ROBOT
POLL
WEEK 1 · 2026 SEASON · RENDERED FROM LOCKED MODEL
0–0
VS THE CLOSING LINE · —
BRIER .190 CALIBRATED ✓
Model-generated rankings, picks, and analysis. Every prediction includes its signal breakdown, is locked before kickoff, and is graded against results.
RANKINGS PICKS VS MARKET BETS SIGNALS SCOREBOARD MISSES HISTORY LAB ABOUT MY HISTORY
RANKINGS — TOP 25
COMPUTED FROM RESULTS · UPDATED WEEKLY
Least-squares ratings from game results: rating = points relative to the average FBS team on a neutral field. Margins are capped at 28 points. Strength of schedule is derived from the same solution.
1
Indiana 0–0 —
SOS #1
+22.3
2
Notre Dame 0–0 —
SOS #36
+19.5
3
Ohio State 0–0 —
SOS #103
+16.7
4
Oregon 0–0 —
SOS #102
+14.9
5
Miami 0–0 —
SOS #101
+13.4
6
Ole Miss 0–0 —
SOS #100
+13.3
7
Penn State 0–0 —
SOS #99
+11.8
8
Iowa 0–0 —
SOS #98
+11.4
9
Texas 0–0 —
SOS #97
+11.1
10
James Madison 0–0 —
SOS #96
+10.5
11
Texas A&M 0–0 —
SOS #95
+9.7
12
USC 0–0 —
SOS #94
+9.6
13
Washington 0–0 —
SOS #93
+9.5
14
Alabama 0–0 —
SOS #92
+9.3
15
SMU 0–0 —
SOS #91
+9.1
16
Oklahoma 0–0 —
SOS #90
+8.0
17
Liberty 0–0 —
SOS #89
+7.9
18
Michigan 0–0 —
SOS #88
+7.3
19
South Florida 0–0 —
SOS #87
+7.0
20
Clemson 0–0 —
SOS #73
+6.5
21
Louisville 0–0 —
SOS #74
+6.4
22
North Texas 0–0 —
SOS #75
+5.3
23
South Carolina 0–0 —
SOS #76
+5.0
24
Washington State 0–0 —
SOS #77
+4.5
25
LSU 0–0 —
SOS #78
+4.4
Validation: across 442 games between ranked teams since 2016, the higher-ranked team won 63.1%. For teams ranked within 5 spots of each other, the rate was 54.7%.
MODEL VS MARKET — 3 LARGEST WIN-PROBABILITY DIVERGENCES
Ranked by disagreement over who WINS the game (model win probability vs. the spread-implied probability). The Betting Analysis below ranks by a different question — edge against the posted number — so its order can differ.
TOLEDO +9.5
+34 PTS APART ON WIN %
MARKET
27%
MODEL
61%
Largest signal contribution toward this side: rating_gap at +6.2 points of lean. Model and market win probabilities differ by 34 percentage points.
TULSA +13.0+32 PTS APART
Market 20% · Model 53%
Largest signal contribution toward this side: lookahead_spot at -2.0 points of lean. Model and market win probabilities differ by 32 percentage points.
COASTAL CAROLINA +21.0+24 PTS APART
Market 9% · Model 33%
Largest signal contribution toward this side: rating_gap at -3.3 points of lean. Model and market win probabilities differ by 24 percentage points.
BETTING ANALYSIS
$100 $250 $500 $1000
Covers every FBS game with a posted line within |spread| ≤ 21. Edge = modeled cover probability minus the vig-implied break-even (52.4%) — a bet on the NUMBER, which is a different question than the win-probability divergences above. Stakes sized by quarter-Kelly, capped at 20% of the weekly budget.
BET $15 · Coastal Carolina +21.0 @ West Virginia edge +29.5LOG
Sat Sep 5 · 12:00 PM ET · model covers 81.9% · market needs 52.4% · breakdown ▾
BET $15 · Tulsa +13.0 vs Oklahoma State edge +29.0LOG
Sat Sep 5 · 3:45 PM ET · model covers 81.4% · market needs 52.4% · breakdown ▾
BET $14 · Toledo +9.5 @ Michigan State edge +28.4LOG
Fri Sep 4 · 8:00 PM ET · model covers 80.8% · market needs 52.4% · breakdown ▾
BET $9 · Clemson +10.0 @ LSU edge +17.8LOG
Sat Sep 5 · 7:30 PM ET · model covers 70.2% · market needs 52.4% · breakdown ▾
BET $8 · Oregon State +20.5 @ Houston edge +16.1LOG
Sat Sep 5 · 12:00 PM ET · model covers 68.5% · market needs 52.4% · breakdown ▾
BET $6 · California +2.0 vs UCLA edge +12.8LOG
Sat Sep 5 · 10:30 PM ET · model covers 65.2% · market needs 52.4% · breakdown ▾
BET $5 · Georgia Tech -6.5 vs Colorado edge +10.5LOG
Thu Sep 3 · 8:00 PM ET · model covers 62.9% · market needs 52.4% · breakdown ▾
BET $5 · Hawai'i +3.0 vs UNLV edge +10.1LOG
Sat Sep 5 · 10:00 PM ET · model covers 62.5% · market needs 52.4% · breakdown ▾
BET $4 · Miami (OH) +16.5 @ Pittsburgh edge +9.0LOG
Sat Sep 5 · 12:30 PM ET · model covers 61.4% · market needs 52.4% · breakdown ▾
BET $3 · Sam Houston +17.0 @ Troy edge +7.4LOG
Sat Sep 5 · 7:00 PM ET · model covers 59.8% · market needs 52.4% · breakdown ▾
remaining games: below edge threshold $16 UNALLOCATED
MY LOGGED BETS
ADD MY OWN BET — any game this week, tagged OWN so your picks and the model's stay separated
ADD
SAVE LOAD
LOG marks a bet you placed; totals update as results grade. Bets are stored in this browser; SAVE with a user + PIN keeps a copy on the site so LOAD works from any device. PIN protection is basic — don't reuse a PIN you care about.
Paper trading only. These are tracked model outputs, not wagering recommendations.
THE WEEK 1 SLATE · TOP 40 BY COMBINED ELO
#1 North Texas @ Indiana ATS SPLIT
Sat Sep 5 · line: Indiana -40.0 · ATS: North Texas +40.0
91%
INDIANA ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+16.6
rest_diff
0.0
opp_pass_def_trend
+0.1
rush_def_trend
−0.1
pass_off_trend
+0.4
lookahead_spot
+1.2
travel_km
+1.0
baseline
+0.7
LEAN +21.7→ 91% INDIANA
⚠ ATS side is NORTH TEXAS +40.0. The straight-up pick is Indiana; against the spread, the model favors the other side.
#2 Louisville @ Ole Miss ATS SPLIT
Sun Sep 6 · line: Ole Miss -7.0 · ATS: Louisville +7.0 · neutral site
65%
OLE MISS ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
0.0
rating_gap
+6.7
rest_diff
0.0
opp_pass_def_trend
+0.2
rush_def_trend
−0.0
pass_off_trend
−0.7
lookahead_spot
−1.2
travel_km
0.0
baseline
+0.7
LEAN +5.7→ 65% OLE MISS
⚠ ATS side is LOUISVILLE +7.0. The straight-up pick is Ole Miss; against the spread, the model favors the other side.
#3 Wisconsin @ Notre Dame ATS SPLIT
Sun Sep 6 · line: Notre Dame -20.5 · ATS: Wisconsin +20.5 · neutral site
88%
NOTRE DAME ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
0.0
rating_gap
+19.3
rest_diff
0.0
opp_pass_def_trend
+0.1
rush_def_trend
+0.0
pass_off_trend
−1.0
lookahead_spot
−0.8
travel_km
0.0
baseline
+0.7
LEAN +18.3→ 88% NOTRE DAME
⚠ ATS side is WISCONSIN +20.5. The straight-up pick is Notre Dame; against the spread, the model favors the other side.
#4 Liberty @ James Madison ATS SPLIT
Sat Sep 5 · line: James Madison -6.5 · ATS: Liberty +6.5
61%
JAMES MADISO ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+2.6
rest_diff
0.0
opp_pass_def_trend
+0.1
rush_def_trend
−0.0
pass_off_trend
−1.1
lookahead_spot
0.0
travel_km
+0.1
baseline
+0.7
LEAN +4.2→ 61% JAMES MADISON
⚠ ATS side is LIBERTY +6.5. The straight-up pick is James Madison; against the spread, the model favors the other side.
#5 Boise State @ Oregon ATS SPLIT
Sat Sep 5 · line: Oregon -24.5 · ATS: Boise State +24.5
83%
OREGON ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+12.4
rest_diff
0.0
opp_pass_def_trend
+0.0
rush_def_trend
−0.1
pass_off_trend
−0.6
lookahead_spot
−0.3
travel_km
+0.5
baseline
+0.7
LEAN +14.5→ 83% OREGON
⚠ ATS side is BOISE STATE +24.5. The straight-up pick is Oregon; against the spread, the model favors the other side.
#6 Washington State @ Washington ATS SPLIT
Sun Sep 6 · line: Washington -23.0 · ATS: Washington State +23.0
66%
WASHINGTON ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+4.9
rest_diff
0.0
opp_pass_def_trend
+0.2
rush_def_trend
−0.2
pass_off_trend
−1.1
lookahead_spot
−0.3
travel_km
+0.4
baseline
+0.7
LEAN +6.4→ 66% WASHINGTON
⚠ ATS side is WASHINGTON STATE +23.0. The straight-up pick is Washington; against the spread, the model favors the other side.
#7 Texas State @ Texas ATS SPLIT
Sat Sep 5 · line: Texas -29.5 · ATS: Texas State +29.5
78%
TEXAS ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+8.3
rest_diff
0.0
opp_pass_def_trend
+0.1
rush_def_trend
−0.3
pass_off_trend
−0.4
lookahead_spot
+1.6
travel_km
+0.0
baseline
+0.7
LEAN +11.8→ 78% TEXAS
⚠ ATS side is TEXAS STATE +29.5. The straight-up pick is Texas; against the spread, the model favors the other side.
#8 East Carolina @ Alabama ATS SPLIT
Sat Sep 5 · line: Alabama -28.0 · ATS: East Carolina +28.0
77%
ALABAMA ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+6.9
rest_diff
0.0
opp_pass_def_trend
+0.4
rush_def_trend
−0.0
pass_off_trend
−0.7
lookahead_spot
+1.3
travel_km
+0.9
baseline
+0.7
LEAN +11.2→ 77% ALABAMA
⚠ ATS side is EAST CAROLINA +28.0. The straight-up pick is Alabama; against the spread, the model favors the other side.
#9 Fresno State @ USC ATS SPLIT
Sat Sep 5 · line: USC -21.5 · ATS: Fresno State +21.5
77%
USC ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+9.0
rest_diff
−0.5
opp_pass_def_trend
+0.1
rush_def_trend
+0.1
pass_off_trend
−0.7
lookahead_spot
+0.2
travel_km
+0.3
baseline
+0.7
LEAN +11.1→ 77% USC
⚠ ATS side is FRESNO STATE +21.5. The straight-up pick is USC; against the spread, the model favors the other side.
#10 SMU @ Florida State
Mon Sep 7 · line: Florida State +3.0 · ATS: SMU -3.0
61%
SMU ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
−1.8
rating_gap
+6.8
rest_diff
+0.3
opp_pass_def_trend
+0.1
rush_def_trend
+0.0
pass_off_trend
+0.6
lookahead_spot
−0.0
travel_km
−1.1
baseline
−0.7
LEAN +4.3→ 61% SMU
#11 Clemson @ LSU ATS SPLIT
Sat Sep 5 · line: LSU -10.0 · ATS: Clemson +10.0
55%
LSU ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
−2.1
rest_diff
0.0
opp_pass_def_trend
−0.1
rush_def_trend
−0.2
pass_off_trend
+0.7
lookahead_spot
+0.1
travel_km
+0.8
baseline
+0.7
LEAN +1.7→ 55% LSU
⚠ ATS side is CLEMSON +10.0. The straight-up pick is LSU; against the spread, the model favors the other side.
#12 Marshall @ Penn State ATS SPLIT
Sat Sep 5 · line: Penn State -24.0 · ATS: Marshall +24.0
87%
PENN STATE ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+12.6
rest_diff
0.0
opp_pass_def_trend
+0.5
rush_def_trend
+0.1
pass_off_trend
+0.3
lookahead_spot
+1.2
travel_km
+0.4
baseline
+0.7
LEAN +17.7→ 87% PENN STATE
⚠ ATS side is MARSHALL +24.0. The straight-up pick is Penn State; against the spread, the model favors the other side.
#13 Miami @ Stanford ATS SPLIT
Sat Sep 5 · line: Stanford +24.5 · ATS: Stanford +24.5
77%
MIAMI ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
−1.8
rating_gap
+16.9
rest_diff
+0.5
opp_pass_def_trend
+0.1
rush_def_trend
−0.4
pass_off_trend
−0.0
lookahead_spot
+0.6
travel_km
−3.7
baseline
−0.7
LEAN +11.4→ 77% MIAMI
⚠ ATS side is STANFORD +24.5. The straight-up pick is Miami; against the spread, the model favors the other side.
#14 Western Michigan @ Michigan ATS SPLIT
Sat Sep 5 · line: Michigan -27.5 · ATS: Western Michigan +27.5
70%
MICHIGAN ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+6.2
rest_diff
0.0
opp_pass_def_trend
+0.3
rush_def_trend
−0.4
pass_off_trend
−1.2
lookahead_spot
+0.4
travel_km
+0.1
baseline
+0.7
LEAN +8.0→ 70% MICHIGAN
⚠ ATS side is WESTERN MICHIGAN +27.5. The straight-up pick is Michigan; against the spread, the model favors the other side.
#15 Tulane @ Duke ATS SPLIT
Sat Sep 5 · line: Duke -7.5 · ATS: Tulane +7.5
63%
DUKE ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+1.0
rest_diff
0.0
opp_pass_def_trend
−0.2
rush_def_trend
−0.1
pass_off_trend
−0.1
lookahead_spot
+0.5
travel_km
+1.1
baseline
+0.7
LEAN +4.8→ 63% DUKE
⚠ ATS side is TULANE +7.5. The straight-up pick is Duke; against the spread, the model favors the other side.
#16 Missouri State @ Texas A&M ATS SPLIT
Sat Sep 5 · line: Texas A&M -40.5 · ATS: Missouri State +40.5
85%
TEXAS A&M ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+14.2
rest_diff
0.0
opp_pass_def_trend
−0.1
rush_def_trend
−0.2
pass_off_trend
−1.2
lookahead_spot
+0.6
travel_km
+0.7
baseline
+0.7
LEAN +16.3→ 85% TEXAS A&M
⚠ ATS side is MISSOURI STATE +40.5. The straight-up pick is Texas A&M; against the spread, the model favors the other side.
#17 Baylor @ Auburn ATS SPLIT
Sat Sep 5 · line: Auburn -7.0 · ATS: Baylor +7.0 · neutral site
66%
AUBURN ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
0.0
rating_gap
+4.2
rest_diff
0.0
opp_pass_def_trend
−0.3
rush_def_trend
−0.1
pass_off_trend
+1.1
lookahead_spot
+0.6
travel_km
0.0
baseline
+0.7
LEAN +6.2→ 66% AUBURN
⚠ ATS side is BAYLOR +7.0. The straight-up pick is Auburn; against the spread, the model favors the other side.
#18 Northern Illinois @ Iowa ATS SPLIT
Sat Sep 5 · line: Iowa -31.0 · ATS: Northern Illinois +31.0
92%
IOWA ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+18.3
rest_diff
0.0
opp_pass_def_trend
+0.0
rush_def_trend
+0.2
pass_off_trend
+1.5
lookahead_spot
+0.7
travel_km
+0.2
baseline
+0.7
LEAN +23.3→ 92% IOWA
⚠ ATS side is NORTHERN ILLINOIS +31.0. The straight-up pick is Iowa; against the spread, the model favors the other side.
#19 Ball State @ Ohio State ATS SPLIT
Sat Sep 5 · line: Ohio State -51.0 · ATS: Ball State +51.0
97%
OHIO STATE ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+29.3
rest_diff
0.0
opp_pass_def_trend
+0.1
rush_def_trend
+0.0
pass_off_trend
−0.6
lookahead_spot
+1.8
travel_km
+0.2
baseline
+0.7
LEAN +33.5→ 97% OHIO STATE
⚠ ATS side is BALL STATE +51.0. The straight-up pick is Ohio State; against the spread, the model favors the other side.
#20 Florida International @ South Florida
Sat Sep 5 · line: South Florida -14.0 · ATS: South Florida -14.0
83%
SOUTH FLORID ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+11.7
rest_diff
0.0
opp_pass_def_trend
−0.1
rush_def_trend
−0.1
pass_off_trend
−0.6
lookahead_spot
+1.2
travel_km
0.0
baseline
+0.7
LEAN +14.6→ 83% SOUTH FLORIDA
#21 Toledo @ Michigan State
Sat Sep 5 · line: Michigan State -9.5 · ATS: Toledo +9.5
61%
TOLEDO ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
−1.8
rating_gap
+6.2
rest_diff
0.0
opp_pass_def_trend
+0.2
rush_def_trend
−0.2
pass_off_trend
−0.1
lookahead_spot
+0.6
travel_km
−0.1
baseline
−0.7
LEAN +4.1→ 61% TOLEDO
#22 UNLV @ Hawai'i
Sun Sep 6 · line: Hawai'i +3.0 · ATS: Hawai'i +3.0
55%
HAWAI'I ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
−4.4
rest_diff
+0.0
opp_pass_def_trend
−0.3
rush_def_trend
+0.0
pass_off_trend
+1.3
lookahead_spot
−1.2
travel_km
+4.0
baseline
+0.7
LEAN +2.0→ 55% HAWAI'I
#23 UAB @ Illinois ATS SPLIT
Fri Sep 4 · line: Illinois -28.0 · ATS: UAB +28.0
73%
ILLINOIS ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+4.7
rest_diff
0.0
opp_pass_def_trend
−0.1
rush_def_trend
+0.1
pass_off_trend
−0.3
lookahead_spot
+1.8
travel_km
+0.7
baseline
+0.7
LEAN +9.4→ 73% ILLINOIS
⚠ ATS side is UAB +28.0. The straight-up pick is Illinois; against the spread, the model favors the other side.
#24 UTEP @ Oklahoma ATS SPLIT
Sat Sep 5 · line: Oklahoma -41.5 · ATS: UTEP +41.5
92%
OKLAHOMA ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+17.6
rest_diff
0.0
opp_pass_def_trend
−0.0
rush_def_trend
+0.3
pass_off_trend
+0.1
lookahead_spot
+1.4
travel_km
+0.8
baseline
+0.7
LEAN +22.7→ 92% OKLAHOMA
⚠ ATS side is UTEP +41.5. The straight-up pick is Oklahoma; against the spread, the model favors the other side.
#25 Colorado @ Georgia Tech
Fri Sep 4 · line: Georgia Tech -6.5 · ATS: Georgia Tech -6.5
78%
GEORGIA TECH ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+6.5
rest_diff
0.0
opp_pass_def_trend
−0.2
rush_def_trend
+0.3
pass_off_trend
−0.3
lookahead_spot
+1.0
travel_km
+1.8
baseline
+0.7
LEAN +11.6→ 78% GEORGIA TECH
#26 Miami (OH) @ Pittsburgh ATS SPLIT
Sat Sep 5 · line: Pittsburgh -16.5 · ATS: Miami (OH) +16.5
78%
PITTSBURGH ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+8.3
rest_diff
0.0
opp_pass_def_trend
−0.1
rush_def_trend
+0.2
pass_off_trend
+0.5
lookahead_spot
+0.3
travel_km
+0.4
baseline
+0.7
LEAN +12.0→ 78% PITTSBURGH
⚠ ATS side is MIAMI (OH) +16.5. The straight-up pick is Pittsburgh; against the spread, the model favors the other side.
#27 Arkansas State @ Memphis ATS SPLIT
Sat Sep 5 · line: Memphis -11.0 · ATS: Arkansas State +11.0
69%
MEMPHIS ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+6.4
rest_diff
−0.4
opp_pass_def_trend
+0.2
rush_def_trend
−0.2
pass_off_trend
−1.4
lookahead_spot
+0.5
travel_km
+0.1
baseline
+0.7
LEAN +7.5→ 69% MEMPHIS
⚠ ATS side is ARKANSAS STATE +11.0. The straight-up pick is Memphis; against the spread, the model favors the other side.
#28 Ohio @ Nebraska ATS SPLIT
Sat Sep 5 · line: Nebraska -24.0 · ATS: Ohio +24.0
73%
NEBRASKA ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+6.5
rest_diff
0.0
opp_pass_def_trend
+0.4
rush_def_trend
−0.2
pass_off_trend
−1.1
lookahead_spot
+0.0
travel_km
+1.1
baseline
+0.7
LEAN +9.2→ 73% NEBRASKA
⚠ ATS side is OHIO +24.0. The straight-up pick is Nebraska; against the spread, the model favors the other side.
#29 Boston College @ Cincinnati ATS SPLIT
Sat Sep 5 · line: Cincinnati -7.0 · ATS: Boston College +7.0
64%
CINCINNATI ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+2.9
rest_diff
0.0
opp_pass_def_trend
+0.4
rush_def_trend
−0.0
pass_off_trend
−1.3
lookahead_spot
+0.0
travel_km
+1.1
baseline
+0.7
LEAN +5.5→ 64% CINCINNATI
⚠ ATS side is BOSTON COLLEGE +7.0. The straight-up pick is Cincinnati; against the spread, the model favors the other side.
#30 Florida Atlantic @ Florida ATS SPLIT
Sat Sep 5 · line: Florida -26.5 · ATS: Florida Atlantic +26.5
81%
FLORIDA ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+10.8
rest_diff
0.0
opp_pass_def_trend
−0.0
rush_def_trend
−0.1
pass_off_trend
+0.1
lookahead_spot
+0.0
travel_km
+0.4
baseline
+0.7
LEAN +13.7→ 81% FLORIDA
⚠ ATS side is FLORIDA ATLANTIC +26.5. The straight-up pick is Florida; against the spread, the model favors the other side.
#31 Central Michigan @ New Mexico
Sun Sep 6 · line: New Mexico -10.0 · ATS: New Mexico -10.0
75%
NEW MEXICO ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+6.8
rest_diff
0.0
opp_pass_def_trend
−0.2
rush_def_trend
−0.1
pass_off_trend
−0.5
lookahead_spot
0.0
travel_km
+1.9
baseline
+0.7
LEAN +10.5→ 75% NEW MEXICO
#32 Kent State @ South Carolina ATS SPLIT
Sat Sep 5 · line: South Carolina -36.5 · ATS: Kent State +36.5
93%
SOUTH CAROLI ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+18.1
rest_diff
0.0
opp_pass_def_trend
+0.2
rush_def_trend
+0.2
pass_off_trend
+0.6
lookahead_spot
+1.8
travel_km
+0.7
baseline
+0.7
LEAN +24.1→ 93% SOUTH CAROLINA
⚠ ATS side is KENT STATE +36.5. The straight-up pick is South Carolina; against the spread, the model favors the other side.
#33 Western Kentucky @ Nevada ATS SPLIT
Sun Sep 6 · line: Nevada +2.5 · ATS: Nevada +2.5
51%
WESTERN KENT ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
−1.8
rating_gap
+3.4
rest_diff
0.0
opp_pass_def_trend
+0.2
rush_def_trend
+0.1
pass_off_trend
+0.6
lookahead_spot
+1.3
travel_km
−2.6
baseline
−0.7
LEAN +0.5→ 51% WESTERN KENTUCKY
⚠ ATS side is NEVADA +2.5. The straight-up pick is Western Kentucky; against the spread, the model favors the other side.
#34 UCLA @ California
Sun Sep 6 · line: California +2.0 · ATS: California +2.0
61%
CALIFORNIA ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+1.0
rest_diff
0.0
opp_pass_def_trend
−0.1
rush_def_trend
−0.2
pass_off_trend
+0.8
lookahead_spot
−0.5
travel_km
+0.5
baseline
+0.7
LEAN +4.1→ 61% CALIFORNIA
#35 Oregon State @ Houston ATS SPLIT
Sat Sep 5 · line: Houston -20.5 · ATS: Oregon State +20.5
80%
HOUSTON ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+9.4
rest_diff
0.0
opp_pass_def_trend
+0.1
rush_def_trend
+0.2
pass_off_trend
−0.4
lookahead_spot
−1.5
travel_km
+2.7
baseline
+0.7
LEAN +13.0→ 80% HOUSTON
⚠ ATS side is OREGON STATE +20.5. The straight-up pick is Houston; against the spread, the model favors the other side.
#36 San José State @ Eastern Michigan
Fri Sep 4 · line: Eastern Michigan -3.0 · ATS: Eastern Michigan -3.0
64%
EASTERN MICH ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
−1.8
rest_diff
−0.0
opp_pass_def_trend
−0.1
rush_def_trend
+0.2
pass_off_trend
+1.4
lookahead_spot
−0.0
travel_km
+3.0
baseline
+0.7
LEAN +5.2→ 64% EASTERN MICHIGAN
#37 UL Monroe @ Mississippi State ATS SPLIT
Sat Sep 5 · line: Mississippi State -28.5 · ATS: UL Monroe +28.5
85%
MISSISSIPPI ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+11.7
rest_diff
0.0
opp_pass_def_trend
+0.6
rush_def_trend
+0.3
pass_off_trend
−0.4
lookahead_spot
+0.9
travel_km
+0.3
baseline
+0.7
LEAN +15.9→ 85% MISSISSIPPI STATE
⚠ ATS side is UL MONROE +28.5. The straight-up pick is Mississippi State; against the spread, the model favors the other side.
#38 Wyoming @ Colorado State ATS SPLIT
Sat Sep 5 · line: Colorado State -3.0 · ATS: Wyoming +3.0
51%
COLORADO STA ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
−3.5
rest_diff
0.0
opp_pass_def_trend
+0.6
rush_def_trend
+0.0
pass_off_trend
+0.6
lookahead_spot
0.0
travel_km
+0.1
baseline
+0.7
LEAN +0.4→ 51% COLORADO STATE
⚠ ATS side is WYOMING +3.0. The straight-up pick is Colorado State; against the spread, the model favors the other side.
#39 Coastal Carolina @ West Virginia ATS SPLIT
Sat Sep 5 · line: West Virginia -21.0 · ATS: Coastal Carolina +21.0
67%
WEST VIRGINI ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+3.3
rest_diff
0.0
opp_pass_def_trend
−0.1
rush_def_trend
+0.0
pass_off_trend
+0.4
lookahead_spot
0.0
travel_km
+0.6
baseline
+0.7
LEAN +6.7→ 67% WEST VIRGINIA
⚠ ATS side is COASTAL CAROLINA +21.0. The straight-up pick is West Virginia; against the spread, the model favors the other side.
#40 Sam Houston @ Troy ATS SPLIT
Sat Sep 5 · line: Troy -17.0 · ATS: Sam Houston +17.0
80%
TROY ▾
SIGNAL BREAKDOWN — CONTRIBUTIONS SUM TO THE LEAN; PROBABILITY IS DERIVED FROM THE LEAN
home_field
+1.8
rating_gap
+10.1
rest_diff
0.0
opp_pass_def_trend
−0.2
rush_def_trend
+0.1
pass_off_trend
−0.7
lookahead_spot
+0.5
travel_km
+0.8
baseline
+0.7
LEAN +13.1→ 80% TROY
⚠ ATS side is SAM HOUSTON +17.0. The straight-up pick is Troy; against the spread, the model favors the other side.
SIGNAL EVALUATION
15 CANDIDATES EVALUATED
A signal is added to the model only if it improves held-out predictions across 3+ seasons with a stable coefficient sign. Rejected signals are retained with their evidence.
✓ ACTIVE — IN THE MODEL
home_field rating_gap rest_diff opp_pass_def_trend rush_def_trend pass_off_trend lookahead_spot travel_km
UNDER EVALUATION
talent_gap 2/3 returning_production 2/3 line_movement 0/3
✗ REJECTED — EVIDENCE RETAINED
rush_off_trend — no lift on held-out seasons (mean Brier delta -0.00003)
turnover_luck — no lift on held-out seasons (mean Brier delta -0.00002)
letdown_spot — no lift on held-out seasons (mean Brier delta -0.00014)
body_clock — no lift on held-out seasons (mean Brier delta -0.00014)
SCOREBOARD
2026 season vs closing line (ATS)0–0 · —
straight-up0–0 (context: favorites win ~70% of these games)
career Brier (2021+).190 baseline .191
career vs closing line1,954–1,915–75 (50.5%)
CALIBRATION
40–50%
47.1% n=444 50–60%
52.6% n=521 59–69%
65.0% n=549 70–80%
72.4% n=557 80–90%
85.6% n=536 90–100%
93.0% n=359
Career held-out calibration, 2021 onward. Bars show actual win rate per predicted-probability bucket.
HISTORICAL PERFORMANCE — BACKTEST VS RESULTS
Every season below was predicted walk-forward: the model only used data from seasons prior to the one being predicted.
SEASONGAMESSTRAIGHT-UPVS THE LINEATS %BRIERML ROI (FLAT $)
2016 76070.3%358–382–2048.4%.190 no ML data
2017 77670.4%390–365–2151.7%.185 no ML data
2018 77272.4%391–361–2052.0%.175 no ML data
2019 77473.1%377–384–1349.5%.176 no ML data
2020 (COVID — excluded from headline metrics) 53471.2%274–251–952.2%.187 no ML data
2021 77069.5%383–378–950.3%.187 -4.0%
2022 77668.7%389–374–1351.0%.199 -4.6%
2023 79272.3%395–381–1650.9%.185 -1.2%
2024 79869.2%391–388–1950.2%.193 -5.4%
2025 80872.5%396–394–1850.1%.184 +1.0%
Moneyline analysis, 3,771 games (2021-2025): flat bets on every straight-up pick returned -2.8%. Two subsets were positive: model-backed underdogs (470 bets, +11.4%) and picks with a ≥5-point modeled edge (968 bets, +2.9%). Both samples are small and were identified alongside other tested strategies, so they are tracked as paper-trading experiments pending live confirmation.
SIGNAL LAB — SUBSET BACKTESTS
TOGGLE SIGNALS · 10 SEASONS OF PRECOMPUTED RESULTS
HOME FIELD ✓ RATING GAP ✓ REST DIFF ✓ OPP PASS DEF TREND ✓ RUSH DEF TREND ✓ PASS OFF TREND ✓ LOOKAHEAD SPOT ✓ TRAVEL KM ✓
—
ATS WIN RATE
—
BRIER SCORE
—
10-SEASON ATS
ROBOT POLL · regenerated weekly · predictions lock at publication and are not edited
2020 excluded from headline metrics · paper trading only · not wagering advice