Omaha 2.7.0
Released 14 October 2026
When a team's lead running back is out, the backs who replace him are now projected to carry the ball more, as they do in real games. Rushing forecasts for backup running backs improve in both seasons we checked. Across every player the eleven figures barely move: the average goes from +10.76% to +10.83%.
How to read these numbers
Every figure below compares the simulation against a player’s own season average so far, scaled to how much of the game is left, roughly what a reader could work out without us. Zero means we add nothing over that. Positive means we beat it; negative means we are worse than it.
Measured across the 2024 season, weeks 1–18, rebuilding the model each week using only what was known at the time. 400 simulations per situation.
The change column is simply this version’s accuracy minus 2.6.0’s, each measured across everything that version covers, so the two columns you can see always account for it. We also measure the change on shared situations alone, which is a more sensitive test; where the two disagree we publish this one.
Accuracy by stat
| Stat | vs. season average | Change from 2.6.0 |
|---|---|---|
| Receiving touchdownsrare event | +20.4% | +0.00+0.00 to +0.00 |
| Rushing touchdownsrare event | +18.2% | +0.00+0.00 to +0.00 |
| Receiving yards | +14.7% | +0.00+0.00 to +0.00 |
| Pass attempts | +12.9% | +0.00+0.00 to +0.00 |
| Rushing yards | +12.9% | +0.19−0.03 to +0.43 |
| Passing yards | +11.0% | +0.00+0.00 to +0.00 |
| Completions | +9.6% | +0.00+0.00 to +0.00 |
| Passing touchdownsrare event | +6.9% | +0.00+0.00 to +0.00 |
| Rush attempts | +4.7% | +0.54+0.20 to +0.92 |
| Targets | +3.9% | +0.00+0.00 to +0.00 |
| Receptions | +3.8% | +0.00+0.00 to +0.00 |
| Average across all figures | +10.8% |
The average is an unweighted mean across the figures above. It is a summary of how we are doing, not a single score for the model: yards and counts are not measured in the same units, so combining them any more cleverly than this would just let passing yards decide the answer.
Each change carries the range the measurement can actually support. A change whose range crosses zero is shown in neutral rather than as a gain: we cannot tell it apart from no change, and we are not claiming it as one.
What changed
The simulation already handed his carries to the backs who dress. But before kickoff every projection is pulled partway toward the player's own season average, and a backup's average was recorded while he was the backup. That pull took most of the extra work back out again, so the player stepping in was projected for carries closer to his usual handful. The average the projection is pulled toward now includes a share of the missing back's.
For backup running backs (each team's second and third back by carries and targets so far that season), the rush attempts forecast improves by 1.3 points across 2024 and 2.0 across 2025, and rushing yards by 0.4 and 0.9. Both rush attempt ranges and the 2025 rushing yards range are clear of zero. Starting running backs are unchanged.
The same adjustment applies when you mark a team's lead back as out in a custom simulation, so the backs who remain are projected the way they would be if he had been listed inactive.
Only running back rushing figures change. Every other figure for every player is identical to 2.6.0, and so is every rushing figure on a team whose lead back plays.
What this version still gets wrong
- Backup running backs' rush attempts still score below a season average in 2025: -2.2%, up from -4.2%. Better, not solved.
- It covers running backs only. The same adjustment for receivers and tight ends made their target and reception forecasts worse, because a missing receiver's targets spread across positions in ways a season average cannot capture. That needs its own work.
- A player given a chance to play between 0% and 100% does not trigger it. Only a player who is out does, because in most simulations a partly available back still plays.
- Backups' projections for big games are still too modest. A backup tight end or receiver reaching four receptions is projected about a fifth less often than it happens.