Week 10 fantasy defense rankings — every defense, ranked by matchup

A standard-scoring defense earns points two ways: a points-allowed bracket (a shutout is 10, 14–20 allowed is 1, 35+ is −4) and counting stats (1 per sack, 2 per interception or fumble recovery, 6 per touchdown). The projection estimates each part, using the same scoring rules as the DST fantasy points calculator:

projected = PA points + expected sacks × 1 + expected takeaways × 2 + big-play base

PA points come from the opponent's implied total. The market's number for the opposing offense — the same figure on the implied team totals board — is the single best public forecast of what a defense will allow. But a team implied for 20 doesn't score exactly 20: across every 2023–25 regular-season game, team scores missed their implied total by a standard deviation of 9 points. So instead of reading one bracket, the projection weighs every bracket by how likely the score is to land in it. An opponent implied for 17 is worth 1.96 PA points, a 20.5 is worth 0.98, and a 27 is worth −0.61. Checked against those same games, that curve was within about 0.2 points of what the bracket actually paid in every band from 15 to 30.

Sacks and takeaways meet in the middle. A pass rush facing a line that protects well, or a ball-hawking secondary facing a careful quarterback, should expect less than either side's numbers alone suggest. So expected sacks are the average of the defense's 2026 sacks per game and the opponent offense's sacks allowed per game; expected takeaways average the defense's interceptions + fumble recoveries with the opponent's giveaways.

Small samples are shrunk toward the league. Two or three games of sacks is mostly noise, so before averaging, each team's rate is pulled toward the league average: its own number gets a weight of games ÷ (games + 3). Right now, with 9 games played, a team's own rate counts 75% and the league average 25%. After six games it is 67%, after twelve 80%.

The big-play base is the same for everyone. Defensive and return touchdowns, safeties and blocked kicks are too rare to rate a team on, so every defense gets the league's 2026 average per game. It moves no one up or down; it keeps the projection on the same scale as a real DST score.

The tiers. Stream at a projection of 7.0 or more, avoid under 5.0, and everything between is viable — worth a start when your options are thin. A league-average defense scores about 5.8 standard points a game, so the streaming line sits about a point above average.

DST scoring is mostly noise. One pick-six is six points, usually more than the gap between the top and the middle of this board. A projection tells you where the odds lean, not what happens — expect the top of the list to disappoint some weeks and a mid-table defense to win yours.

Lines and injuries move. A quarterback ruled out on Friday moves the opponent's implied total, and this page moves with it only at the next data refresh. The lines stamp above the table says when they were pulled. For a line that has moved since, run it through the implied team total calculator and read the PA column off the curve above.

Standard scoring only. Yahoo's yards-allowed bracket and big-play formats are not projected here — the order rarely changes, but the totals do. Score your own league's format with the DST calculator.

Weather is flagged, not modeled. Sustained wind of 15 mph or more does help defenses a little; the NFL weather report has the forecast behind the weather column. Treat a wind flag as a tiebreaker between two close rows.

A fantasy tool, not betting advice. FantasyTools analysis is opinion for entertainment purposes only — not betting, financial, or professional advice. The lines are a median across sportsbooks; no book is named or quoted.