Johan Lange

[tomav84]

43 minutes ago, LondonLax said:

Surely there is no top flight team that doesn’t rely heavily on statistical analysis these days. It’s not some sort of secret only us message board fans have stumbled onto but hasn’t yet been discovered by the professional coaching and scouting staff at Premier League level.

reminds me when everyone was purring over Bielsa's little powerpoint presentation around the time of spygate. as if that same level of analysis doesn't happen at every single club in the top 2 leagues

[Zatman]

2 hours ago, mikeyp102 said:

I'm pretty sure Liverpool use statistical analysis, seems to work for them ok. 

They do follow the same approach thanks to the owners 

[sparrow1988]

11 hours ago, mikeyp102 said:

I'm pretty sure Liverpool use statistical analysis, seems to work for them ok. 

There was an article about a year back, in the NY Times I think, about their system. It's seriously impressive. They talk to some guy in it and he says he doesn't watch the games just analyses the data as watching the game distorts the data (a bit extreme if you ask me) but he apparently said to Klopp when he arrived that his last season at Dortmund wasn't as bad as everyone said and apparently they should have finished 2nd not 9th or something like that. Klopp says in the article only for the data side of the Liverpool set up he wouldn't have been appointed.

Now obviously not watching the games and basing your decision purely on data is a bit far fetched, but if you've got an data analyst who summarises the game and then provides this summary to the management team to be able to make more informed decisions then it's a no brainer for me. However, it's finding that balance that is the key.

[Mjvilla]

19 hours ago, TRO said:

Sorry, that was not meant.....but I enjoy our ding dongs.

I think, sifting through stuff, we both agree on our club becoming stronger is our prime mission.

UTV

 

100%. We just have different ideas on how we get stronger. There's no malice and it's why I love football. No 2 people have the same opinion.

[lexicon]

1 hour ago, sparrow1988 said:

There was an article about a year back, in the NY Times I think, about their system. It's seriously impressive. They talk to some guy in it and he says he doesn't watch the games just analyses the data as watching the game distorts the data (a bit extreme if you ask me) but he apparently said to Klopp when he arrived that his last season at Dortmund wasn't as bad as everyone said and apparently they should have finished 2nd not 9th or something like that. Klopp says in the article only for the data side of the Liverpool set up he wouldn't have been appointed.

Now obviously not watching the games and basing your decision purely on data is a bit far fetched, but if you've got an data analyst who summarises the game and then provides this summary to the management team to be able to make more informed decisions then it's a no brainer for me. However, it's finding that balance that is the key.

This, really - there's a lot of interpretation needed and it does have significant limitations, but it's definitely useful and serves a purpose. I'm not sold on all aspects of it, though, as I think things like 'expected goals' are an absolute nonsense. 

 

[sparrow1988]

47 minutes ago, lexicon said:

This, really - there's a lot of interpretation needed and it does have significant limitations, but it's definitely useful and serves a purpose. I'm not sold on all aspects of it, though, as I think things like 'expected goals' are an absolute nonsense. 

 

Yep that's exactly what I think as well. You'd be crazy to just ignore it.

However, I disagree on the expected goals. It gives you a purely statistical feeling (not the only metric by any means) of how a team are performing. For instance at the start of the season Leicester were well outperforming their expected goal difference stats - in the second half of the season the weren't getting the results that they were early in the season. Conversely, Southampton were totally underperforming - how many teams would have made a knee jerk reaction to the 9-0 considering their league position at that time? As they're a club that have relied heavily on data analysis, I'm sure Southampton's "decisions" department would have been made aware of the 'expected results or goal difference' of the team during that period and it would have played a part in sticking with the manager.

It seems a silly stat in isolation, but it's a decent indicator of how well a team is performing irrespective of the results. And the theory is that eventually the real goal difference stat will level itself out over the course of the season hence the respective turnaround in fortunes for those two clubs. Again it's finding the balance is what makes the difference not relying on one method or the other.

[TRO]

5 hours ago, lexicon said:

This, really - there's a lot of interpretation needed and it does have significant limitations, but it's definitely useful and serves a purpose. I'm not sold on all aspects of it, though, as I think things like 'expected goals' are an absolute nonsense. 

 

I am not sure, I understand "expected goals"....never devled in to it.

 

[TreeVillan]

20 hours ago, mikeyp102 said:

I'm pretty sure Liverpool use statistical analysis, seems to work for them ok. 

Aye it led them to signing Downing and Benteke.

[mikeyp102]

3 hours ago, TreeVillan said:

Aye it led them to signing Downing and Benteke.

And also Mane and Salah

[DaveAV1]

3 hours ago, TreeVillan said:

Aye it led them to signing Downing and Benteke.

I thought they signed Downing because of that fake video of him kicking balls into dustbins about half a mile away?

[Johnnyp]

Interesting watch this. Came across it this evening. Thought it would be applicable as we are told Lange is a big data guy. I'd be curious to see how Brentford apply their very unique method if they get promoted but they might surprise a few.

[CosmicVillan]

49 minutes ago, Johnnyp said:

Interesting watch this. Came across it this evening. Thought it would be applicable as we are told Lange is a big data guy. I'd be curious to see how Brentford apply their very unique method if they get promoted but they might surprise a few.

Interesting. Shef Utd had the equal least shots in the Premier League last season and next to bottom for shots on target. But top 4 for least shots conceded.

We were far and away the worst for shots conceded but mid table for shots taken. I wonder what the post lock-down stats were vs pre-lockdown. 

I think Shef Utd are punching above their stats and will do rather worse next season, whereas we would only need to be as good at the back as we have been for the last 10 games to jump up the table.

[Tomaszk]

7 hours ago, TreeVillan said:

Aye it led them to signing Downing and Benteke.

Ah now, that was plenty of the brodge. The brodge who fought against their director of football Edwards to sign his own players. 

As the first team manager, they let him do it and he did with gusto bringing in players like Benteke and Balotelli.

Salah incidentally was very much an Edwards signing. He had to convince Klopp to bring him in, Klopp wasn't totally sold initially.

Stats isn't the only way, but it's certainly a sound way of increasing your chances of success.

Then again, it's really easy not to buy bad players. Hopefully we'll show that this summer.

Edited August 3, 2020 by Tomaszk

[Muriel]

On 02/08/2020 at 23:24, tomav84 said:

reminds me when everyone was purring over Bielsa's little powerpoint presentation around the time of spygate. as if that same level of analysis doesn't happen at every single club in the top 2 leagues

1. There are for sure teams who barely use any statistical analysis in their recruitment but rely heavily on agents and connections (Probably us last year)
2. It is not as easy as using statistics. Brighton, Brentford and Liverpool have some of smartest guys in the country to create very complicated models. I still believe it's the most obvious and easiest competetive edge you can get in the game today, especially because british people are incredibly conservative and stubborn when it comes to football 

[Johnnyp]

22 minutes ago, Muriel said:

1. There are for sure teams who barely use any statistical analysis in their recruitment but rely heavily on agents and connections (Probably us last year)
2. It is not as easy as using statistics. Brighton, Brentford and Liverpool have some of smartest guys in the country to create very complicated models. I still believe it's the most obvious and easiest competetive edge you can get in the game today, especially because british people are incredibly conservative and stubborn when it comes to football 

I think live scouting will always be a key component in looking at a player but i think data and analytics are very important too and as the TED talk showed with the Brentford guy you don't make emotional human decisions when you've had some success and likewise when you are going through a sticky patch. I mean Thomas Frank went on a horrid run last season when he took over from Smith. If someone like the late doug was in charge of them, he'd of been gone long ago. Their data ( and they are very data driven ) was telling them otherwise. 

[TreeVillan]

Haven't watched the above video yet, but I'm sure I saw a Tifo video saying Brentford tried data analysis (moneyball) then quickly changed their minds to a more traditional approach?

I get a feeling people like the excitement of these new ideas such as moneyball as they have an air of mystery surrounding them, as it did with Tikka Taka ten years ago with Pep's Barca revolution, but only it turns out Pep doesn't play tikka taka and actually doesn't like the idea of it. 

[MotoMkali]

1 hour ago, TreeVillan said:

Haven't watched the above video yet, but I'm sure I saw a Tifo video saying Brentford tried data analysis (moneyball) then quickly changed their minds to a more traditional approach?

I get a feeling people like the excitement of these new ideas such as moneyball as they have an air of mystery surrounding them, as it did with Tikka Taka ten years ago with Pep's Barca revolution, but only it turns out Pep doesn't play tikka taka and actually doesn't like the idea of it. 

The guy in the video runs fc midjytallen (or whatever it is called) which is owned by the same guys who own Brentford. He knows what he is talking about 

[Knutserov13]

I would have expected an interview with Lange on the OS by now.

[mikeyp102]

1 hour ago, Knutserov13 said:

I would have expected an interview with Lange on the OS by now.

Personally I couldn’t care less if we never hear from him. Any statement from someone on a role like this is always going to be criticised,  if he says we want to buy low price and sell on, criticised. If he says we want to but top talent, criticised. If he says we have money to spend, clubs put prices up. 
Why as fans do we need to know what the sporting director is thinking. Let him be judged on his performance not on his words.

[mjmooney]

On 03/08/2020 at 08:15, sparrow1988 said:

he says he doesn't watch the games just analyses the data as watching the game distorts the data (a bit extreme if you ask me) but he apparently said to Klopp when he arrived that his last season at Dortmund wasn't as bad as everyone said and apparently they should have finished 2nd not 9th or something like that.

And there you have the problem with data analysis in a nutshell. 'Should have' finished 2nd according to the data, but actually finished 9th. Want to know why? Watch the games.