NFL EDGES

Live game edges — model vs. Kalshi, ranked

Gridiron Edge · 5-min refresh

NFL 2026

Power rankings, futures & win totals — all 32 teams

COMBO EDGE

Correlation-aware Kalshi combos — free

NFL Injury Report, Week 3: Which Defenses Are Breaking

The official injury report, read by unit instead of by name. One missing starter rarely moves a game; several out of the same front seven or secondary does, because the replacements were not in the rotation. This board weights every absence by the player’s share of snaps and flags the defenses that are close to breaking.

Week 3 defensive injury board

Every defense on the Week 3 slate, ranked by how much of its front seven (DL + LB) and secondary (CB + S) is out or questionable. Snap share is summed across the unit — 1.00 is one every-down player. 2 defenses are flagged.

DefenseGameLineFront out (+Q)Secondary out (+Q)Status
SFvs ARI−8.51.91 — NEAR
Out: Nick Bosa DL 64%, Romello Height DL 52%, CJ West DL 41%, James Thompson DL 35%
CLEvs CAR+2.5— — (+1.88)NEAR
Questionable: Grant Delpit S 100%, Tyson Campbell CB 88%
ARI@ SF+8.5— (+0.59)0.90 (+0.37)—
Out: Will Johnson CB 90%
Questionable: Roy Lopez DL 59%, Max Melton CB 37%
NYGvs TEN−2.5— (+0.76)— (+1.00)—
Questionable: Tyler Nubin S 100%, Brian Burns LB 76%
TBvs MIN+1.51.49 — —
Out: Josiah Trotter LB 87%, Rueben Bain LB 62%
NE@ JAX+30.55 — (+0.83)—
Out: Dre'Mont Jones LB 55%
Questionable: Craig Woodson S 83%
CAR@ CLE−2.50.85 0.52 —
Out: Devin Lloyd LB 85%, Nick Scott S 52%
LV@ NO+3— (+0.56)0.73 —
Out: Treydan Stukes S 73%
Questionable: Kwity Paye DL 56%
DETvs NYJ−6.5— 1.17 —
Out: Avonte Maddox CB 65%, Thomas Harper S 52%
WASvs SEA+7.5— 0.99 —
Out: Nick Cross S 99%
SEA@ WAS−7.5— 0.98 —
Out: Julian Love S 98%
INDvs HOU+1.50.26 — (+0.70)—
Out: Micheal Clemons DL 26%
Questionable: Charvarius Ward CB 70%
PITvs CIN+3.5— — (+0.92)—
Questionable: Jamel Dean CB 92%
CHIvs PHI+4.5— — (+0.83)—
Questionable: Tyrique Stevenson CB 83%
DALvs BAL+3.5— 0.62 —
Out: Cobie Durant CB 62%
LAC@ BUF+70.47 — —
Out: Dalvin Tomlinson DL 46%
NYJ@ DET+6.50.46 — —
Out: David Onyemata DL 46%
BAL@ DAL−3.5— — —
BUFvs LAC−7— — —
CIN@ PIT−3.5— — —
DENvs LAR+2.5— — —
HOU@ IND−1.5— — —
JAXvs NE−3— — —
KC@ MIA−10.5— — —
LAR@ DEN−2.5— — —
MIAvs KC+10.5— — —
MIN@ TB−1.5— — —
NOvs LV−3— — —
PHI@ CHI−4.5— — —
TEN@ NYG+2.5— — —

Percentages are each player’s share of defensive snaps over his team’s last four games. Out = listed Out or Doubtful, on injured reserve, or no longer on the roster. Updated nightly Thursday to Sunday from the official report.

This week’s trigger calls

The one spot the history backs: a clustered defense that the market still favours. From 2018–2025, scoring each season only with rules set on the seasons before it, the opponent covered 45-27 (62.5%).

2026 ledger, graded in public

Every defense that actually played a game in a cluster this season, measured from snap counts after the game, and whether its opponent covered the closing line. No trigger spot has been graded yet.

None yet — a defense needs two games of snap history before a cluster can be measured.

Injury clusters, explained

What is an NFL injury cluster?

A cluster is when one unit of a defense loses several regulars at once: three or more starters out of the front seven (defensive line and linebackers) or the secondary (cornerbacks and safeties), or two full players’ worth of that unit’s snaps. A team can usually absorb one missing starter because the replacement is still an NFL player. Several at once means players who were not in the rotation are playing together, and offenses keep attacking that patched unit.

Does the market price defensive injuries correctly?

Mostly, yes. The exception in our 2013–2025 study is a clustered defense that the market still favours. Scoring each season only with rules set on the seasons before it, the opponent covered 45-27 (62.5%) from 2018–2025, about 1.6 points better than an ordinary game. When the clustered team is the underdog, the line already reflects it. It is a small sample, which is why every 2026 case is graded on this page.

How is “out” decided before the game?

From the official injury report: a player listed Out or Doubtful, on injured reserve, or no longer on the roster counts as out. Questionable players are shown separately, and a defense that becomes a cluster only if its questionable players sit is marked WATCH. Each player’s weight is his share of defensive snaps over his team’s last four games, so a rotational backup counts for less than an every-down starter.

Which matters more, the front seven or the secondary?

In the history, the front seven. Front-seven clusters ran about 1.5 points worse than expected for the defense; secondary clusters barely moved the result, because offenses throw more but the extra passes are not more efficient. The one secondary case that did break was three starting cornerbacks out at once.

More NFL Markets