PPDA punishes the team that is winning

The most quoted pressing metric is contaminated by scoreline, and comparing two clubs on a season figure compares their league positions.

Passes allowed per defensive action is a neat idea and a badly behaved number. It divides opposition passes by your own tackles, interceptions, fouls and challenges in a defined area, and the lower it goes the more aggressively you are supposed to be pressing.

The problem is that the denominator depends on the game and the numerator depends on the scoreline.

Go a goal up against a weaker side in the twentieth minute. Your opponent now has to attack, so they take the ball more often, and each of their possessions lasts longer, which inflates the pass count. Meanwhile you have every reason to sit slightly deeper, so your defensive actions happen further from their goal and fall outside the measured zone. Your PPDA gets worse over the course of a match you are comfortably controlling. Now go behind against a stronger side and chase the game with a desperate high press for twenty-five minutes. Your PPDA improves.

Read the season table and the strongest teams often look like moderate pressers while mid-table sides that spend a lot of time losing look ferocious. Some of that is real. Most of it is game state.

The fix is not complicated and it is not new. Compute the metric only during periods of parity, at level scoreline, and preferably weight it by field position rather than a binary zone. Do that and the ordering changes noticeably, because you have removed the sides whose pressing reputation is really a record of how often they have needed to chase.

There is a second contamination worth naming. PPDA counts fouls as defensive actions, which means a team that presses badly and hacks people down looks identical to a team that presses well and wins the ball cleanly. Those are not the same tactic and they do not have the same consequences by minute seventy or by the third yellow card.

I still use the number. It is quick, it is available, and for a single match against a single opponent it does tell you something about intent. What it cannot do is what it is most often used for, which is ranking clubs across a season as though it described a philosophy rather than a circumstance.

If your pressing metric improves when you concede, it is not measuring your pressing.