This is interesting, I remember commentators talking about this stuff during the EUROs. While football stats are nearly always irrelevant because they’re stats of the wrong things, they’re interpreted by people with no understanding of statistics or maths or the scientific method whatsoever, clearly there is plenty of room for improvement and I like this impect and packing stuff, it at least represents some things better than the usual suspects, as the article points out, passing %, shots, possession, corners etc. The Brazil - Germany game is a great example.
However clearly this system is still far from claiming completness, this argument for example:
. Impect disregards all passes that go backwards or don’t beat any defender. It’s a correct assessment, a pass that doesn’t travel forwards, doesn’t help you score. Can’t argue with that.
I can most definitely argue with that. Another example:
Kroos beat 85 defenders per game (!), the average of all defensive midfielders is at 28.
A team that expects to own possession plays higher up the pitch and will have fewer opponents playing higher up the pitch intercepting passes that look to go past defenders from a (supposed) defensive (?) midfielder like Kroos. There are examples of other defensive midfielders who could do a far better job for what their tactics demand from them than Kroos, which doesn’t say anything about passes beating defenders. Defensive midfielders who are brilliant at killing off spaces and passes by showing the opponent to a poorer passing alternative is not covered at all in any statistic afaik but might be absolutely massive in terms of how the game plays out just to take one example. so:
Impect metrics can totally tell you who played “better” and “should” have won or not lost, which could be huge on many levels.
talent evaluation will be revolutionized
The stats Reinartz and Hegeler came up with are nothing less than a quantum leap for football analytics
Is of course an exaggeration (and quantum leap is a funny misuse of the word since it is a leap so short it happens within the size of an atom ). But it is definitely a statistic that adds something different to the regular crap of possession, corners and shots that we’ve been fed since forever. Ideally one wants the whole picture and this can only help. I’m looking forward to what else kind of stats we can get in the future.