The best goals-per-90 records in Premier League history

Goals per 90 is often described as measuring finishing, and that is not quite right. It measures the whole chain: how good the team was at creating chances, how central the player was to that creation, what share of the chances fell to him, and then how well he took them. A very good finisher in a poor side will rate below a good finisher in an excellent one, and no amount of statistical adjustment inside this table will change that.

Which is why the clubs column repays attention. The highest rates in Premier League history have overwhelmingly been produced at teams that dominated possession and generated an unusual volume of chances — and the striker in that system is receiving a quality and quantity of service that a forward at a mid-table club will never see.

None of that makes the achievement smaller. It does mean a rate should be read as a description of a player in a situation, rather than a portable measure of ability that would have travelled anywhere.

Any list of scoring rates invites the same objection, and it deserves a straight answer: these totals include penalties, because the dataset counts goals rather than goal types. A designated penalty taker at a side that attacks a lot will accumulate a meaningful number of them over a career, and there is no way to separate that out here.

Whether that matters is a genuine argument rather than a settled one. Penalties are part of a striker's job and missing them is a real failure; equally, being the nominated taker is partly a political fact about a dressing room. Reasonable people land in different places.

What is not reasonable is quoting a rate as though the question had been resolved. If you are using one of these numbers to win an argument, say what it includes — the figure is strong enough to survive the caveat.

Rate statistics are supposed to neutralise era, and they mostly do. But scoring environments have changed: the volume of chances a leading side generates now is higher than it was in the mid-nineties, and the defensive structures that produced the lowest-scoring Premier League seasons are gone. A rate set in a low-scoring era against organised, deep defences is not measuring quite the same task as one set against a modern high line.

The practical version, if you want to compare two players fairly, is to look at when each played and how many seasons the rate covers. A rate held across a decade through two or three tactical cycles is a stronger claim than one produced across three or four seasons in a single system, even if the second number is higher.

That is the honest reading, and it is also the one that makes the top of the table more impressive rather than less — the players who are up there across long spans have done something the format does not make easy.

A related point about longevity: a rate can only fall once a career extends past its peak, so a player who retired at thirty with a high rate is measured on his best years alone, while one who played until thirty-six has his worst included. Rates quietly punish the longest careers, which is the mirror image of what totals do to them.

One structural point about the floor is worth making explicit, because it decides who appears at all. Nine thousand minutes is roughly four full seasons of regular football, and it excludes a specific and interesting group: forwards who had two devastating years and then left, got injured, or declined. Several of those had rates that would top this table. They are absent not because the rate was fake but because four seasons is where a rate stops being a fluke, and any floor has to be drawn somewhere a reader can understand.