Omaha 1.10.0
Released 15 September 2026
When a team's starting tight end is out, real teams line up differently. The simulation kept calling the same formation and sent the backup out instead.
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 1.9.0’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 1.9.0 |
|---|---|---|
| Receiving touchdownsrare event | +20.1% | −0.03−0.12 to +0.06 |
| Rushing touchdownsrare event | +17.9% | −0.17−0.26 to +0.06 |
| Receiving yards | +13.7% | +0.01−0.02 to +0.03 |
| Pass attempts | +13.0% | +0.03−0.01 to +0.08 |
| Rushing yards | +12.5% | −0.01−0.05 to +0.02 |
| Passing yards | +11.0% | +0.01−0.03 to +0.06 |
| Completions | +9.8% | +0.01−0.06 to +0.07 |
| Passing touchdownsrare event | +6.9% | +0.01−0.21 to +0.08 |
| Rush attempts | +3.8% | −0.02−0.06 to +0.04 |
| Targets | +1.5% | +0.04−0.00 to +0.08 |
| Receptions | +1.3% | +0.03−0.03 to +0.06 |
| Average across all figures | +10.1% |
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
A formation is a choice, and losing your main tight end changes it. Real teams drop from two tight ends to one and put a third receiver on the field. The simulation chose formations from what a team normally does on that down and distance, which carries no information about who is healthy, so it kept asking for two tight ends and handed the second slot to a backup who does not really play that role.
With their main tight end out, real teams use two or more tight ends on 19.7% of snaps instead of 28.0%. The simulation moved by a tenth of a percentage point. It now lands on 19.4%.
This is the change you are most likely to notice, because it is the one you can trigger. Before, ruling out a tight end moved his work almost entirely to other tight ends, and receivers barely registered. That was the formation model holding the shape of the offense fixed. Now the whole offense reshapes, the way a real one does.
About 85% of a ruled-out tight end's lost yards previously went to other tight ends, with receivers moving at the noise floor.
This did not make our projections measurably better, and we expected that before we measured it. Roughly 2.5% of snaps across a season involve a formation this changes, which is too few to move a season-long accuracy figure. We are shipping it because it makes the simulation behave correctly when you ask it a question about a missing player, not because it improved a score.
Across the 2024 season every one of the twelve statistics we track moved less than its margin of error. The average change was two hundredths of a percentage point.
What this version still gets wrong
- This decides how many tight ends or receivers are on the field, not which ones. Who fills the slot is still chosen the same way it was.
- It is fitted as one league-wide pattern. A team that leans on two tight ends and a team that rarely uses them get the same response to losing one.
- Two starters out at once is handled by combining the two adjustments, which assumes they act independently. We have not measured whether they do.
- The simulation still does not go quite as far as real teams do toward a single tight end: it lands about six points short on the most common formation.
- Running plays were what we measured. Whether the right receivers are targeted when a starter is missing is a separate question we have not looked at.