The Stats Thread

Are TWTD claiming they've won the 2020 derby match?

 

Not in the same league, or even close for the first half of the year. And it looks likely that situation will continue for all of this new year.

 

I posted this on the season thread, but will include it here for stats fans wondering whether we 'deserve' to be top of the table.

https://norwichcity.myfootballwrite...-of-the-championship-what-do-the-numbers-say/

 

1986 would be 34 years not 44

 

Comparing the two seasons, we have the same number of points this season after 23 games as in 2018-2019, despite scoring 10 fewer goals. Part of the reason is that all goals are not equal when it comes to earning points. At this stage in 2018-2019, 10 of the 41 goals we'd scored made no difference to the number of points earned; this season, on the other hand, we have only twice had the luxury of a 2+ goal margin. Put another way, we don't need Pukki to match his 2018-19 scoring feat -- as long as the goals he does score keep the points total ticking upwards.

 

I seem to recall one of the commentators for the QPR game saying that Pukki's current total was the same as he had at this stage two years ago. Buendia looks odds on to easily exceed his goals total that year and probably his assists as well.

 

Buendia is so important to us , when his tail is up he's everywhere on the pitch defending corners, breaking up play he switches both wings he's honestly like 3 players for us .

IF we sold him then I would have serious doubts about promotion he's immense and that finish yesterday was World class like vintage Bergkamp.

I can see him playing at the very highest level and getting into the Argentina squad .

Possibly the best player to ever wear the yellow shirt imho

 

Currently 11 goals compared to 12 two years ago. That season we had several players chipping in with more than the odd goal -- Rhodes, Onel, Stiepi, even Klose. At the moment, Pukki and Vrancic (3) apart, Emi is single-handedly making up for goal-shy team mates.

 

The attacking trio which started the game for the Tykes had a combined 87 touches during the game, according to WhoScored.com. Buendia had one fewer than that on his own

 

Your rightful correction of the dates aside, it is some stat. Krul’s terrific

However I cannot help but feel that our clean sheet yesterday was more down to Gibson’s return to the starting line up.

When you look at how much Zimmermann has been partly or largely responsible for goals we have conceded recently, I think that adds to the conclusion.

I say that through gritted teeth because Zimmo was brilliant for us until that twat at West Ham scythed him down (and should undoubtedly have been red carded). Unfortunately the injury seems to have cost him more than a yard of pace and, worst of all, his previously impeccable timing of a tackle. Hopefully he just needs more recovery time.

 

It's possible that injuries have effected Zimbo's performances, but I suspect that the time out with injuries could be a more important factor and that he can get back to good form soon. The other aspect is that Handley and Gibson have been playing so well, lifting the bar.

 

My teeth gritted too, Rob. CZ looks as if the injury is still affecting him psychologically, almost as if he expects to break down again at any minute. Hence the hesitation, which is critical.

 

https://www.canaries.co.uk/News/2021/january/in-numbers-citys-2020-21-season-so-far/

 

For a while now I've wanted to start a discussion about stats and how useful they are in football analysis. I admit my own bias tends towards the sceptical and a belief that stats are too often an attempt to deny the inevitability of subjectivity (in a range of activities, not only in football), but I am genuinely interested in people like Robbie who clearly think otherwise. But the atmosphere around stats seems to have become so polarised and rancorous on here that this probably isn't a good time to do it. Anyone up for it, with a promise not to end up name-calling?

 

I also like to objectify aspects of sport through stats as much as possible because the alternative is simple gut feeling about this or that player or how the team as a whole has done. This is why I use WhoScored's player ratings based on a range of achievements in a particular match. These include goals scored and assists, but also total shots, shots on target, pass success ratios (PS%), dribbles, aerials won, tackles, interceptions and so on. These aren't judgements, they are totals of key contributions for each player and then the team as a whole to allow an objective assessment of players and teams. They also amalgamate these into season totals based on total points divided by number of matches played. The beauty is that they do all the calculations for easy to read graphics.

Other stats like expected goals (xG) and expected goals against (xGa) are also useful, but require judgements about the likelihood of scoring from each shot for or against, which is why different xG and xGa compilations produce different totals for the same match. They also don't include match issues like one team defending a lead and therefore defending more and attacking less. I therefore use WhoScored first, then other factors like possession%, corners, xG and xGa to add further dimensions.

 

Pretty much agree with this as long as people realise that stats are a tool to be used, but not the be all and end all. As I think I've mentioned before one of the main benefits of stats, particularly with developing players and teams is to identify trends. After all a Championship winning team is not established through a single individual stat (Game result), but from a series of results over a season (league table).

Judgements should not be made on small samples i.e. a single game, after all even the best players and teams have off days.

 

Luckily he can still tackle with his head! He'll be back. I hope he finds his mojo soon though otherwise he'll miss the chance to prove himself with the way Gibson and Hanley are performing.

Likely Coventry will be that chance.

 

As River End says, the best thing about stats is that they can prevent confirmation bias, where we see every mistake a player we don't like makes and don't notice the good stuff he does.

Some genuine questions for Robbie: has there been any attempt to change the situation where a ball that thumps off the crossbar or post is not a shot on target, and yet a weak effort that goes straight into the goalkeeper's hands and is close to a back pass is a shot on target? And who (or what, perhaps, if it's an algorithm?) decides what is a 'key pass' or 'big chance'? Also, is any effort made to distinguish between a real assist (where a player intended to make the killing pass and has shown great vision a la Buendia) and an assist which is counted just because player A is the last person to touch the ball before player B goes on a brilliant run and scores a wonderful solo goal? And do you believe that it all evens out in the end?