PPDA is a bad number and we should stop quoting it in arguments

It rewards teams that never have the ball, punishes teams that keep it, and tells you almost nothing about whether a press worked.

Passes per defensive action. Take the passes the opposition completed, divide by your tackles, interceptions, fouls and challenges in the attacking sixty percent of the pitch. Low number means you press aggressively. That is the pitch, and it caught on because it is easy to calculate from event data everyone already had.

It is also close to useless as a standalone measure, and it gets quoted constantly as though it settles things.

Start with the obvious. A team with thirty percent possession faces far more opposition passes, and their defenders make far more actions, simply because the ball is near them all afternoon. A team with seventy percent possession barely gets the chance to press. So PPDA quietly tracks possession share, and comparing a possession-dominant side's number to a counter-attacking side's number is comparing two different sports.

Then there is the definitional problem. A defensive action counts whether it succeeded or not. A midfielder who lunges in, misses, and gets bypassed has generated a defensive action. He has also just broken the line and put his team in trouble. The stat cannot tell that apart from a clean interception in the same spot.

Worse, the very best pressing does not generate defensive actions at all. If your shape is so good that the opposition's centre-back looks up, sees nothing on, and hits it long into the channel for your keeper, that is a total pressing success. No tackle, no interception, no foul. PPDA records it as a long pass completed to nobody and gives you no credit whatsoever. The best press produces the fewest actions, which is exactly backwards.

What I would want instead is not complicated. Where does the opposition's possession end, on average, measured in metres up the pitch. How long do their possessions last. What share of their build-ups end with a long ball inside eight seconds. Those get at the actual question, which is whether you are stopping them playing, rather than counting how often somebody stuck a leg out.

None of this is a secret. Analysts at clubs moved past PPDA years ago and use their own composite measures built on tracking data. It survives in public discussion because it is free, it is a single number, and a single number is easy to put in a graphic.

That is the whole reason. Not because it is good.