Edge Labs EDGE LABS Market intelligence
CFB NFL
← The slate
CFB · Week 2 · Sat, Sep 12 · 12:00 PM ET · Boone Pickens Stadium · FINAL Oregon 31, Oklahoma State 39

Oregon at Oklahoma State. Our picks: Oregon -22.5 (high). Under 57.5 (medium).

The picks come first on this page. The market's number, our number and the gap between them sit under them as MODEL CONTEXT: that is the input our analyst panel argued from, not a call of its own. Tiers are how decisively the evidence agreed, never a win probability.

ATS pick
Oregon -22.5
high free LOSS
locked Tue 12:38am ET
locked -22.5 · closed -23.5
Oregon 31, Oklahoma State 39
free pick, released Tue 9:07am ET
Total pick
Under 57.5
medium LOSS
locked 57.5 · closed 55.5
panel 3-0

Panel reasoning

On the spread

Oregon at Oklahoma State, Oregon -22.5. Oklahoma State lost starting LT Jacob Sexton for the season on the opening drive at Tulsa and lost the game 24-10 with four turnovers and one third-down conversion in eleven; the replacements at tackle struggled in relief, and Drew Mestemaker has seven interceptions over his last stretch. Oregon put up 497 yards on Boise State with Dante Moore throwing for 378, gets WR Iverson Hooks back and is without Jalen Lott. The line moved from -17.5 and -18.5 to -22.5 (-23.5 at Pinnacle) on 95 percent of the handle. Our number (Oregon -40.8) is not credible and grades a roster that no longer exists, so it is not the reason. One panel took Oklahoma State, and its case deserves a hearing: Eric Morris's North Texas offense led FBS in scoring last year, only 24 players remain from the 69-3 team, and five points of steam are already in the number. Take Oregon -22.5 at DraftKings or FanDuel, not -23.5.

On the total

Under 57.5. DraftKings and FanDuel opened 60.5 and every book sits 57.5 while 62 percent of the money is on the over, a three-point move against the money. Oklahoma State's protection and turnovers, an Oregon lead that can slow the fourth quarter, and a 40 percent storm chance all point down. Our totals model's 48.4 is built on a 13 mph wind input the forecast does not support (about 6 mph), so treat it as direction only. The over case is Oregon's 497 yards and a transfer-built Cowboys offense. Under 57.5 everywhere.

Model context

The numbers the panel argued from. Our models price a game, the panel makes the call, and the picks above are the call. Nothing in this block is a pick.

Locked line
Oregon -22.5 · Oklahoma State +22.5
the spread we locked at, it does not move
Our model number
Oregon by 40.8
the calibrated EL CFB rating number, home field included
Spread gap
16.3 points
the distance between our model number and the market, on the Oregon -24.5 side. Context for the panel, not a call of its own.
Market total
55.5
consensus of 5 books at the model run, Sep 12, 12:38 PM ET
Our total
49.7
EL CFB Totals v1
Total gap
-5.8
the distance between our model total and the market total. Context for the panel, not a call of its own.

Context, not a call. The spread number is our rating difference plus home field. The totals number is a registered candidate that loses to the closing total: across 2,264 held out games it missed the combined score by 12.862 points on average and the closing total missed by 12.635, ahead of us in all 3 of those seasons, so it never posts to the board and never enters the record on its own.

Over and under splits

Descriptive context on this game, every line with the games it counts. None of it is a reason a pick wins and none of it enters a tier.

  • Early-season total of 55.5 fits the validated under window, unders 55% historically (n=943)
    total · under · moderate · n=943
    H64 (validated): Aug through mid-Sep games with a closing total of 47+ landed under 55% of 943 games, 2015-2025. Recent seasons: 2022: 58% (n=92) · 2023: 53% (n=81) · 2024: 57% (n=103) · 2025: 52% (n=119). 2025 ran soft at 52.1%, below the -110 breakeven, and the 2026 forward test is pre-committed on the early-totals page. Early-season totals are shaded up by public over-action; the market corrects as real form data arrives.
  • Oklahoma State games go OVER 61% at home (n=71)
    total · over · moderate · n=71
    Multi-season split: Oklahoma State at home landed over 61% of 71 gradable games (2013-2025). A genuine venue tendency, one of several reasons to lean over, not a standalone edge.
  • Oregon games go UNDER 61% on the road (n=64)
    total · under · moderate · n=64
    Multi-season split: Oregon on the road landed under 61% of 64 gradable games (2013-2025). A genuine venue tendency, one of several reasons to lean under, not a standalone edge.

The panel

Panel A

Oregon -22.5

Under 57.5

Take Oregon -22.5, available in the dossier’s freshest FanDuel/DraftKings snapshot and corroborated by BetMGM’s public listing. EL’s -40.8 is not credible as a fair price: calibration exaggerates the mismatch, and Oklahoma State’s coaching and roster inputs are materially stale. My judgment is closer to Oregon -25 to -27, with substantial uncertainty. The strongest matchup evidence is Oklahoma State losing starting left tackle Jacob Sexton for the season after already struggling with protection against Tulsa. Oregon’s passing attack generated 378 yards in its opener, suggesting its seven-point victory understates offensive effectiveness. However, the market has already moved substantially toward Oregon, so the injury is not an undiscovered advantage. Morris’s rebuilt roster also deserves more respect than a mechanical extrapolation of 2025. The early road kickoff and potential late Oklahoma State scores make laying more than three touchdowns uncomfortable. This is a moderate lean based on protection and offensive quality, not the model’s advertised 18-point edge.

Total. Under 57.5 is a modest lean; my judgment is roughly 54, above EL’s 48.38. Oklahoma State’s protection problems and Sexton’s confirmed absence threaten sustained drives, and an Oregon lead could eventually reduce offensive urgency. The over case is substantial: Oregon just produced 497 yards, Morris’s transfer-built offense is poorly represented by last year’s numbers, and turnovers can create short fields rather than suppress scoring. Oregon also allowed 27 points in its opener. The market has already removed three points from the earliest 60.5 totals, reducing the under’s value. Forecasts suggest heat, moderate wind and possible storms, but disagree enough that I would not apply another mechanical weather deduction. Protection concerns narrowly outweigh explosive-play and turnover risks.

Where our model looks wrong

Oklahoma State's first_year_head_coach=false is wrong: Eric Morris is in year one. Zero returning production and portal net -9 inadequately describe 91 newcomers and a highly ranked incoming class. Last season's poor scoring therefore deserves much less weight. Sexton is a starting left tackle and season-ending absence, not merely a questionable guard. The spread fields also mix margin and spread signs; ratings imply an approximately 33.4-point Oregon raw advantage, which calibration stretches to 40.8 without demonstrated ATS value. The totals documentation says no weather, yet an explicit wind adjustment subtracts 2.616 points; that inconsistency needs correction.

Panel B

Oklahoma State +22.5

Under 57.5

Our number (Oregon -40.8) sits 18 points past a market that has already moved five points its way, from -18.5 at DraftKings on 9/1 to -22.5/-23.5 on 9/9. That gap is the model's documented failure mode, not an edge: the rating is still mostly Oklahoma State's 1-11 prior and last year's 69-3, and only 24 players remain from that roster (SI, 9/9). The calibration map also stretches a -33.4 raw margin to -40.8, exactly the big-favorite band the dossier calls least reliable. The market has correctly priced the bad news, the 24-10 loss at Tulsa with four turnovers and 1-of-11 on third down, and Jacob Sexton out for the season (Yahoo/Eric Morris, 9/8). What it prices less well is Oregon's side: a 9am PT body-clock road opener in mid-80s heat that Lanning himself flagged (9/7), after a 34-27 home escape in which Oregon is 0-1 ATS (NBC Sports, 9/9). Morris's North Texas system led FBS in scoring at 45.1 ppg; Mestemaker's arm produces the garbage-time points that kill three-touchdown covers. Take the points.

Total. DraftKings and FanDuel opened 60.5 on 9/1 and every one of the five books sits 57.5 on 9/9, while 62.31% of the money is on the Over (SportsBettingDime, 9/7). A three-point move against the money is the cleanest signal in this game. Our model's 48.4 is directionally right but built on a 13.3 mph wind input the NWS does not support: the Stillwater hourly for Saturday shows S 6-8 mph and 79-86F through the first half (checked 9/9), so the honest model number is nearer 51 and the edge is smaller than the dossier claims. The floor case is the real one. Oklahoma State scored 10 at Tulsa, went 1-of-11 on third down and just lost its best lineman for the season; a team that cannot stay on the field caps the total from below even inside a blowout.

Where our model looks wrong

Two places. First, the rating is grading a team that no longer exists: Oklahoma State returns 24 players from the roster that lost 69-3, and 91 newcomers under a first-year staff running a different offense are invisible to a rating that learns from final margins only. The -14.4 in-season value is essentially the 1-11 prior. Second, the calibration map turns a -33.4 raw margin into -40.8, 7.4 points of stretch in the exact big-favorite regime the dossier flags as least reliable, and the held-out record is 46-50% ATS in every gap bucket. A 17.9-point gap here should read as 'the model has no opinion,' not 'high tier.' The tier label is the thing that looks most wrong. Third, smaller: Oregon still sits at 21.5 after a 34-27 escape it failed to cover, so the Week 1 refit absorbed almost nothing. And the injuries table is stale on the single most important item in the game (Sexton is out for the year, not questionable).

Panel C

Oregon -22.5

Under 57.5

Our model's calibrated Oregon -40.8 is not a credible price on its own, it has zero proven ATS edge in any gap bucket historically (46-50%) and is built off one Week 1 sample for a rating that doesn't see injuries, coaching turnover, or depth. Discount the magnitude heavily, but the direction is well supported by live reporting: Oklahoma State lost starting LT Jacob Sexton for the season (confirmed by HC Eric Morris, 9/8), on top of a genuinely bad 10-24 loss at Tulsa with four turnovers and 50% red-zone execution under a first-year staff. Our own injuries table still under-reports this, listing Sexton as merely 'Questionable.' Oregon looked explosive in its opener (497 total yards, 378 passing) with no real QB concern, and gets Iverson Hooks back from a minor hamstring tweak. The market has moved from -18.5 to -22.5/-23.5 even with public tickets reportedly split toward the dog, sharp money confirming the mismatch, not public overreaction. I'll lay the number, but a 3-plus-possession road spread carries real backdoor-cover risk once Oregon empties the bench, so confidence stays moderate.

Total. Our totals model reads 48.38 against a 57.5 market, a large gap that still falls short of its own 12.86-point lean threshold, so treat it as a soft signal rather than a call on its own. Real-world evidence points the same way: OSU scored just 10 points on 50% red-zone execution at Tulsa and now plays without starting LT Jacob Sexton (out for the season, confirmed 9/8), which should further slow a struggling offense's ability to sustain drives. Oregon's offense is explosive enough (497 yards in Week 1) to score fast, but a 3-score-plus spread raises the odds Dan Lanning empties the bench in the second half rather than chase garbage-time points, capping the total. Books have already trimmed the total from an opening 58-60.5 range down to 57.5, agreeing with this lean. Main over risk: Oregon's defense allowed 27 to Boise State in Week 1, and OSU's backups may be no better defensively than its starters once Oregon's reserves enter.

Where our model looks wrong

Two concrete errors: OSU's preseason input flags first_year_head_coach=false, but Eric Morris is in fact in his first season after Gundy's in-season firing last year, that understates the roster/scheme uncertainty around this team. More importantly, our own injuries feed lists Jacob Sexton as 'Questionable - Leg' as of 9/9, when OSU's staff confirmed on 9/8 he is out for the season; the single most consequential injury in this game is under-reported in our own data. Separately, the model's -40.8 calibrated number is likely a one-game overreaction to OSU's Tulsa margin layered onto an already-extreme preseason gap, and it carries no demonstrated ATS success at this or any gap size, treat the size of the model's lean as suspect even where the direction checks out.

The write-up

The case

Our model makes Oregon a 40.8 point favorite on the road, after giving Oklahoma State 2.5 points of home field. The market had Oregon by 22.5 when we made the pick. The gap is 18.3 points, and it points to Oregon -22.5. Locked Tue 12:38am ET, and it grades at that number.

Confidence: high

Confidence is a decisiveness label, not a win probability. It says how firmly the gap and the independent signals agree. This one runs on an 18.3 point gap, past the 11 point bar that reads high on the gap alone. No independent signal picked a side on the spread here, so the read rests on the gap.

The independent signals

The trend engine has 4 independent reads on this game, none of them on the side. Rating-adjacent categories are left out, so none of these is our own number in different words.

  • Early-season total of 54.5 fits the validated under window, unders 55% historically (n=943). On the total, not the side (under).
  • Oklahoma State games go OVER 61% at home (n=71). On the total, not the side (over).
  • Oregon games go UNDER 61% on the road (n=64). On the total, not the side (under).
  • Oregon favored by 23.5, blowout-range spread. Context, no side taken.

Trends

Descriptive context, each line with the games it counts. Filters like these did not hold up as predictors in testing, so none of this sits in the confidence read and none of it is a reason the pick wins. Situational splits look back 3 seasons and every line states its own span.

  • Oregon: won 4 of their last 5 (n=5, 2025 to 2026)
  • Oregon: 12-0 straight up in night divisional games (n=12, 2024 to 2025)
  • Oklahoma State: won 0 of their last 5 (n=5, 2025 to 2026)
  • Oklahoma State: lost 12 straight
  • Oklahoma State: 0-18 straight up in divisional games (n=18, 2024 to 2025)

Scoring profiles

2025: Oregon scored 36.9 and allowed 17.9 per game (n=15). Oklahoma State scored 14.2 and allowed 33.3 per game (n=12). 2026 to date: Oregon scored 34.0 and allowed 27.0 per game (n=1). Oklahoma State scored 10.0 and allowed 24.0 per game (n=1).

written before kickoff, frozen at kickoff

Trends

Descriptive context, every line with the games it counts. Filters like these did not hold up as predictors in our testing, so none of this is in the confidence read and none of it is a reason a pick wins. Situational splits look back 3 seasons and need at least 8 decided games to be shown at all. Records against the number and on the total join these once our closing-line history is restored; the straight-up splits are live now.

Oregon
  • won 4 of their last 5 (n=5, 2025 to 2026)
  • 12-0 straight up in night divisional games (n=12, 2024 to 2025)
Oklahoma State
  • won 0 of their last 5 (n=5, 2025 to 2026)
  • lost 12 straight
  • 0-18 straight up in divisional games (n=18, 2024 to 2025)
The rest of the splits we can compute on this game
Oregon
  • won 9 of their last 10 (n=10, 2025 to 2026)
  • 10-0 straight up on the road (n=10, 2024 to 2025)
  • 9-0 straight up in home night games (n=9, 2024 to 2025)
  • 9-0 straight up in road divisional games (n=9, 2024 to 2025)
  • 18-1 straight up in divisional games (n=19, 2024 to 2025)
  • 15-1 straight up at home (n=16, 2024 to 2026)
  • 15-1 straight up in night games (n=16, 2024 to 2025)
  • 8-1 straight up in home divisional games (n=9, 2024 to 2025)
  • 12-2 straight up in day games (n=14, 2024 to 2026)
Oklahoma State
  • won 0 of their last 10 (n=10, 2025 to 2026)
  • 0-17 straight up in divisional games after a straight up loss (n=17, 2024 to 2025)
  • 0-15 straight up in day divisional games (n=15, 2024 to 2025)
  • 0-15 straight up in day games after a straight up loss (n=15, 2024 to 2026)
  • 0-10 straight up on the road after a straight up loss (n=10, 2024 to 2026)
  • 0-9 straight up in home divisional games (n=9, 2024 to 2025)
  • 0-9 straight up in road divisional games (n=9, 2024 to 2025)
  • 1-19 straight up after a straight up loss (n=20, 2024 to 2026)
  • 1-11 straight up on the road (n=12, 2024 to 2026)
  • 1-9 straight up at home after a straight up loss (n=10, 2024 to 2025)
  • 1-9 straight up in road day games (n=10, 2024 to 2026)
  • 3-17 straight up in day games (n=20, 2024 to 2026)
  • 2-8 straight up in home day games (n=10, 2024 to 2025)
  • 3-10 straight up at home (n=13, 2024 to 2025)

The signals and the work

Members

7 independent signals on this game

The signal tally per side, the totals read, every angle with its sample size, and the line movement from open to now. 3 picks are released free on the slate every week. $24/mo, founder rate.

The two numbers underneath

TeamEL ratingConference
Oregon 21.544 Big Ten
Oklahoma State -14.401 Big 12

Rating difference plus 2.5 points of home field is the rating read above. Full board: the power ratings.

Conditions at kickoff

Clear. 87F · wind 7 mph

Provenance: schedule and finals from the Edge Labs database (2026 season); the line is the latest capture for this game with the book named; our number is the EL rating read (in-season once a team has played, preseason before that) plus 2.5 home field, none at neutral sites, and it is context on this page rather than the call. Signals come from the trend engine, each with its real sample; a tier is how decisively the evidence agreed and is never a win probability. Once the game kicks off this page stops computing and reads our pick lock ledger instead: the pick and the spread we were locked at, written once at kickoff, never updated, graded against that same number. Line movement comes from our permanent capture log, one book across both ends and never a capture taken at or after kickoff, so the second number is the last pregame line and after kickoff it is the close; a game the log has captured only once shows no movement rather than an invented one. Weather is the captured forecast for the venue, and the chip appears only when it is worth saying (wind at 12 mph or more, a 50 percent or better chance of rain, or 35F or colder), never on an indoor venue. Injury counts are our latest daily scan, real report rows only, and a team the scan does not cover is left out rather than shown as zero. The write-up is assembled from those same stored rows, never written around them: the case comes from the lock row, the signals from the trend engine with their own samples, the series from our game database from 2013 forward, and the scoring profiles from completed games only, each with the number of games it averages. A section with no data behind it is left out instead of filled in, it refreshes while the game is pregame, and it can never be edited once the game has kicked off. Trends are descriptive only, computed by the same shared module the write-up uses so the two cannot disagree: straight-up splits come from completed games in our database, situational splits look back three seasons and need at least eight decided games, records against the number and on the total arrive with the closing-line restore, and every line carries the games it counts. Filters like these did not hold up as predictors in our testing, so none of them enter the confidence read. The totals block is a read and not a pick: our number is EL CFB Totals v1 (edgelabs.el_cfb_totals, season 2026, model_version v1, docs/el-cfb-totals-v1.md), a registered candidate that loses to the closing total, so it never posts to the board or the record; the market total beside it is the same capture the spread comes from before kickoff and the frozen consensus the model priced against after it, and the gap is measured against whichever number is printed. The totals trends under it are the trend engine's own rows for this game, deduped on the headline, each carrying its sample. On the games the analyst panel works each week (the games our model prices off the market plus the ranked and Power Four games, at most twenty), the pick and the total call are the panel's synthesis: three independent analyst panels, the same evidence pack, their own research, one side each, tiered by how firmly they agreed, never a win probability. A panel majority on a side is what makes a pick; without one the game is a published no-pick, printed as No pick rather than left blank. Research and context, never a guarantee, 21+.