy 5.35 ypc
  • North Texas OL (+10.0 pts): run-block proxy 5.30 ypc
  • 76F, wind 8mph at kickoff
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 55.5 (-115/-105) · sharp consensus 55.5 across 3 books · market mean 55.9 (model off by 5.6) · blend 35/65 model/market

Clemson @ LSU · CFB · week 1

Under 51.5 (total) -118 (1.85)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile
EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+5.2%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+4.2%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.57.0%
MarketThe market's fair probability for this side — the price with the vig removed.51.9%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).25
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.86pp
Projected Clemson 22 — LSU 23 · margin +0.9 · total 45.6 · stake 1.0u
Why this pick · matchup breakdown

Net edge -9.8 (Weak O vs strong D)

  • LSU pass offense (50th pct) vs Clemson defense (66th pct): edge -16
  • LSU rush offense (28th pct) vs Clemson defense (78th pct): edge -50
  • Clemson pass offense (60th pct) vs LSU defense (62nd pct): edge -2
  • Clemson rush offense (38th pct) vs LSU defense (83rd pct): edge -45
  • LSU OL (-10.0 pts): run-block proxy 3.66 ypc
  • Clemson OL (-10.0 pts): run-block proxy 3.95 ypc
  • 88F, wind 1mph at kickoff
  • Line movement: sharp consensus 1.9pp toward over vs draftkings (line 51.0 vs 51.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.9pp toward over vs draftkings (line 51.0 vs 51.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 51.5 (-102/-118) · sharp consensus 51.0 across 3 books · market mean 51.0 (model off by 5.4) · blend 35/65 model/market

Liberty @ James Madison · CFB · week 1

James Madison -7.0 (spread) -105 (1.95)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile
EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+5.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+4.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.52.6%
MarketThe market's fair probability for this side — the price with the vig removed.48.8%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).46
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+2.35pp
Projected Liberty 20 — James Madison 32 · margin +11.4 · total 51.8 · stake 0.8u
Why this pick · matchup breakdown

Net edge +78.1 (Dual edge) — James Madison holds both the pass and rush matchup, the highest-conviction spot.

  • James Madison pass offense (67th pct) vs Liberty defense (46th pct): edge +21
  • James Madison rush offense (91st pct) vs Liberty defense (20th pct): edge +71
  • Liberty pass offense (35th pct) vs James Madison defense (65th pct): edge -30
  • Liberty rush offense (62nd pct) vs James Madison defense (79th pct): edge -17
  • James Madison OL (+10.0 pts): run-block proxy 5.71 ypc
  • Liberty OL (-0.5 pts): run-block proxy 4.63 ypc
  • Line movement: sharp consensus 1.6pp toward home vs draftkings (line -7.0 vs -7.0)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.6pp toward home vs draftkings (line -7.0 vs -7.0)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -7.0 (-105/-115) · sharp consensus -7.0 across 3 books · market mean 7.2 (model off by 4.2) · blend 35/65 model/market

Western Kentucky @ Nevada · CFB · week 1

Western Kentucky -1.5 (spread) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile
EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+4.9%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+3.9%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.54.9%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).29
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.16pp
Projected Western Kentucky 26 — Nevada 19 · margin -7.3 · total 45.2 · stake 0.8u
Why this pick · matchup breakdown

Net edge -31.1 (Weak O vs strong D)

  • Nevada pass offense (7th pct) vs Western Kentucky defense (51st pct): edge -44
  • Nevada rush offense (35th pct) vs Western Kentucky defense (46th pct): edge -11
  • Western Kentucky pass offense (43rd pct) vs Nevada defense (49th pct): edge -6
  • Western Kentucky rush offense (26th pct) vs Nevada defense (16th pct): edge +10
  • Nevada OL (-4.6 pts): run-block proxy 4.49 ypc
  • Western Kentucky OL (-10.0 pts): run-block proxy 4.24 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 1.5 (-110/-110) · sharp consensus 1.5 across 3 books · market mean -1.5 (model off by 5.8) · blend 35/65 model/market

Texas State @ Texas · CFB · week 1

Under 59.5 (total) -108 (1.93)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Strong O vs weak DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile
EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+4.2%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+3.2%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.54.1%
MarketThe market's fair probability for this side — the price with the vig removed.49.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).20
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.46pp
Projected Texas State 28 — Texas 27 · margin -0.6 · total 55.0 · stake 0.8u
Why this pick · matchup breakdown

Net edge +6.3 (Strong O vs weak D)

  • Texas pass offense (71st pct) vs Texas State defense (40th pct): edge +31
  • Texas rush offense (45th pct) vs Texas State defense (54th pct): edge -9
  • Texas State pass offense (77th pct) vs Texas defense (72nd pct): edge +5
  • Texas State rush offense (87th pct) vs Texas defense (86th pct): edge +0
  • Texas OL (-10.0 pts): run-block proxy 4.24 ypc
  • Texas State OL (+10.0 pts): run-block proxy 5.19 ypc
  • 98F, wind 4mph at kickoff
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 59.5 (-112/-108) · sharp consensus 60.0 across 3 books · market mean 60.0 (model off by 5.0) · blend 35/65 model/market

Oregon State @ Houston · CFB · week 1

Under 51.5 (total) -115 (1.87)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile
EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+4.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+3.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.55.7%
MarketThe market's fair probability for this side — the price with the vig removed.51.2%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).37
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.87pp
Projected Oregon State 20 — Houston 27 · margin +7.0 · total 46.5 · stake 0.8u
Why this pick · matchup breakdown

Net edge +53.0 (Weak O vs strong D)

  • Houston pass offense (56th pct) vs Oregon State defense (40th pct): edge +16
  • Houston rush offense (40th pct) vs Oregon State defense (47th pct): edge -7
  • Oregon State pass offense (19th pct) vs Houston defense (62nd pct): edge -42
  • Oregon State rush offense (15th pct) vs Houston defense (71st pct): edge -56
  • Houston OL (-10.0 pts): run-block proxy 4.26 ypc
  • Oregon State OL (-10.0 pts): run-block proxy 3.75 ypc
  • 90F, wind 3mph at kickoff
  • Line movement: sharp consensus 1.2pp toward over vs draftkings (line 51.0 vs 51.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp toward over vs draftkings (line 51.0 vs 51.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 51.5 (-105/-115) · sharp consensus 51.0 across 3 books · market mean 51.0 (model off by 4.5) · blend 35/65 model/market

Analyzed · below threshold or guardrail-held

49 markets

Oklahoma State @ Tulsa · CFB · week 1

Tulsa +13.5 (spread) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +8.5 vs market mean -13.8 (22.3 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+23.2%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+22.2%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.64.5%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).23
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.87pp
Projected Oklahoma State 20 — Tulsa 28 · margin +8.5 · total 48.2
Why this pick · matchup breakdown

Net edge +13.8 (Weak O vs strong D)

  • Tulsa pass offense (24th pct) vs Oklahoma State defense (45th pct): edge -21
  • Tulsa rush offense (36th pct) vs Oklahoma State defense (48th pct): edge -12
  • Oklahoma State pass offense (17th pct) vs Tulsa defense (43rd pct): edge -26
  • Oklahoma State rush offense (19th pct) vs Tulsa defense (54th pct): edge -35
  • Tulsa OL (-8.4 pts): run-block proxy 4.35 ypc
  • Oklahoma State OL (-10.0 pts): run-block proxy 3.53 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 13.5 (-110/-110) · sharp consensus 13.5 across 3 books · market mean -13.8 (model off by 22.3) · blend 35/65 model/market

East Carolina @ Alabama · CFB · week 1

East Carolina +27.5 (spread) -105 (1.95)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +8.3 vs market mean +27.8 (19.5 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+22.4%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+21.4%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.62.7%
MarketThe market's fair probability for this side — the price with the vig removed.48.8%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).20
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.01pp
Projected East Carolina 20 — Alabama 28 · margin +8.3 · total 47.4
Why this pick · matchup breakdown

Net edge +4.4 (Weak O vs strong D)

  • Alabama pass offense (71st pct) vs East Carolina defense (49th pct): edge +22
  • Alabama rush offense (27th pct) vs East Carolina defense (77th pct): edge -50
  • East Carolina pass offense (55th pct) vs Alabama defense (62nd pct): edge -7
  • East Carolina rush offense (38th pct) vs Alabama defense (71st pct): edge -34
  • Alabama OL (-10.0 pts): run-block proxy 3.36 ypc
  • East Carolina OL (-10.0 pts): run-block proxy 4.10 ypc
  • Line movement: sharp consensus 1.6pp against home vs draftkings (line -28.0 vs -27.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.6pp against home vs draftkings (line -28.0 vs -27.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -27.5 (-115/-105) · sharp consensus -28.0 across 3 books · market mean 27.8 (model off by 19.5) · blend 35/65 model/market

Texas State @ Texas · CFB · week 1

Texas State +29.5 (spread) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Strong O vs weak DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +8.7 vs market mean +29.5 (20.7 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+22.4%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+21.4%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.64.1%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).20
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-2.32pp
Projected Texas State 23 — Texas 32 · margin +8.7 · total 55.0
Why this pick · matchup breakdown

Net edge +6.3 (Strong O vs weak D)

  • Texas pass offense (71st pct) vs Texas State defense (40th pct): edge +31
  • Texas rush offense (45th pct) vs Texas State defense (54th pct): edge -9
  • Texas State pass offense (77th pct) vs Texas defense (72nd pct): edge +5
  • Texas State rush offense (87th pct) vs Texas defense (86th pct): edge +0
  • Texas OL (-10.0 pts): run-block proxy 4.24 ypc
  • Texas State OL (+10.0 pts): run-block proxy 5.19 ypc
  • Line movement: sharp consensus 1.2pp toward home vs draftkings (line -29.0 vs -29.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp toward home vs draftkings (line -29.0 vs -29.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -29.5 (-110/-110) · sharp consensus -29.0 across 3 books · market mean 29.5 (model off by 20.7) · blend 35/65 model/market

Ohio @ Nebraska · CFB · week 1

Ohio +23.5 (spread) -105 (1.95)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Strong O vs weak DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +5.9 vs market mean +24.0 (18.1 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+21.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+20.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.62.0%
MarketThe market's fair probability for this side — the price with the vig removed.48.8%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).25
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-2.35pp
Projected Ohio 25 — Nebraska 31 · margin +5.9 · total 55.7
Why this pick · matchup breakdown

Net edge -19.6 (Strong O vs weak D)

  • Nebraska pass offense (55th pct) vs Ohio defense (42nd pct): edge +13
  • Nebraska rush offense (50th pct) vs Ohio defense (47th pct): edge +3
  • Ohio pass offense (46th pct) vs Nebraska defense (51st pct): edge -6
  • Ohio rush offense (80th pct) vs Nebraska defense (25th pct): edge +54
  • Nebraska OL (-9.4 pts): run-block proxy 4.32 ypc
  • Ohio OL (+10.0 pts): run-block proxy 5.37 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -23.5 (-115/-105) · sharp consensus -23.5 across 3 books · market mean 24.0 (model off by 18.1) · blend 35/65 model/market

Western Michigan @ Michigan · CFB · week 1

Western Michigan +27.5 (spread) -105 (1.95)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +11.7 vs market mean +28.1 (16.4 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+19.3%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+18.3%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.61.1%
MarketThe market's fair probability for this side — the price with the vig removed.48.8%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).40
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.43pp
Projected Western Michigan 18 — Michigan 29 · margin +11.7 · total 47.2
Why this pick · matchup breakdown

Net edge +59.9 (Weak O vs strong D)

  • Michigan pass offense (57th pct) vs Western Michigan defense (57th pct): edge +0
  • Michigan rush offense (100th pct) vs Western Michigan defense (54th pct): edge +46
  • Western Michigan pass offense (16th pct) vs Michigan defense (67th pct): edge -50
  • Western Michigan rush offense (58th pct) vs Michigan defense (79th pct): edge -22
  • Michigan OL (+10.0 pts): run-block proxy 5.52 ypc
  • Western Michigan OL (-0.7 pts): run-block proxy 4.63 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -27.5 (-115/-105) · sharp consensus -27.5 across 3 books · market mean 28.1 (model off by 16.4) · blend 35/65 model/market

Missouri State @ Texas A&M · CFB · week 1

Missouri State +40.5 (spread) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: divergence guard: model +24.3 vs market mean +40.0 (15.7 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+18.5%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+17.5%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.62.1%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).54
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+1.16pp
Projected Missouri State 16 — Texas A&M 40 · margin +24.3 · total 56.4
Why this pick · matchup breakdown

Net edge +130.1 (Dual edge) — Texas A&M holds both the pass and rush matchup, the highest-conviction spot.

  • Texas A&M pass offense (90th pct) vs Missouri State defense (32nd pct): edge +59
  • Texas A&M rush offense (87th pct) vs Missouri State defense (12th pct): edge +75
  • Missouri State pass offense (52nd pct) vs Texas A&M defense (92nd pct): edge -40
  • Missouri State rush offense (5th pct) vs Texas A&M defense (54th pct): edge -50
  • Texas A&M OL (+6.0 pts): run-block proxy 4.86 ypc
  • Missouri State OL (-10.0 pts): run-block proxy 3.60 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -40.5 (-110/-110) · sharp consensus -40.0 across 3 books · market mean 40.0 (model off by 15.7) · blend 35/65 model/market

North Texas @ Indiana · CFB · week 1

North Texas +40.5 (spread) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: divergence guard: model +26.8 vs market mean +40.0 (13.1 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+15.7%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+14.7%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.60.6%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).49
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-0.70pp
Projected North Texas 17 — Indiana 44 · margin +26.8 · total 61.5
Why this pick · matchup breakdown

Net edge +86.5 (Dual edge) — Indiana holds both the pass and rush matchup, the highest-conviction spot.

  • Indiana pass offense (100th pct) vs North Texas defense (38th pct): edge +62
  • Indiana rush offense (100th pct) vs North Texas defense (19th pct): edge +81
  • North Texas pass offense (85th pct) vs Indiana defense (72nd pct): edge +13
  • North Texas rush offense (80th pct) vs Indiana defense (96th pct): edge -16
  • Indiana OL (+10.0 pts): run-block proxy 5.35 ypc
  • North Texas OL (+10.0 pts): run-block proxy 5.30 ypc
  • Line movement: sharp consensus 1.2pp toward home vs draftkings (line -39.5 vs -40.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp toward home vs draftkings (line -39.5 vs -40.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -40.5 (-110/-110) · sharp consensus -39.5 across 3 books · market mean 40.0 (model off by 13.1) · blend 35/65 model/market

Oklahoma State @ Tulsa · CFB · week 1

Under 59.5 (total) -115 (1.87)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +48.2 vs market mean +59.0 (10.8 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+13.9%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+12.9%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.60.9%
MarketThe market's fair probability for this side — the price with the vig removed.51.2%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).21
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+1.16pp
Projected Oklahoma State 20 — Tulsa 28 · margin +8.5 · total 48.2
Why this pick · matchup breakdown

Net edge +13.8 (Weak O vs strong D)

  • Tulsa pass offense (24th pct) vs Oklahoma State defense (45th pct): edge -21
  • Tulsa rush offense (36th pct) vs Oklahoma State defense (48th pct): edge -12
  • Oklahoma State pass offense (17th pct) vs Tulsa defense (43rd pct): edge -26
  • Oklahoma State rush offense (19th pct) vs Tulsa defense (54th pct): edge -35
  • Tulsa OL (-8.4 pts): run-block proxy 4.35 ypc
  • Oklahoma State OL (-10.0 pts): run-block proxy 3.53 ypc
  • weather feed unavailable
  • Line movement: sharp consensus 1.2pp toward over vs draftkings (line 59.0 vs 59.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp toward over vs draftkings (line 59.0 vs 59.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: outdoor total without weather feed
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 59.5 (-105/-115) · sharp consensus 59.0 across 3 books · market mean 59.0 (model off by 10.8) · blend 35/65 model/market

UL Monroe @ Mississippi State · CFB · week 1

UL Monroe +28.5 (spread) -112 (1.89)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +16.9 vs market mean +28.0 (11.1 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+12.8%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+11.8%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.59.6%
MarketThe market's fair probability for this side — the price with the vig removed.50.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).27
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+1.16pp
Projected UL Monroe 18 — Mississippi State 35 · margin +16.9 · total 52.4
Why this pick · matchup breakdown

Net edge +25.3 (Weak O vs strong D)

  • Mississippi State pass offense (48th pct) vs UL Monroe defense (34th pct): edge +14
  • Mississippi State rush offense (40th pct) vs UL Monroe defense (41st pct): edge -2
  • UL Monroe pass offense (13th pct) vs Mississippi State defense (43rd pct): edge -30
  • UL Monroe rush offense (31st pct) vs Mississippi State defense (42nd pct): edge -11
  • Mississippi State OL (-10.0 pts): run-block proxy 3.73 ypc
  • UL Monroe OL (-2.6 pts): run-block proxy 4.56 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -28.5 (-108/-112) · sharp consensus -28.5 across 3 books · market mean 28.0 (model off by 11.1) · blend 35/65 model/market

Coastal Carolina @ West Virginia · CFB · week 1

Coastal Carolina +21.0 (spread) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +8.9 vs market mean +21.1 (12.2 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+12.7%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+11.7%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.58.1%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).36
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.23pp
Projected Coastal Carolina 20 — West Virginia 29 · margin +8.9 · total 48.1
Why this pick · matchup breakdown

Net edge +49.4 (Weak O vs strong D)

  • West Virginia pass offense (20th pct) vs Coastal Carolina defense (35th pct): edge -15
  • West Virginia rush offense (26th pct) vs Coastal Carolina defense (14th pct): edge +13
  • Coastal Carolina pass offense (5th pct) vs West Virginia defense (52nd pct): edge -47
  • Coastal Carolina rush offense (23rd pct) vs West Virginia defense (74th pct): edge -51
  • West Virginia OL (-10.0 pts): run-block proxy 3.69 ypc
  • Coastal Carolina OL (-9.6 pts): run-block proxy 4.31 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -21.0 (-110/-110) · sharp consensus -21.0 across 3 books · market mean 21.1 (model off by 12.2) · blend 35/65 model/market

Boston College @ Cincinnati · CFB · week 1

Over 50.5 (total) -105 (1.95)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: divergence guard: model +59.2 vs market mean +50.0 (9.2 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+12.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+11.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.57.4%
MarketThe market's fair probability for this side — the price with the vig removed.48.8%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).48
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.16pp
Projected Boston College 24 — Cincinnati 35 · margin +11.7 · total 59.3
Why this pick · matchup breakdown

Net edge +63.7 (Dual edge) — Cincinnati holds both the pass and rush matchup, the highest-conviction spot.

  • Cincinnati pass offense (83rd pct) vs Boston College defense (37th pct): edge +46
  • Cincinnati rush offense (87th pct) vs Boston College defense (35th pct): edge +51
  • Boston College pass offense (48th pct) vs Cincinnati defense (37th pct): edge +11
  • Boston College rush offense (12th pct) vs Cincinnati defense (40th pct): edge -29
  • Cincinnati OL (+10.0 pts): run-block proxy 5.75 ypc
  • Boston College OL (-10.0 pts): run-block proxy 3.32 ypc
  • 94F, wind 10mph at kickoff
  • Line movement: sharp consensus 1.2pp toward over vs draftkings (line 50.0 vs 50.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp toward over vs draftkings (line 50.0 vs 50.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 50.5 (-105/-115) · sharp consensus 50.0 across 3 books · market mean 50.0 (model off by 9.3) · blend 35/65 model/market

Ohio @ Nebraska · CFB · week 1

Over 46.5 (total) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Strong O vs weak DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +55.7 vs market mean +47.0 (8.7 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+11.8%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+10.8%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.58.6%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).25
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-0.46pp
Projected Ohio 25 — Nebraska 31 · margin +5.9 · total 55.7
Why this pick · matchup breakdown

Net edge -19.6 (Strong O vs weak D)

  • Nebraska pass offense (55th pct) vs Ohio defense (42nd pct): edge +13
  • Nebraska rush offense (50th pct) vs Ohio defense (47th pct): edge +3
  • Ohio pass offense (46th pct) vs Nebraska defense (51st pct): edge -6
  • Ohio rush offense (80th pct) vs Nebraska defense (25th pct): edge +54
  • Nebraska OL (-9.4 pts): run-block proxy 4.32 ypc
  • Ohio OL (+10.0 pts): run-block proxy 5.37 ypc
  • 92F, wind 4mph at kickoff
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 46.5 (-110/-110) · sharp consensus 47.0 across 3 books · market mean 47.0 (model off by 8.7) · blend 35/65 model/market

Clemson @ LSU · CFB · week 1

Clemson +10.0 (spread) -112 (1.89)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model -1.4 vs market mean +9.5 (11.0 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+11.5%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+10.5%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.57.9%
MarketThe market's fair probability for this side — the price with the vig removed.50.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).25
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+1.16pp
Projected Clemson 24 — LSU 22 · margin -1.4 · total 45.6
Why this pick · matchup breakdown

Net edge -9.9 (Weak O vs strong D)

  • LSU pass offense (50th pct) vs Clemson defense (66th pct): edge -16
  • LSU rush offense (28th pct) vs Clemson defense (78th pct): edge -50
  • Clemson pass offense (60th pct) vs LSU defense (62nd pct): edge -2
  • Clemson rush offense (38th pct) vs LSU defense (83rd pct): edge -45
  • LSU OL (-10.0 pts): run-block proxy 3.66 ypc
  • Clemson OL (-10.0 pts): run-block proxy 3.95 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -10.0 (-108/-112) · sharp consensus -10.0 across 3 books · market mean 9.5 (model off by 11.0) · blend 35/65 model/market

UCLA @ California · CFB · week 1

California +2.5 (spread) -108 (1.93)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Strong O vs weak DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +7.5 vs market mean -2.9 (10.4 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+11.5%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+10.5%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.57.9%
MarketThe market's fair probability for this side — the price with the vig removed.49.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).24
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-3.48pp
Projected UCLA 22 — California 29 · margin +7.5 · total 51.5
Why this pick · matchup breakdown

Net edge +6.8 (Strong O vs weak D)

  • California pass offense (48th pct) vs UCLA defense (32nd pct): edge +16
  • California rush offense (10th pct) vs UCLA defense (23rd pct): edge -13
  • UCLA pass offense (27th pct) vs California defense (53rd pct): edge -26
  • UCLA rush offense (56th pct) vs California defense (26th pct): edge +29
  • California OL (-10.0 pts): run-block proxy 2.79 ypc
  • UCLA OL (-5.2 pts): run-block proxy 4.47 ypc
  • Line movement: sharp consensus 1.9pp against home vs draftkings (line 2.0 vs 2.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.9pp against home vs draftkings (line 2.0 vs 2.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 2.5 (-108/-112) · sharp consensus 2.0 across 3 books · market mean -2.9 (model off by 10.4) · blend 35/65 model/market

Western Kentucky @ Nevada · CFB · week 1

Under 51.5 (total) -102 (1.98)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +43.6 vs market mean +51.4 (7.8 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+11.4%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+10.4%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.56.2%
MarketThe market's fair probability for this side — the price with the vig removed.48.1%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).29
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-2.32pp
Projected Western Kentucky 26 — Nevada 19 · margin -7.3 · total 43.6 · weather x0.969
Why this pick · matchup breakdown

Net edge -31.3 (Weak O vs strong D)

  • Nevada pass offense (7th pct) vs Western Kentucky defense (51st pct): edge -44
  • Nevada rush offense (35th pct) vs Western Kentucky defense (46th pct): edge -11
  • Western Kentucky pass offense (43rd pct) vs Nevada defense (49th pct): edge -6
  • Western Kentucky rush offense (26th pct) vs Nevada defense (16th pct): edge +10
  • Nevada OL (-4.6 pts): run-block proxy 4.49 ypc
  • Western Kentucky OL (-10.0 pts): run-block proxy 4.24 ypc
  • 64F, wind 17mph at kickoff -> total x0.969
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 51.5 (-118/-102) · sharp consensus 51.0 across 3 books · market mean 51.4 (model off by 7.8) · blend 35/65 model/market

Tulane @ Duke · CFB · week 1

Over 51.5 (total) -102 (1.98)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Strong O vs weak DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +59.3 vs market mean +50.8 (8.5 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+11.3%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+10.3%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.56.2%
MarketThe market's fair probability for this side — the price with the vig removed.48.1%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).19
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.63pp
Projected Tulane 29 — Duke 30 · margin +1.1 · total 59.3
Why this pick · matchup breakdown

Net edge +2.5 (Strong O vs weak D)

  • Duke pass offense (71st pct) vs Tulane defense (50th pct): edge +21
  • Duke rush offense (39th pct) vs Tulane defense (48th pct): edge -8
  • Tulane pass offense (54th pct) vs Duke defense (34th pct): edge +20
  • Tulane rush offense (45th pct) vs Duke defense (57th pct): edge -12
  • Duke OL (-10.0 pts): run-block proxy 4.26 ypc
  • Tulane OL (-3.4 pts): run-block proxy 4.53 ypc
  • 98F, wind 6mph at kickoff
  • Line movement: sharp consensus 1.4pp toward over vs draftkings (line 51.0 vs 51.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.4pp toward over vs draftkings (line 51.0 vs 51.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 51.5 (-102/-118) · sharp consensus 51.0 across 3 books · market mean 50.8 (model off by 8.5) · blend 35/65 model/market

Baylor @ Auburn · CFB · week 1

Under 59.5 (total) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +50.7 vs market mean +58.8 (8.1 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+11.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+10.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.58.2%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).34
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-0.70pp
Projected Baylor 20 — Auburn 31 · margin +10.4 · total 50.7
Why this pick · matchup breakdown

Net edge +30.3 (Weak O vs strong D)

  • Auburn pass offense (40th pct) vs Baylor defense (49th pct): edge -9
  • Auburn rush offense (79th pct) vs Baylor defense (42nd pct): edge +37
  • Baylor pass offense (62nd pct) vs Auburn defense (55th pct): edge +7
  • Baylor rush offense (44th pct) vs Auburn defense (93rd pct): edge -49
  • Auburn OL (+1.0 pts): run-block proxy 4.69 ypc
  • Baylor OL (-10.0 pts): run-block proxy 4.15 ypc
  • indoor (dome/closed roof)
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 59.5 (-110/-110) · sharp consensus 59.0 across 3 books · market mean 58.8 (model off by 8.1) · blend 35/65 model/market

Ball State @ Ohio State · CFB · week 1

Under 56.5 (total) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: divergence guard: model +47.8 vs market mean +56.0 (8.2 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+10.9%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+10.0%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.58.1%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).54
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.16pp
Projected Ball State 3 — Ohio State 45 · margin +41.2 · total 47.8
Why this pick · matchup breakdown

Net edge +161.3 (Dual edge) — Ohio State holds both the pass and rush matchup, the highest-conviction spot.

  • Ohio State pass offense (94th pct) vs Ball State defense (24th pct): edge +71
  • Ohio State rush offense (68th pct) vs Ball State defense (24th pct): edge +43
  • Ball State pass offense (0th pct) vs Ohio State defense (77th pct): edge -77
  • Ball State rush offense (6th pct) vs Ohio State defense (97th pct): edge -91
  • Ohio State OL (-1.4 pts): run-block proxy 4.60 ypc
  • Ball State OL (-10.0 pts): run-block proxy 3.54 ypc
  • 81F, wind 3mph at kickoff
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 56.5 (-110/-110) · sharp consensus 56.0 across 3 books · market mean 56.0 (model off by 8.2) · blend 35/65 model/market

Florida Atlantic @ Florida · CFB · week 1

Florida Atlantic +26.5 (spread) -112 (1.89)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +16.4 vs market mean +26.5 (10.1 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+10.8%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+9.8%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.58.6%
MarketThe market's fair probability for this side — the price with the vig removed.50.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).35
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.16pp
Projected Florida Atlantic 19 — Florida 35 · margin +16.4 · total 54.0
Why this pick · matchup breakdown

Net edge +47.6 (Weak O vs strong D) Turnover pairing: Florida's high-takeaway defense meets a turnover-prone offense.

  • Florida pass offense (39th pct) vs Florida Atlantic defense (52nd pct): edge -13
  • Florida rush offense (48th pct) vs Florida Atlantic defense (14th pct): edge +34
  • Florida Atlantic pass offense (53rd pct) vs Florida defense (69th pct): edge -16
  • Florida Atlantic rush offense (12th pct) vs Florida defense (62nd pct): edge -50
  • Florida OL (-10.0 pts): run-block proxy 4.27 ypc
  • Florida Atlantic OL (-10.0 pts): run-block proxy 3.56 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -26.5 (-108/-112) · sharp consensus -26.5 across 3 books · market mean 26.5 (model off by 10.1) · blend 35/65 model/market

Sam Houston @ Troy · CFB · week 1

Under 53.5 (total) -105 (1.95)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model +46.0 vs market mean +54.0 (8.0 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+10.0%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+9.0%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.56.4%
MarketThe market's fair probability for this side — the price with the vig removed.48.8%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).29
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.16pp
Projected Sam Houston 13 — Troy 33 · margin +19.8 · total 46.0
Why this pick · matchup breakdown

Net edge +37.0 (Weak O vs strong D)

  • Troy pass offense (22nd pct) vs Sam Houston defense (25th pct): edge -3
  • Troy rush offense (3rd pct) vs Sam Houston defense (13th pct): edge -11
  • Sam Houston pass offense (0th pct) vs Troy defense (61st pct): edge -61
  • Sam Houston rush offense (17th pct) vs Troy defense (46th pct): edge -28
  • Troy OL (-10.0 pts): run-block proxy 2.79 ypc
  • Sam Houston OL (-9.8 pts): run-block proxy 4.31 ypc
  • weather feed unavailable
  • Line movement: sharp consensus 1.2pp against over vs draftkings (line 54.0 vs 53.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp against over vs draftkings (line 54.0 vs 53.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: outdoor total without weather feed
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 53.5 (-115/-105) · sharp consensus 54.0 across 3 books · market mean 54.0 (model off by 8.0) · blend 35/65 model/market

Ball State @ Ohio State · CFB · week 1

Ball State +49.5 (spread) -108 (1.93)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: divergence guard: model +41.2 vs market mean +50.0 (8.8 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+9.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+8.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.56.6%
MarketThe market's fair probability for this side — the price with the vig removed.49.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).54
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.72pp
Projected Ball State 3 — Ohio State 45 · margin +41.2 · total 47.8
Why this pick · matchup breakdown

Net edge +161.3 (Dual edge) — Ohio State holds both the pass and rush matchup, the highest-conviction spot.

  • Ohio State pass offense (94th pct) vs Ball State defense (24th pct): edge +71
  • Ohio State rush offense (68th pct) vs Ball State defense (24th pct): edge +43
  • Ball State pass offense (0th pct) vs Ohio State defense (77th pct): edge -77
  • Ball State rush offense (6th pct) vs Ohio State defense (97th pct): edge -91
  • Ohio State OL (-1.4 pts): run-block proxy 4.60 ypc
  • Ball State OL (-10.0 pts): run-block proxy 3.54 ypc
  • Line movement: sharp consensus 1.6pp against home vs draftkings (line -50.5 vs -49.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.6pp against home vs draftkings (line -50.5 vs -49.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -49.5 (-112/-108) · sharp consensus -50.5 across 3 books · market mean 50.0 (model off by 8.8) · blend 35/65 model/market

Kent State @ South Carolina · CFB · week 1

Kent State +35.5 (spread) -108 (1.93)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: divergence guard: model +28.2 vs market mean +35.7 (7.5 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+7.6%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+6.6%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.55.9%
MarketThe market's fair probability for this side — the price with the vig removed.49.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).50
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.26pp
Projected Kent State 10 — South Carolina 38 · margin +28.2 · total 47.7
Why this pick · matchup breakdown

Net edge +89.6 (Dual edge) — South Carolina holds both the pass and rush matchup, the highest-conviction spot.

  • South Carolina pass offense (45th pct) vs Kent State defense (25th pct): edge +20
  • South Carolina rush offense (31st pct) vs Kent State defense (20th pct): edge +10
  • Kent State pass offense (31st pct) vs South Carolina defense (60th pct): edge -29
  • Kent State rush offense (0th pct) vs South Carolina defense (91st pct): edge -91
  • South Carolina OL (-10.0 pts): run-block proxy 3.28 ypc
  • Kent State OL (-10.0 pts): run-block proxy 2.99 ypc
  • Line movement: sharp consensus 1.2pp against home vs draftkings (line -36.0 vs -35.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp against home vs draftkings (line -36.0 vs -35.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -35.5 (-112/-108) · sharp consensus -36.0 across 3 books · market mean 35.7 (model off by 7.5) · blend 35/65 model/market

Wyoming @ Colorado State · CFB · week 1

Wyoming +3.0 (spread) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: divergence guard: model -5.1 vs market mean +2.5 (7.6 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+7.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+6.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.54.9%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).24
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+2.32pp
Projected Wyoming 27 — Colorado State 22 · margin -5.1 · total 48.3
Why this pick · matchup breakdown

Net edge -15.7 (Weak O vs strong D)

  • Colorado State pass offense (20th pct) vs Wyoming defense (48th pct): edge -27
  • Colorado State rush offense (13th pct) vs Wyoming defense (52nd pct): edge -39
  • Wyoming pass offense (8th pct) vs Colorado State defense (31st pct): edge -23
  • Wyoming rush offense (15th pct) vs Colorado State defense (22nd pct): edge -7
  • Colorado State OL (-10.0 pts): run-block proxy 3.58 ypc
  • Wyoming OL (-10.0 pts): run-block proxy 4.06 ypc
  • Line movement: sharp consensus 1.2pp against home vs draftkings (line -3.0 vs -3.0)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp against home vs draftkings (line -3.0 vs -3.0)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -3.0 (-110/-110) · sharp consensus -3.0 across 3 books · market mean 2.5 (model off by 7.6) · blend 35/65 model/market

Florida International @ South Florida · CFB · week 1

South Florida -14.5 (spread) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: divergence guard: model +21.5 vs market mean +14.5 (7.0 > 6.0)

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+6.8%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+5.8%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.56.0%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).44
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.00pp
Projected Florida International 17 — South Florida 38 · margin +21.5 · total 55.0
Why this pick · matchup breakdown

Net edge +72.6 (Dual edge) — South Florida holds both the pass and rush matchup, the highest-conviction spot.

  • South Florida pass offense (89th pct) vs Florida International defense (53rd pct): edge +36
  • South Florida rush offense (86th pct) vs Florida International defense (28th pct): edge +58
  • Florida International pass offense (37th pct) vs South Florida defense (64th pct): edge -26
  • Florida International rush offense (63rd pct) vs South Florida defense (69th pct): edge -6
  • South Florida OL (+10.0 pts): run-block proxy 5.47 ypc
  • Florida International OL (+6.2 pts): run-block proxy 4.87 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -14.5 (-110/-110) · sharp consensus -14.5 across 3 books · market mean 14.5 (model off by 7.0) · blend 35/65 model/market

Arkansas State @ Memphis · CFB · week 1

Memphis -11.5 (spread) -105 (1.95)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: adjusted EV +2.0% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+3.0%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+2.0%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.52.8%
MarketThe market's fair probability for this side — the price with the vig removed.48.8%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).41
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-0.46pp
Projected Arkansas State 19 — Memphis 35 · margin +15.6 · total 54.1
Why this pick · matchup breakdown

Net edge +63.4 (Weak O vs strong D)

  • Memphis pass offense (62nd pct) vs Arkansas State defense (53rd pct): edge +10
  • Memphis rush offense (86th pct) vs Arkansas State defense (36th pct): edge +50
  • Arkansas State pass offense (33rd pct) vs Memphis defense (47th pct): edge -14
  • Arkansas State rush offense (7th pct) vs Memphis defense (63rd pct): edge -57
  • Memphis OL (+10.0 pts): run-block proxy 5.19 ypc
  • Arkansas State OL (-10.0 pts): run-block proxy 3.58 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -11.5 (-105/-115) · sharp consensus -11.0 across 3 books · market mean 10.8 (model off by 4.8) · blend 35/65 model/market

Boston College @ Cincinnati · CFB · week 1

Cincinnati -7.5 (spread) -108 (1.93)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: adjusted EV +1.7% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+2.7%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+1.7%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.53.3%
MarketThe market's fair probability for this side — the price with the vig removed.49.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).48
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+1.16pp
Projected Boston College 24 — Cincinnati 35 · margin +11.7 · total 59.3
Why this pick · matchup breakdown

Net edge +63.7 (Dual edge) — Cincinnati holds both the pass and rush matchup, the highest-conviction spot.

  • Cincinnati pass offense (83rd pct) vs Boston College defense (37th pct): edge +46
  • Cincinnati rush offense (87th pct) vs Boston College defense (35th pct): edge +51
  • Boston College pass offense (48th pct) vs Cincinnati defense (37th pct): edge +11
  • Boston College rush offense (12th pct) vs Cincinnati defense (40th pct): edge -29
  • Cincinnati OL (+10.0 pts): run-block proxy 5.75 ypc
  • Boston College OL (-10.0 pts): run-block proxy 3.32 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -7.5 (-108/-112) · sharp consensus -7.5 across 3 books · market mean 7.5 (model off by 4.2) · blend 35/65 model/market

Central Michigan @ New Mexico · CFB · week 1

Over 46.5 (total) -118 (1.85)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: adjusted EV +0.7% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+0.7%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+0.7%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.54.5%
MarketThe market's fair probability for this side — the price with the vig removed.51.9%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).30
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+4.11pp
Projected Central Michigan 21 — New Mexico 29 · margin +7.8 · total 49.9
Why this pick · matchup breakdown

Net edge +32.5 (Weak O vs strong D)

  • New Mexico pass offense (36th pct) vs Central Michigan defense (44th pct): edge -8
  • New Mexico rush offense (41st pct) vs Central Michigan defense (48th pct): edge -7
  • Central Michigan pass offense (32nd pct) vs New Mexico defense (61st pct): edge -29
  • Central Michigan rush offense (18th pct) vs New Mexico defense (71st pct): edge -53
  • New Mexico OL (-10.0 pts): run-block proxy 4.27 ypc
  • Central Michigan OL (-10.0 pts): run-block proxy 3.99 ypc
  • 84F, wind 4mph at kickoff
  • Line movement: sharp consensus 2.3pp toward over vs draftkings (line 46.0 vs 46.5)
  • Adjustment: sharp support +2.3pp: +1.0%
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 2.3pp toward over vs draftkings (line 46.0 vs 46.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 46.5 (-118/-102) · sharp consensus 46.0 across 3 books · market mean 47.5 (model off by 2.4) · blend 35/65 model/market

Baylor @ Auburn · CFB · week 1

Auburn -7.5 (spread) -102 (1.98)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: adjusted EV +0.6% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+1.6%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+0.6%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.51.3%
MarketThe market's fair probability for this side — the price with the vig removed.48.1%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).34
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.88pp
Projected Baylor 20 — Auburn 31 · margin +10.4 · total 50.7
Why this pick · matchup breakdown

Net edge +30.3 (Weak O vs strong D)

  • Auburn pass offense (40th pct) vs Baylor defense (49th pct): edge -9
  • Auburn rush offense (79th pct) vs Baylor defense (42nd pct): edge +37
  • Baylor pass offense (62nd pct) vs Auburn defense (55th pct): edge +7
  • Baylor rush offense (44th pct) vs Auburn defense (93rd pct): edge -49
  • Auburn OL (+1.0 pts): run-block proxy 4.69 ypc
  • Baylor OL (-10.0 pts): run-block proxy 4.15 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -7.5 (-102/-118) · sharp consensus -7.5 across 3 books · market mean 6.9 (model off by 3.4) · blend 35/65 model/market

Florida International @ South Florida · CFB · week 1

Over 51.5 (total) -115 (1.87)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: outdoor total without weather -> analysis-only

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+1.4%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+0.4%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.54.2%
MarketThe market's fair probability for this side — the price with the vig removed.51.2%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).41
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+1.52pp
Projected Florida International 17 — South Florida 38 · margin +21.5 · total 55.0
Why this pick · matchup breakdown

Net edge +72.6 (Dual edge) — South Florida holds both the pass and rush matchup, the highest-conviction spot.

  • South Florida pass offense (89th pct) vs Florida International defense (53rd pct): edge +36
  • South Florida rush offense (86th pct) vs Florida International defense (28th pct): edge +58
  • Florida International pass offense (37th pct) vs South Florida defense (64th pct): edge -26
  • Florida International rush offense (63rd pct) vs South Florida defense (69th pct): edge -6
  • South Florida OL (+10.0 pts): run-block proxy 5.47 ypc
  • Florida International OL (+6.2 pts): run-block proxy 4.87 ypc
  • weather feed unavailable
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: outdoor total without weather feed
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 51.5 (-115/-105) · sharp consensus 51.5 across 3 books · market mean 51.9 (model off by 3.1) · blend 35/65 model/market

Northern Illinois @ Iowa · CFB · week 1

Over 44.5 (total) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: outdoor total without weather -> analysis-only

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+1.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.+0.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.53.0%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).48
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.62pp
Projected Northern Illinois 9 — Iowa 38 · margin +28.9 · total 47.5
Why this pick · matchup breakdown

Net edge +93.7 (Dual edge) — Iowa holds both the pass and rush matchup, the highest-conviction spot.

  • Iowa pass offense (39th pct) vs Northern Illinois defense (23rd pct): edge +16
  • Iowa rush offense (70th pct) vs Northern Illinois defense (21st pct): edge +48
  • Northern Illinois pass offense (1st pct) vs Iowa defense (63rd pct): edge -62
  • Northern Illinois rush offense (42nd pct) vs Iowa defense (78th pct): edge -36
  • Iowa OL (-2.5 pts): run-block proxy 4.56 ypc
  • Northern Illinois OL (-4.8 pts): run-block proxy 4.48 ypc
  • weather feed unavailable
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: outdoor total without weather feed
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 44.5 (-110/-110) · sharp consensus 44.5 across 3 books · market mean 44.3 (model off by 3.2) · blend 35/65 model/market

Missouri State @ Texas A&M · CFB · week 1

Over 53.5 (total) -112 (1.89)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: outdoor total without weather -> analysis-only

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+0.6%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-0.4%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.53.1%
MarketThe market's fair probability for this side — the price with the vig removed.50.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).50
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+1.16pp
Projected Missouri State 16 — Texas A&M 40 · margin +24.3 · total 56.4
Why this pick · matchup breakdown

Net edge +130.1 (Dual edge) — Texas A&M holds both the pass and rush matchup, the highest-conviction spot.

  • Texas A&M pass offense (90th pct) vs Missouri State defense (32nd pct): edge +59
  • Texas A&M rush offense (87th pct) vs Missouri State defense (12th pct): edge +75
  • Missouri State pass offense (52nd pct) vs Texas A&M defense (92nd pct): edge -40
  • Missouri State rush offense (5th pct) vs Texas A&M defense (54th pct): edge -50
  • Texas A&M OL (+6.0 pts): run-block proxy 4.86 ypc
  • Missouri State OL (-10.0 pts): run-block proxy 3.60 ypc
  • weather feed unavailable
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: outdoor total without weather feed
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 53.5 (-112/-108) · sharp consensus 53.5 across 3 books · market mean 53.5 (model off by 2.9) · blend 35/65 model/market

Sam Houston @ Troy · CFB · week 1

Troy -16.5 (spread) -115 (1.87)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: adjusted EV -0.8% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+0.2%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-0.8%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.53.6%
MarketThe market's fair probability for this side — the price with the vig removed.51.2%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).31
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+1.86pp
Projected Sam Houston 13 — Troy 33 · margin +19.8 · total 46.0
Why this pick · matchup breakdown

Net edge +37.0 (Weak O vs strong D)

  • Troy pass offense (22nd pct) vs Sam Houston defense (25th pct): edge -3
  • Troy rush offense (3rd pct) vs Sam Houston defense (13th pct): edge -11
  • Sam Houston pass offense (0th pct) vs Troy defense (61st pct): edge -61
  • Sam Houston rush offense (17th pct) vs Troy defense (46th pct): edge -28
  • Troy OL (-10.0 pts): run-block proxy 2.79 ypc
  • Sam Houston OL (-9.8 pts): run-block proxy 4.31 ypc
  • Line movement: sharp consensus 1.2pp against home vs draftkings (line -16.5 vs -16.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp against home vs draftkings (line -16.5 vs -16.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -16.5 (-115/-105) · sharp consensus -16.5 across 3 books · market mean 16.5 (model off by 3.3) · blend 35/65 model/market

UL Monroe @ Mississippi State · CFB · week 1

Under 54.5 (total) -105 (1.95)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: outdoor total without weather -> analysis-only

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+0.2%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-0.8%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.51.3%
MarketThe market's fair probability for this side — the price with the vig removed.48.8%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).25
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+2.32pp
Projected UL Monroe 18 — Mississippi State 35 · margin +16.9 · total 52.4
Why this pick · matchup breakdown

Net edge +25.3 (Weak O vs strong D)

  • Mississippi State pass offense (48th pct) vs UL Monroe defense (34th pct): edge +14
  • Mississippi State rush offense (40th pct) vs UL Monroe defense (41st pct): edge -2
  • UL Monroe pass offense (13th pct) vs Mississippi State defense (43rd pct): edge -30
  • UL Monroe rush offense (31st pct) vs Mississippi State defense (42nd pct): edge -11
  • Mississippi State OL (-10.0 pts): run-block proxy 3.73 ypc
  • UL Monroe OL (-2.6 pts): run-block proxy 4.56 ypc
  • weather feed unavailable
  • Line movement: sharp consensus 1.2pp against over vs draftkings (line 54.0 vs 54.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp against over vs draftkings (line 54.0 vs 54.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: outdoor total without weather feed
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 54.5 (-115/-105) · sharp consensus 54.0 across 3 books · market mean 54.0 (model off by 1.6) · blend 35/65 model/market

UNLV @ Hawai'i · CFB · week 1

Over 55.5 (total) -108 (1.93)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: adjusted EV -0.9% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).+0.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-0.9%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.52.0%
MarketThe market's fair probability for this side — the price with the vig removed.49.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).37
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-0.46pp
Projected UNLV 32 — Hawai'i 26 · margin -5.7 · total 57.8 · weather x0.993
Why this pick · matchup breakdown

Net edge -52.8 (Dual edge) — UNLV holds both the pass and rush matchup, the highest-conviction spot.

  • Hawai'i pass offense (50th pct) vs UNLV defense (41st pct): edge +9
  • Hawai'i rush offense (8th pct) vs UNLV defense (21st pct): edge -14
  • UNLV pass offense (80th pct) vs Hawai'i defense (39th pct): edge +40
  • UNLV rush offense (89th pct) vs Hawai'i defense (37th pct): edge +52
  • Hawai'i OL (-10.0 pts): run-block proxy 3.40 ypc
  • UNLV OL (+10.0 pts): run-block proxy 5.27 ypc
  • 82F, wind 13mph at kickoff -> total x0.993
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (1 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 55.5 (-108/-112) · sharp consensus 55.5 across 3 books · market mean 55.5 (model off by 2.3) · blend 35/65 model/market

Marshall @ Penn State · CFB · week 1

Marshall +23.5 (spread) -115 (1.87)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: adjusted EV -1.1% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-0.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-1.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.53.4%
MarketThe market's fair probability for this side — the price with the vig removed.51.2%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).32
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.00pp
Projected Marshall 17 — Penn State 37 · margin +20.4 · total 53.9
Why this pick · matchup breakdown

Net edge +38.2 (Dual edge) — Penn State holds both the pass and rush matchup, the highest-conviction spot.

  • Penn State pass offense (59th pct) vs Marshall defense (33rd pct): edge +26
  • Penn State rush offense (82nd pct) vs Marshall defense (61st pct): edge +21
  • Marshall pass offense (51st pct) vs Penn State defense (66th pct): edge -15
  • Marshall rush offense (53rd pct) vs Penn State defense (60th pct): edge -6
  • Penn State OL (+3.5 pts): run-block proxy 4.77 ypc
  • Marshall OL (+3.9 pts): run-block proxy 4.78 ypc
  • Line movement: sharp consensus 1.2pp toward home vs draftkings (line -23.5 vs -23.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp toward home vs draftkings (line -23.5 vs -23.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -23.5 (-105/-115) · sharp consensus -23.5 across 3 books · market mean 23.5 (model off by 3.1) · blend 35/65 model/market

UCLA @ California · CFB · week 1

Under 53.5 (total) -108 (1.93)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Strong O vs weak DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: adjusted EV -1.4% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-0.4%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-1.4%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.51.7%
MarketThe market's fair probability for this side — the price with the vig removed.49.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).24
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-0.70pp
Projected UCLA 22 — California 29 · margin +7.5 · total 51.5
Why this pick · matchup breakdown

Net edge +6.8 (Strong O vs weak D)

  • California pass offense (48th pct) vs UCLA defense (32nd pct): edge +16
  • California rush offense (10th pct) vs UCLA defense (23rd pct): edge -13
  • UCLA pass offense (27th pct) vs California defense (53rd pct): edge -26
  • UCLA rush offense (56th pct) vs California defense (26th pct): edge +29
  • California OL (-10.0 pts): run-block proxy 2.79 ypc
  • UCLA OL (-5.2 pts): run-block proxy 4.47 ypc
  • 64F, wind 4mph at kickoff
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 53.5 (-112/-108) · sharp consensus 53.0 across 3 books · market mean 53.4 (model off by 1.9) · blend 35/65 model/market

Boise State @ Oregon · CFB · week 1

Boise State +24.5 (spread) -108 (1.93)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: adjusted EV -1.6% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-0.7%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-1.7%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.51.6%
MarketThe market's fair probability for this side — the price with the vig removed.49.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).46
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.00pp
Projected Boise State 15 — Oregon 38 · margin +22.3 · total 52.9
Why this pick · matchup breakdown

Net edge +77.4 (Dual edge) — Oregon holds both the pass and rush matchup, the highest-conviction spot.

  • Oregon pass offense (98th pct) vs Boise State defense (72nd pct): edge +26
  • Oregon rush offense (100th pct) vs Boise State defense (25th pct): edge +75
  • Boise State pass offense (57th pct) vs Oregon defense (71st pct): edge -14
  • Boise State rush offense (61st pct) vs Oregon defense (76th pct): edge -15
  • Oregon OL (+10.0 pts): run-block proxy 5.42 ypc
  • Boise State OL (-2.4 pts): run-block proxy 4.57 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -24.5 (-112/-108) · sharp consensus -24.5 across 3 books · market mean 24.5 (model off by 2.2) · blend 35/65 model/market

Central Michigan @ New Mexico · CFB · week 1

Central Michigan +10.5 (spread) -115 (1.87)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: adjusted EV -1.7% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-0.7%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-1.7%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.53.1%
MarketThe market's fair probability for this side — the price with the vig removed.51.2%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).30
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.00pp
Projected Central Michigan 21 — New Mexico 29 · margin +7.8 · total 49.9
Why this pick · matchup breakdown

Net edge +32.5 (Weak O vs strong D)

  • New Mexico pass offense (36th pct) vs Central Michigan defense (44th pct): edge -8
  • New Mexico rush offense (41st pct) vs Central Michigan defense (48th pct): edge -7
  • Central Michigan pass offense (32nd pct) vs New Mexico defense (61st pct): edge -29
  • Central Michigan rush offense (18th pct) vs New Mexico defense (71st pct): edge -53
  • New Mexico OL (-10.0 pts): run-block proxy 4.27 ypc
  • Central Michigan OL (-10.0 pts): run-block proxy 3.99 ypc
  • Line movement: sharp consensus 1.2pp toward home vs draftkings (line -10.5 vs -10.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp toward home vs draftkings (line -10.5 vs -10.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -10.5 (-105/-115) · sharp consensus -10.5 across 3 books · market mean 10.5 (model off by 2.7) · blend 35/65 model/market

Western Michigan @ Michigan · CFB · week 1

Under 49.5 (total) -115 (1.87)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: outdoor total without weather -> analysis-only

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-0.9%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-1.9%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.53.0%
MarketThe market's fair probability for this side — the price with the vig removed.51.2%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).37
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.24pp
Projected Western Michigan 18 — Michigan 29 · margin +11.7 · total 47.2
Why this pick · matchup breakdown

Net edge +59.9 (Weak O vs strong D)

  • Michigan pass offense (57th pct) vs Western Michigan defense (57th pct): edge +0
  • Michigan rush offense (100th pct) vs Western Michigan defense (54th pct): edge +46
  • Western Michigan pass offense (16th pct) vs Michigan defense (67th pct): edge -50
  • Western Michigan rush offense (58th pct) vs Michigan defense (79th pct): edge -22
  • Michigan OL (+10.0 pts): run-block proxy 5.52 ypc
  • Western Michigan OL (-0.7 pts): run-block proxy 4.63 ypc
  • weather feed unavailable
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: outdoor total without weather feed
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 49.5 (-105/-115) · sharp consensus 49.5 across 3 books · market mean 49.4 (model off by 2.2) · blend 35/65 model/market

Northern Illinois @ Iowa · CFB · week 1

Northern Illinois +31.5 (spread) -115 (1.87)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: cfb mismatch cap: |spread| 31.5 > 28.0

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-0.9%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-1.9%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.53.0%
MarketThe market's fair probability for this side — the price with the vig removed.51.2%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).52
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.00pp
Projected Northern Illinois 9 — Iowa 38 · margin +28.9 · total 47.5
Why this pick · matchup breakdown

Net edge +93.7 (Dual edge) — Iowa holds both the pass and rush matchup, the highest-conviction spot.

  • Iowa pass offense (39th pct) vs Northern Illinois defense (23rd pct): edge +16
  • Iowa rush offense (70th pct) vs Northern Illinois defense (21st pct): edge +48
  • Northern Illinois pass offense (1st pct) vs Iowa defense (63rd pct): edge -62
  • Northern Illinois rush offense (42nd pct) vs Iowa defense (78th pct): edge -36
  • Iowa OL (-2.5 pts): run-block proxy 4.56 ypc
  • Northern Illinois OL (-4.8 pts): run-block proxy 4.48 ypc
  • Line movement: sharp consensus 1.2pp toward home vs draftkings (line -31.5 vs -31.5)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.2pp toward home vs draftkings (line -31.5 vs -31.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -31.5 (-105/-115) · sharp consensus -31.5 across 3 books · market mean 31.5 (model off by 2.6) · blend 35/65 model/market

Miami (OH) @ Pittsburgh · CFB · week 1

Under 47.5 (total) -108 (1.93)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: adjusted EV -2.1% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-1.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-2.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.51.4%
MarketThe market's fair probability for this side — the price with the vig removed.49.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).32
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.46pp
Projected Miami (OH) 16 — Pittsburgh 30 · margin +14.8 · total 45.8
Why this pick · matchup breakdown

Net edge +38.1 (Weak O vs strong D)

  • Pittsburgh pass offense (66th pct) vs Miami (OH) defense (62nd pct): edge +5
  • Pittsburgh rush offense (25th pct) vs Miami (OH) defense (61st pct): edge -36
  • Miami (OH) pass offense (23rd pct) vs Pittsburgh defense (73rd pct): edge -50
  • Miami (OH) rush offense (30th pct) vs Pittsburgh defense (88th pct): edge -58
  • Pittsburgh OL (-10.0 pts): run-block proxy 3.56 ypc
  • Miami (OH) OL (-10.0 pts): run-block proxy 4.29 ypc
  • 72F, wind 3mph at kickoff
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 47.5 (-112/-108) · sharp consensus 47.5 across 3 books · market mean 47.5 (model off by 1.7) · blend 35/65 model/market

Oregon State @ Houston · CFB · week 1

Oregon State +21.0 (spread) -118 (1.85)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: adjusted EV -2.1% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-1.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-2.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.52.2%
MarketThe market's fair probability for this side — the price with the vig removed.51.9%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).37
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.92pp
Projected Oregon State 14 — Houston 32 · margin +17.5 · total 46.4
Why this pick · matchup breakdown

Net edge +52.9 (Weak O vs strong D)

  • Houston pass offense (56th pct) vs Oregon State defense (40th pct): edge +16
  • Houston rush offense (40th pct) vs Oregon State defense (47th pct): edge -7
  • Oregon State pass offense (19th pct) vs Houston defense (62nd pct): edge -42
  • Oregon State rush offense (15th pct) vs Houston defense (71st pct): edge -56
  • Houston OL (-10.0 pts): run-block proxy 4.26 ypc
  • Oregon State OL (-10.0 pts): run-block proxy 3.75 ypc
  • Line movement: sharp consensus 1.4pp toward home vs draftkings (line -21.0 vs -21.0)
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 1.4pp toward home vs draftkings (line -21.0 vs -21.0)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -21.0 (-102/-118) · sharp consensus -21.0 across 3 books · market mean 20.8 (model off by 3.3) · blend 35/65 model/market

Wyoming @ Colorado State · CFB · week 1

Over 46.5 (total) -110 (1.91)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: outdoor total without weather -> analysis-only

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-1.2%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-2.2%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.51.8%
MarketThe market's fair probability for this side — the price with the vig removed.50.0%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).22
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.16pp
Projected Wyoming 27 — Colorado State 22 · margin -5.1 · total 48.3
Why this pick · matchup breakdown

Net edge -15.7 (Weak O vs strong D)

  • Colorado State pass offense (20th pct) vs Wyoming defense (48th pct): edge -27
  • Colorado State rush offense (13th pct) vs Wyoming defense (52nd pct): edge -39
  • Wyoming pass offense (8th pct) vs Colorado State defense (31st pct): edge -23
  • Wyoming rush offense (15th pct) vs Colorado State defense (22nd pct): edge -7
  • Colorado State OL (-10.0 pts): run-block proxy 3.58 ypc
  • Wyoming OL (-10.0 pts): run-block proxy 4.06 ypc
  • weather feed unavailable
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: outdoor total without weather feed
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 46.5 (-110/-110) · sharp consensus 46.0 across 3 books · market mean 46.0 (model off by 2.3) · blend 35/65 model/market

Boise State @ Oregon · CFB · week 1

Over 51.5 (total) -108 (1.93)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: adjusted EV -2.6% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-1.6%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-2.6%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.51.1%
MarketThe market's fair probability for this side — the price with the vig removed.49.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).46
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.46pp
Projected Boise State 15 — Oregon 38 · margin +22.3 · total 52.9
Why this pick · matchup breakdown

Net edge +77.4 (Dual edge) — Oregon holds both the pass and rush matchup, the highest-conviction spot.

  • Oregon pass offense (98th pct) vs Boise State defense (72nd pct): edge +26
  • Oregon rush offense (100th pct) vs Boise State defense (25th pct): edge +75
  • Boise State pass offense (57th pct) vs Oregon defense (71st pct): edge -14
  • Boise State rush offense (61st pct) vs Oregon defense (76th pct): edge -15
  • Oregon OL (+10.0 pts): run-block proxy 5.42 ypc
  • Boise State OL (-2.4 pts): run-block proxy 4.57 ypc
  • 63F, wind 2mph at kickoff
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 51.5 (-108/-112) · sharp consensus 51.5 across 3 books · market mean 51.7 (model off by 1.2) · blend 35/65 model/market

UNLV @ Hawai'i · CFB · week 1

UNLV -3.0 (spread) -112 (1.89)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: adjusted EV -2.7% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-1.8%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-2.8%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.50.6%
MarketThe market's fair probability for this side — the price with the vig removed.50.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).37
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.00pp
Projected UNLV 32 — Hawai'i 26 · margin -5.7 · total 57.8 · weather x0.993
Why this pick · matchup breakdown

Net edge -52.8 (Dual edge) — UNLV holds both the pass and rush matchup, the highest-conviction spot.

  • Hawai'i pass offense (50th pct) vs UNLV defense (41st pct): edge +9
  • Hawai'i rush offense (8th pct) vs UNLV defense (21st pct): edge -14
  • UNLV pass offense (80th pct) vs Hawai'i defense (39th pct): edge +40
  • UNLV rush offense (89th pct) vs Hawai'i defense (37th pct): edge +52
  • Hawai'i OL (-10.0 pts): run-block proxy 3.40 ypc
  • UNLV OL (+10.0 pts): run-block proxy 5.27 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (1 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 3.0 (-108/-112) · sharp consensus 3.0 across 3 books · market mean -3.0 (model off by 2.7) · blend 35/65 model/market

Miami (OH) @ Pittsburgh · CFB · week 1

Miami (OH) +16.5 (spread) -112 (1.89)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: adjusted EV -3.0% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-2.0%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-3.0%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.51.8%
MarketThe market's fair probability for this side — the price with the vig removed.50.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).32
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+0.46pp
Projected Miami (OH) 16 — Pittsburgh 30 · margin +14.8 · total 45.8
Why this pick · matchup breakdown

Net edge +38.1 (Weak O vs strong D)

  • Pittsburgh pass offense (66th pct) vs Miami (OH) defense (62nd pct): edge +5
  • Pittsburgh rush offense (25th pct) vs Miami (OH) defense (61st pct): edge -36
  • Miami (OH) pass offense (23rd pct) vs Pittsburgh defense (73rd pct): edge -50
  • Miami (OH) rush offense (30th pct) vs Pittsburgh defense (88th pct): edge -58
  • Pittsburgh OL (-10.0 pts): run-block proxy 3.56 ypc
  • Miami (OH) OL (-10.0 pts): run-block proxy 4.29 ypc
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings -16.5 (-108/-112) · sharp consensus -16.5 across 3 books · market mean 16.5 (model off by 1.7) · blend 35/65 model/market

Marshall @ Penn State · CFB · week 1

Over 51.5 (total) -115 (1.87)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: outdoor total without weather -> analysis-only

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-0.6%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-3.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.53.2%
MarketThe market's fair probability for this side — the price with the vig removed.51.2%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).29
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.-1.62pp
Projected Marshall 17 — Penn State 37 · margin +20.4 · total 53.9
Why this pick · matchup breakdown

Net edge +38.2 (Dual edge) — Penn State holds both the pass and rush matchup, the highest-conviction spot.

  • Penn State pass offense (59th pct) vs Marshall defense (33rd pct): edge +26
  • Penn State rush offense (82nd pct) vs Marshall defense (61st pct): edge +21
  • Marshall pass offense (51st pct) vs Penn State defense (66th pct): edge -15
  • Marshall rush offense (53rd pct) vs Penn State defense (60th pct): edge -6
  • Penn State OL (+3.5 pts): run-block proxy 4.77 ypc
  • Marshall OL (+3.9 pts): run-block proxy 4.78 ypc
  • weather feed unavailable
  • Line movement: sharp consensus 2.3pp against over vs draftkings (line 52.0 vs 51.5)
  • Adjustment: sharp fade -2.3pp: -1.5%
  • Adjustment: fragile edge: -1.0%
  • Adjustment: noted: sharp consensus 2.3pp against over vs draftkings (line 52.0 vs 51.5)
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: outdoor total without weather feed
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 51.5 (-115/-105) · sharp consensus 52.0 across 3 books · market mean 51.6 (model off by 2.4) · blend 35/65 model/market

Liberty @ James Madison · CFB · week 1

Over 50.5 (total) -112 (1.89)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Dual edgeOne team holds BOTH the passing and rushing matchup edge — the model's highest-conviction archetype.fragile

Held to analysis-only: adjusted EV -3.6% below threshold

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-2.6%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-3.6%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.51.5%
MarketThe market's fair probability for this side — the price with the vig removed.50.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).46
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+2.32pp
Projected Liberty 20 — James Madison 32 · margin +11.4 · total 51.7
Why this pick · matchup breakdown

Net edge +77.8 (Dual edge) — James Madison holds both the pass and rush matchup, the highest-conviction spot.

  • James Madison pass offense (67th pct) vs Liberty defense (46th pct): edge +21
  • James Madison rush offense (91st pct) vs Liberty defense (20th pct): edge +71
  • Liberty pass offense (35th pct) vs James Madison defense (65th pct): edge -30
  • Liberty rush offense (62nd pct) vs James Madison defense (79th pct): edge -17
  • James Madison OL (+10.0 pts): run-block proxy 5.71 ypc
  • Liberty OL (-0.5 pts): run-block proxy 4.63 ypc
  • 83F, wind 6mph at kickoff
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 50.5 (-112/-108) · sharp consensus 50.0 across 3 books · market mean 50.4 (model off by 1.3) · blend 35/65 model/market

Arkansas State @ Memphis · CFB · week 1

Over 53.5 (total) -108 (1.93)

PaperSimulated pick. Football is paper-only until it shows positive CLV vs the sharp close over a meaningful sample — never sized with real money.Weak O vs strong DThe primary matchup archetype driving this game's edge (strong unit vs weak opposing unit).fragile

Held to analysis-only: outdoor total without weather -> analysis-only

EVRaw expected value from the blended cover/total probability vs the price. The threshold and tiers read ADJUSTED EV (sharp support/fade and fragile-edge haircuts applied once).-3.1%
Adj EVAdjusted EV: raw EV plus small signed context adjustments — sharp support/fade vs the sharp consensus and a fragile-edge haircut. Applied exactly once, here.-4.1%
ModelThe model's probability for THIS side, after blending 35/65 toward the de-vigged market — the same market-heavy anchoring philosophy as the MLB models.50.3%
MarketThe market's fair probability for this side — the price with the vig removed.49.5%
ConfidenceConfidence (0–100) from edge size and data completeness, with college-noise haircuts: CFB scores lower than NFL for the same edge, Group-of-Five games lower still, and explosive offenses cut totals confidence (variance, not projection).38
CLVClosing-line value vs the SHARP close, in probability points. + = the market moved toward our side after commit — the gate for football ever going live.+1.16pp
Projected Arkansas State 19 — Memphis 35 · margin +15.6 · total 54.1
Why this pick · matchup breakdown

Net edge +63.4 (Weak O vs strong D)

  • Memphis pass offense (62nd pct) vs Arkansas State defense (53rd pct): edge +10
  • Memphis rush offense (86th pct) vs Arkansas State defense (36th pct): edge +50
  • Arkansas State pass offense (33rd pct) vs Memphis defense (47th pct): edge -14
  • Arkansas State rush offense (7th pct) vs Memphis defense (63rd pct): edge -57
  • Memphis OL (+10.0 pts): run-block proxy 5.19 ypc
  • Arkansas State OL (-10.0 pts): run-block proxy 3.58 ypc
  • weather feed unavailable
  • Adjustment: fragile edge: -1.0%
  • Fragile signal: small sample (0 games) — early-season prior blend active
  • Fragile signal: outdoor total without weather feed
  • Fragile signal: OL layer on sack/YPC proxy only
Market: draftkings 53.5 (-108/-112) · sharp consensus 53.5 across 3 books · market mean 53.5 (model off by 0.6) · blend 35/65 model/market