Gridiron · Game Center · 2024
The Rush Room
Every dropback in this database carries a count of how many defenders came after the quarterback. Sending a fifth is a trade — one more man in the rush is one fewer in coverage — and the trade is not the same trade for every defence. This room is that trade, priced, 2018 to 2025.
Across 19,946 dropbacks, a five-or-more rush hit the quarterback on 18.1% of snaps where a four-man rush hit him on 13.4%. 19,946 charted dropbacks across the 2024 regular season, each one carrying a count of how many defenders rushed. 1,032 of them are scrambles — the box score files those as runs, and a quarterback who took a snap and ran from a rush is on a dropback, so this room counts them. 2,701 penalty-wiped snaps are set aside: they carry a rusher count averaging two, and no hit or sack is ever recorded on one, so they would deflate every rate here while looking like extra sample. 269 more are dropped because the charted rush was zero, which is a missing value rather than a play nobody rushed, and 4 are left out of the box average for the same reason.
Defense. The rush this unit SENT. Expected points stay in the offence's frame, so on a defence row a LOWER number is the better one — the ladder is sorted and worded in this defence's own frame so you never have to invert it by hand. The situation pills are in this team's frame too: Trailing on a defence row means this defence's own team was behind. Every dropback. Every charted dropback of the regular season, sacks and scrambles included. 32 rows clear the 80-dropback floor in this split.
Every dropback in this database carries a count of how many defenders came after the quarterback. Sending a fifth is a trade — one more man in the rush is one fewer in coverage — and the trade is not the same trade for every defence.
19,946 charted dropbacks across the 2024 regular season, each one carrying a count of how many defenders rushed. 1,032 of them are scrambles — the box score files those as runs, and a quarterback who took a snap and ran from a rush is on a dropback, so this room counts them. 2,701 penalty-wiped snaps are set aside: they carry a rusher count averaging two, and no hit or sack is ever recorded on one, so they would deflate every rate here while looking like extra sample. 269 more are dropped because the charted rush was zero, which is a missing value rather than a play nobody rushed, and 4 are left out of the box average for the same reason.
Two denominators exist for this season and both are honest: 19,946 charted dropbacks after the exclusions above, and 22,916 before them. Every rate on this page divides by the charted figure — a rate over the raw count would credit a defence for snaps on which nobody counted the rush.
Painted Monday 04:00 ET · the room refreshes with the weekly play-by-play sync
Before you judge anybody for blitzing
What an extra rusher is worth, league-wide
Across 19,946 dropbacks in this situation, a five-or-more rush hit the quarterback on 18.1% of snaps against 13.4% for a four-man rush — and sacked him on 7.7% against 6.1%. The extra man reliably buys CONTACT. Whether it buys anything on the scoreboard is the column on the right, and it is the question the rest of this room is about.
| Rushers | Share | Dropbacks | Sack | Hit or sacked | Box | EPA / drop |
|---|---|---|---|---|---|---|
| 3 or fewerera | 3.8% | 756 | 5.0% | 9.7% | 5.2 | +0.11 |
| 4 | 69.4% | 13,844 | 6.1% | 13.4% | 5.8 | +0.05 |
| 5 | 20.3% | 4,043 | 7.4% | 17.1% | 6.1 | +0.04 |
| 6+ | 6.5% | 1,303 | 8.6% | 21.2% | 6.3 | +0.04 |
Defense · Every dropback · 32 rows
What the extra man bought
Expected points on a five-or-more rush minus expected points on a four-man rush, same key, same situation. Expected points are the offence’s, so on a defence row a NEGATIVE delta is the blitz working. The ladder is sorted best-for-this-team first either way, and every row carries the plain word. Withheld as an em-dash below 80 dropbacks in the situation or 25 inside either rung.
| # | Team | Drops | 5+ share | On 4 | On 5+ | The extra man | |
|---|---|---|---|---|---|---|---|
| 1 | 599 | 27.5% | +0.14 | −0.11 | −0.25paid | ||
| 2 | 569 | 24.3% | +0.05 | −0.18 | −0.24paid | ||
| 3 | 624 | 24.2% | +0.08 | −0.15 | −0.24paid | ||
| 4 | 622 | 23.5% | +0.02 | −0.12 | −0.14paid | ||
| 5 | 613 | 24.5% | +0.13 | −0.01 | −0.14paid | ||
| 6 | 685 | 36.5% | −0.02 | −0.15 | −0.12paid | ||
| 7 | 638 | 20.5% | +0.09 | −0.03 | −0.12paid | ||
| 8 | 636 | 26.9% | +0.03 | −0.08 | −0.10paid | ||
| 9 | 646 | 24.5% | +0.08 | −0.02 | −0.10paid | ||
| 10 | 593 | 29.3% | +0.20 | +0.12 | −0.09paid | ||
| 11 | 653 | 34.8% | +0.10 | +0.02 | −0.09paid | ||
| 12 | 610 | 19.7% | +0.10 | +0.04 | −0.06paid | ||
| 13 | 544 | 24.3% | +0.08 | +0.05 | −0.03paid | ||
| 14 | 713 | 39.0% | −0.04 | −0.07 | −0.03paid | ||
| 15 | 597 | 30.2% | +0.05 | +0.02 | −0.03paid | ||
| 16 | 621 | 28.3% | +0.09 | +0.08 | −0.01level | ||
| 17 | 684 | 36.8% | +0.03 | +0.03 | 0.00level | ||
| 18 | 624 | 19.1% | −0.05 | −0.05 | 0.00level | ||
| 19 | 604 | 24.2% | +0.04 | +0.04 | 0.00level | ||
| 20 | 677 | 22.2% | +0.03 | +0.04 | +0.01level | ||
| 21 | 658 | 16.3% | +0.23 | +0.27 | +0.04cost | ||
| 22 | 577 | 32.2% | +0.15 | +0.20 | +0.05cost | ||
| 23 | 575 | 34.8% | 0.00 | +0.07 | +0.07cost | ||
| 24 | 640 | 28.1% | +0.01 | +0.08 | +0.07cost | ||
| 25 | 579 | 31.1% | +0.05 | +0.16 | +0.11cost | ||
| 26 | 575 | 24.5% | +0.05 | +0.17 | +0.11cost | ||
| 27 | 663 | 19.2% | −0.03 | +0.09 | +0.12cost | ||
| 28 | 703 | 36.1% | −0.13 | +0.01 | +0.14cost | ||
| 29 | 592 | 31.1% | +0.08 | +0.26 | +0.18cost | ||
| 30 | 645 | 18.6% | −0.09 | +0.11 | +0.19cost | ||
| 31 | 618 | 23.9% | +0.08 | +0.32 | +0.24cost | ||
| 32 | 569 | 18.5% | +0.01 | +0.26 | +0.25cost |
1st of 32 · Every dropback
NYJ rung by rung
The same four rungs as the league card, for this team, with the league’s expected points drawn behind as a hairline. A rung under 25 dropbacks prints an em-dash rather than a mean of nine snaps.
- Three or fewer rushed24 drops · 4.0% of them
EPA / drop
—
lg +0.11
Sack
—
lg 5.0%
Hit or sacked
—
lg 9.7%
Box
—
lg 5.2
Under 25 dropbacks on this rung, so no mean is printed.
- Four rushed410 drops · 68.4% of them
EPA / drop
+0.14
lg +0.05
Sack
6.1%
lg 6.1%
Hit or sacked
13.2%
lg 13.4%
Box
5.9
lg 5.8
+0.09 against the league on this rung — worse than average for a defence.
- Five rushed110 drops · 18.4% of them
EPA / drop
−0.16
lg +0.04
Sack
10.0%
lg 7.4%
Hit or sacked
13.6%
lg 17.1%
Box
6.1
lg 6.1
−0.20 against the league on this rung — better than average for a defence.
- Six or more rushed55 drops · 9.2% of them
EPA / drop
−0.02
lg +0.04
Sack
7.3%
lg 8.6%
Hit or sacked
20.0%
lg 21.2%
Box
6.4
lg 6.3
−0.05 against the league on this rung — better than average for a defence.
EPA / drop
−0.11
lg +0.04
Sack
9.1%
Hit or sacked
15.8%
This is the pool the ladder above subtracts from, and it is why the ladder can print a delta on a slice where the five and six-plus rungs are each too thin to print on their own. Pooled by counts, not by averaging two means — the six-plus rung is several times the smaller of the two and a mean of means would weight it as though it were not.
Where the extra man paid
NYJ by situation
The same delta across all 8 situations. They overlap on purpose — a trailing two-minute dropback is in both — because forcing a snap to pick one would hide the ones most worth reading.
- −0.25
- −0.34
- —
- +0.30
- −0.69
- —
- −0.47
- −0.19
Clicking a situation here drives the whole room. Emerald is the extra man paying for this team, red is it costing them — the sign is already flipped into the defence's frame.
What this room does not have
There is no pressure or hurry column in this database — the two that exist are empty on every row of every season — so nothing here is a pressure rate. What is counted is the rush itself, and whether the quarterback was hit or brought down. A sack is not always recorded as a hit, so the two are reported as one quantity rather than as a conversion. Quarterbacks appear only above 150 dropbacks in the season.