The NFL Week 1 Totals Problem: Cutdown Day Opens a Short Repricing Window
At 6 PM ET today, every NFL team must submit its 53-man roster. Each club cuts as many as 37 players off a 90-man summer limit. Across 32 teams, roughly 1,200 player transactions happen inside a single afternoon. The depth charts bettors, DFS players, and sportsbook traders have been reading all August get rewritten in real time.
This is the most concentrated information event of the NFL calendar before Week 1 kickoff. And the Week 1 totals board has been sitting largely unchanged since May. Today is the day those numbers start to move.
But there is a second problem built into Week 1 totals that roster cuts alone do not explain. The market has a structural pricing error that shows up every year, and academic research has identified exactly why it persists.
The Pattern Is Real
Since the NFL moved the extra point back to the 15-yard line in 2015, Week 1 games have gone under the total at a rate of 96-79. That is a 54.9% hit rate. Going further back, the under rate in Week 1 since 2005 sits at 55%.
A 54.9% win rate at -110 generates consistent profit. The real question is not whether the pattern exists, it is why it persists in a market that has had a decade to price it in.
Two academic papers provide the mechanism. Together they explain why the Week 1 board is systematically softer than any other week of the season.
Holdover Bias: What the Research Says
In 2013, Fodor and DiFilippo published "Inefficient pricing from holdover bias in NFL point spread markets" in Applied Financial Economics. Their data covered 2004 through 2012, and the finding is stark.
Prior playoff teams win only 35.6% of their opening week games against the spread when they face non-playoff teams. Bettors and linemakers alike overweight what they saw from a team in November and December of the prior season. The result: the playoff team gets priced too high, and the non-playoff team covers at a rate that generates average returns above 22% per game over the study period.
The mechanism is straightforward. Bettors form strong perceptions of teams late in the prior season, when the stakes are highest and the viewership is biggest. Those perceptions linger into the offseason. Linemakers, aware of where public money will flow, shade toward what their customers believe. The spread ends up reflecting last year's version of both teams, not the current one.
A 2025 paper in Economics Letters (Patterson, Shank, and Fodor) extends this analysis with anchoring bias evidence. The authors found that sportsbooks continue to incorporate pre-season Super Bowl odds into their closing lines throughout the entire season. Both bettors and sportsbooks anchor to the pre-season market. The result is a pricing inefficiency that persists into Week 1 and, to a diminishing degree, into the first two months of the regular season.
Why Totals Are More Exposed Than Spreads
Spreads adjust faster. A quarterback change, a coordinator departure, a key free agent loss, all of these move spreads within days because the public and the sharp money both react to headline news. Totals are slower. They depend on projected scoring output, which is a product of offensive scheme, personnel depth, and defensive adjustments that take weeks of preseason practice to evaluate.
Books set Week 1 totals in May, shortly after the schedule releases. The projection at that point uses last year's scoring rates, adjusted for known offseason changes. Five months of preseason camp, injuries, and scheme installs happen between that opening number and today. The total on the board right now reflects May's assumptions. The roster cuts happening this afternoon are the last major information update before Week 1 kicks off September 9.
Gray and Gray's 1997 Journal of Finance paper on NFL market efficiency found that probit-based betting strategies generated statistically significant in-sample profits, particularly when betting on less-publicized games and situations with informational lag. Week 1, with a five-month pricing gap and major roster uncertainty, fits that description more than any other week.
The 2026 Board: Three Games to Watch
The Week 1 slate kicks off Wednesday, September 9 with the Seahawks hosting the Patriots at 8:20 PM ET. The Rams face the 49ers in Melbourne on Thursday morning. The rest of the conference schedule fills out that weekend. Here are the three most relevant cases from a totals standpoint.
How Cutdowns Move Totals
Not every cut moves the line. A backup linebacker being released does not change the total. Three categories of cuts are relevant for totals today.
An offensive skill-player cut at the top of the depth chart moves a total down. A starter-level receiver, tight end, or running back who unexpectedly clears waivers tells you the team's Week 1 offensive projection was built on a player who is no longer there. Books adjust within minutes of the transaction wire.
A secondary defensive cut moves a total up. If a team releases a starting corner or safety, the opponent's receiving corps now has a better matchup in Week 1. The total should reflect a higher projected offensive output from the team with the easier coverage assignment.
A depth cut at an interior offensive line position is the subtlest signal. Books frequently underweight O-line depth changes in totals, since the connection between depth chart quality and projected points is less obvious to the public. Sharp bettors who track this category get the best prices because the market lags.
The Two-Hour Window
The 6 PM deadline means the transaction wire gets busy between 4 PM and 6 PM as teams file their final rosters. Books watch the wire and update lines within 10 to 20 minutes of a notable transaction. That window, between the transaction filing and the public recognition of what it means, is where a sharp bettor operates.
The pattern to watch today, starting around 4 PM ET:
- Monitor the CBS Sports or NFL.com live roster cuts tracker for skill-position releases involving Week 1 matchups.
- For each notable cut, identify which team benefits offensively and which is weakened.
- Check the current total at BetMGM, DraftKings, and FanDuel within five minutes of the transaction.
- If the total has not moved within 15 minutes of a significant skill-player cut, the market has not fully priced the news. That is when to act.
Applying the Holdover Bias Framework to 2026
The Seahawks are the defending Super Bowl champions. They opened as -3.5 favorites. The number is now -4 at most books, with some at -4.5. That line drift represents public and market anchoring to Seattle's championship form, exactly the holdover bias that Fodor and DiFilippo documented.
Their Super Bowl MVP running back is gone. Their primary backup is on IR with a torn ACL. The total is 44.5, a number that assumes the Seahawks offense performs at something close to its 2025 playoff level. The market has not adjusted the total despite a significant personnel change at the most ball-touch position on the offense.
The Patriots, on the other side, missed the playoffs in 2025. Holdover bias says they are underpriced as the away dog. The research from 2004-2012 found that non-playoff teams beat prior playoff teams by a margin significant enough to produce 22%+ returns when bet systematically. The Patriots fit the underdog-non-playoff profile exactly.
This does not mean bet the Patriots or bet the under. It means the structural pricing error identified in the academic literature is present in this specific matchup. The holdover bias is not a lock, it is a thumb on the scale, and right now the scale has not been recalibrated for the season-opening information it is about to receive at 6 PM today.
Framework Summary
| Signal | What It Means for Totals | Timing |
|---|---|---|
| Prior Super Bowl/playoff team as Week 1 favorite | Total may be inflated by holdover bias. Lean under, especially if total is 44+. | Now (before cuts) |
| Offensive skill-player cut from a Week 1 team | Total should move down. Bet under before or shortly after the total adjusts. | 4-6 PM ET today |
| Starting secondary cut from a Week 1 defense | Total should move up. Bet over if the market is slow to respond. | 4-6 PM ET today |
| Total 50+ with two offenses both coming off strong prior seasons | Double holdover bias exposure. 54.9% historical under rate applies with additional tailwind. | Watch Bengals-Bucs cutdowns today |
| Rebuilt-roster game with lowest total on board | Maximum cutdown sensitivity. Watch Jets-Titans for any move off 39.5. | Post-6 PM ET today |
What the Under Rate Tells You
A 54.9% under rate since 2015 is significant, and it is not fully explained by the holdover bias alone. Part of it is the lag between August depth charts and actual Week 1 rosters. Part of it is that offensive schemes take weeks of live game reps to function at their projected rate. And part of it is that the market consistently prices Week 1 totals higher than the actual results support, because the public wants to see the stars and the big numbers from the prior season.
Today changes the information set. After 6 PM, the rosters are final. Books will adjust several totals based on the transaction wire. By tomorrow morning, the market will have incorporated most of the cutdown information.
The 10-day window between today's deadline and Week 1 kickoff on September 9 is when the market finishes its repricing. The sharper play, for most of the totals on the board, is to understand the structural bias first and then watch for the cutdown-driven moves that confirm or contradict what the holdover research predicts.
The 54.9% under rate in Week 1 is not a coincidence. It is a predictable output of a market that anchors to last year's rosters, faces a concentrated information event today at 6 PM, and then has to recalibrate in 10 days. The games that fit the holdover bias profile the tightest are the ones worth tracking most closely this afternoon.