Opta report

Hi all,

The chron have published Optas latest stats for the season so far. I'm certainly not going to pore over all the details, although everyone else feel free! A couple of mad things to point out though: Matthieu Debuchy is our highest overall rated player this season so far. Even more surprising is that we have played the fewest long balls of any club in the top flight. This is quite a turnaround and shows in our improved results so far.

Any other thoughts?

Link is here: http://www.chroniclelive.co.uk/spor...s/stats-show-newcastle-united-sharper-6311292

 

Keep playing the way we did against Spurs and we'll be Champions next season.

 

To think WhoScored will probably have hired actual mathematicians and they come out with that drivel.

 

Thought you were away on holiday.

 

You can pul anything you want from stats. Still think its pretty interesting, and Opta are pretty widely used.

 

If nufczec is online, you can't leave rep if you've got zero rep power.

Maybe a Mod can explain it to him.

 

I'm not surprised at all Debuchy has been class the last 5 games or so. Also we clearly have stopped pumping the ball upfield. I think pards is doing a pretty good job at the moment.

 

I found that incredibly boring would be my thoughts. I just think the whole analysis by stats is a bit mind numbing.

I definitely would not rely on stats to work out who is the top performer, your own eyes do a better job of that. While there is no doubt that we have played less long balls this season, it doesn't tell the full story because it doesn't define what a long ball is and it is too easy to say "we're doing better because we are passing it shorter". We're passing it shorter because confidence has been restored and players are more willing to take possession. The fear has gone slowly over the course of the games so far.

Pretty simple for me, as the season has progressed we've got ourselves a bit of confidence, got ourselves a bit more organised and we're doing ok as a result.

 

1 Arsenal 11 12 25
2 Liverpool 11 11 23
3 South'pton 11 10 22
4 Chelsea 11 8 21
5 Man Utd 11 5 20
6 Everton 11 4 20
7 Tottenham 11 3 20
8 Man City 11 16 19
9 Newcastle 11 -1 17
10 West Brom 11 0 14
11 Aston Villa 11 -1 14
12 Hull 11 -5 14

Ninth place, just 3 points behind Man Utd, Everton and Spurs and two behind City isn't too bad, I think we'd have all settled for that before the season started.

 

please log in to view this image

 

Good article by Lee Ryder.

 

I'd be over the moon if we finished the season in this position!

 

Pardew has specifically mentioned he's using statistical analysis this year, and I think it's showing. It's a fact that on average the team that has the most shots on target wins the game, and we've consistently outshot our opponents this year (other than spurs this weekend where we became the lucky statistical anomaly thanks to Tim Krul) giving us the best possible chance of winning, unlike last year.

I know there's a lot of resistance to stats from some people, but they are the future as far as I'm concerned. A few things that stats are predicting this year, let's see if they happen:

Spurs will be top 4
Cardiff will be relegated
Arsenal won't win the title

 

Do you know how hard it is to model 22 free bodies, in a sport as dynamic as football, where the only well-defined, non-subjective aspect of it is that you have to score more goals than the opponent? And even if you were capable of constructing a model within a set of given subjective parameters and extracting something vaguely useful - which I don't think you really can - it's application in coaching is pretty limited and it's not something that's ever going to made available to the general public, because then the teams lose whatever advantage they might have gained. And even if, even if, you were capable of constructing a singular metric for a specified component of the game, you can't just go 'well, X>Y, so team/player A does this better than team/player B' - that's mathematically incorrect.

But blah, blah, blah - you watched MoneyBall, you might have read a book, maybe you frequent a blog with pretty graphs, and anyone opposed to extensive statistical analysis in football is part of some archaic anti-intelligentsia. Well, just remember, this is you Billy Beane:

please log in to view this image

Pardew's always used stats, FWIW. Him and Allardyce. Analytics4Eva.

 

Can't say I've seen anywhere Pardew suggest he's using statistical analysis more this year? Certainly stats are not the reason for our improvement in performance

 

The basic reality being that stats have their place, its just a small one.

 

Get a signed statement from him when you have your weekly meeting.

 

Will do. This week me, him and Jim will be looking at the stats on how many statistical analyst's it takes to completely bamboozle a set of PL footballers to the point they forget how to kick a ball.

 

Ignorance is bliss eh?

Point about Allardyce is that stats have disproved his approach to the game.long ball was developed as a result of statistical analysis, and then disproved when more detailed statistics became available. Not that you'd care as you'd prefer head in the sand "I can see it with my eyes I don't need stats". Good luck with that :-£

 

I don't think he said it in so many words but the language he's used and his behaviour with substitutions etc has a pretty blatant stats focus