The answer depends on which lens you use, but by points allowed per game the 2025 Dallas Cowboys have fielded the NFL's worst defense. They entered Week 14 allowing roughly 30.5 points per game, worst in the league, ahead of only a handful of units that had already fired coaches or benched starters. Total yards tells a similar story, though pass defense and run defense disagree on the details. This article explains why defenses fail, how to measure them properly, and where the public conversation gets the logic wrong.
How a Defense Actually Generates Outcomes
A defense does not "allow points" in a vacuum. Points emerge from a short chain of events: field position, down and distance, play call, execution, and finishing. Each link can fail, and bad units usually fail at two or three links simultaneously.
Start with field position. A defense that starts possessions at its own 25-yard line faces a different problem than one starting at the opponent's 40. Turnovers, special teams, and offensive three-and-outs all set that starting point. A unit can look worse than it is because its offense punts poorly or throws interceptions in bad spots.
Down and distance is the second link. Third-and-short converts at high rates against almost any defense. Third-and-long is the opposite. A defense that allows early-down gains of 5 to 6 yards repeatedly forces itself into short-yardage situations, where the opponent can run or pass efficiently. That compounding is why "third-down defense" is often a symptom, not a cause.
The third link is explosive plays. A defense can allow eight plays of 4 yards and one of 50. The long gain costs more points than the eight medium ones combined in expected-points terms. Bad defenses typically leak explosive runs and passes because of missed tackles, poor pursuit angles, or coverage busts.
The fourth link is finishing in the red zone. Once an offense reaches the 20, the field compresses. Defenses that cannot stop the run or defend the back-shoulder fade give up touchdowns instead of field goals. A 3-point difference per trip, multiplied across a season, separates the worst unit from the middle of the pack.
Plain language: points are the final symptom, but the disease usually starts on first down and in the tackling.
The Numbers Behind the 2025 Season
Through Week 13 of the 2025 season, the Cowboys' defensive profile was historically poor. They were allowing about 30.5 points per game and roughly 6.0 yards per play, both near the bottom of the league. Their run defense was among the five worst by yards per carry allowed, and their pass defense gave up explosive plays at a rate that compounded the damage.
Context matters here. The league average in 2025 sits near 22 points per game. The gap between the Cowboys and the league average was roughly 8.5 points per game, which is enormous. For comparison, the difference between the league average and the best defense in a typical season is usually 6 to 8 points per game.
The Washington Commanders were the closest competitor for "worst" honors for much of the season. They allowed more total yards per game but had a slightly better points-per-game figure because they generated more takeaways. That divergence is a useful teaching point: yards and points are correlated but not identical, and takeaways can mask a bad unit.
| Metric | Cowboys (approx.) | Commanders (approx.) | League average |
|---|---|---|---|
| Points allowed per game | 30.5 | 28.0 | 22.0 |
| Yards allowed per game | 380 | 395 | 340 |
| Yards per play allowed | 6.0 | 5.9 | 5.4 |
| Takeaways per game | 0.9 | 1.4 | 1.2 |
| Red-zone TD rate allowed | 68% | 64% | 57% |
| Explosive plays allowed per game | 7.5 | 7.0 | 5.5 |
The pattern is consistent: the Cowboys give up more points, more yards per play, and more red-zone touchdowns, while generating fewer takeaways. The Commanders are worse by raw yards but slightly better by points because they take the ball away more often.
One counterintuitive detail: a defense that allows many yards but forces turnovers can rank better in points than a defense that allows fewer yards but never takes the ball away. Turnovers are high-variance and not fully controllable. That is why professional evaluators weight points and expected points added (EPA) per play more heavily than raw yards.
Edge Cases That Distort the Ranking
Defensive rankings are noisy. Several factors can push a unit up or down the list without reflecting true quality.
Strength of schedule. Facing top-five offenses inflates points allowed. A defense that plays the Chiefs, Bills, and Ravens in a five-week stretch will look worse than one playing bottom-tier offenses. Offensive pace and turnovers. A fast, turnover-prone offense gives the defense more possessions and worse field position. Injuries to key players. A single elite pass rusher or cornerback can swing a unit by 2 to 3 points per game. Garbage time. Blowout losses produce meaningless late points that can inflate the season total. Weather and home-field. Cold, windy games suppress scoring for both teams.
The professional caveat: do not judge a defense on a single metric or a single game. The worst defense by points allowed in a given week may be a middle-tier unit that faced a top offense in a hurricane. Rankings are useful over 12 to 17 games, not over two.
Misconceptions About "Worst Defense"
The most common misconception is that the worst defense is the one allowing the most yards. Yards matter, but points decide games. A unit that bends between the 20s and stiffens in the red zone can rank poorly in yards and acceptably in points.
The second misconception is that defensive failure is purely a talent problem. Scheme, communication, and coaching matter. A defense with average talent but poor communication allows explosive plays because players are out of position. A defense with elite talent but a predictable scheme can be attacked repeatedly.
The third misconception is that takeaways are a stable skill. They are partly skill and largely variance. A defense that relies on turnovers to survive will regress toward the mean. The Cowboys' low takeaway rate in 2025 is one reason their points allowed stayed high.
The fourth misconception is that a bad defense is uniformly bad. Most units are strong in one area and weak in another. The Cowboys, for example, were especially vulnerable to explosive runs and play-action passes, while their third-down defense was closer to league average. Treating them as uniformly terrible misses where the real damage occurs.
What Coaches and Analysts Actually Do
Professionals do not rank defenses by a single number. They use a layered approach.
First, they look at EPA per play allowed, which weights each play by its actual impact on scoring. Second, they separate pass and run EPA, because the modern game is pass-heavy and pass defense matters more. Third, they adjust for opponent strength using metrics like DVOA (Defense-adjusted Value Over Average) or similar schedule-adjusted models. Fourth, they examine situational splits: red zone, third down, two-minute drill, and goal line.
A practical process for a fan or analyst:
Start with points allowed per game and EPA per play allowed. Adjust for strength of schedule using a public schedule-adjusted metric. Split pass versus run to find the leak. Check red-zone and third-down efficiency to see whether the leak is situational. Review explosive plays allowed per game to identify structural problems. Cross-check takeaways per game to see whether the ranking is turnover-dependent.
A useful warning: if a defense ranks well in yards but poorly in points, suspect red-zone failure or a high takeaway rate that is masking inefficiency. If it ranks well in points but poorly in yards, suspect good red-zone defense and turnover luck that may not persist.
Limitations of Any "Worst Defense" Label
Any single-season ranking is a snapshot, not a verdict. Defenses change week to week with injuries, scheme adjustments, and opponent quality. The 2025 Cowboys' ranking reflects their performance through Week 13, not a permanent identity.
Sample size is another limit. Twelve games is enough to identify a trend but not enough to separate a truly bad unit from a mediocre one with bad luck. Statisticians generally want 16 to 17 games, plus playoff data, before making strong claims. Even then, one season is a small sample in a sport with high play-to-play variance.
Measurement limits matter too. Public metrics like points allowed and yards allowed are not opponent-adjusted. Advanced metrics like DVOA and EPA are better but still model-dependent. Different models can disagree by several ranking spots.
The honest answer: the Cowboys have the worst defense in the NFL by the most common public metrics through Week 13 of 2025, but the gap between them and the next-worst unit is smaller than the headline suggests, and the ranking could shift with a few strong or weak performances.
This article is general educational information about sports analytics and is not a prediction, betting recommendation, or professional evaluation of any team or player. For betting or fantasy decisions, consult a qualified professional and follow applicable regulations in your jurisdiction.