Moneyball and Footy

[festa5]

I think we are trying and will play more direct that Carrick. But it won't be route 1 direct. It'll be 4 or 5 quick pass direct.
Rather than trying to build up with 20 passes.
More Liverpool than Man City.

We've got good enough technical players to play better football than route 1.

Yeah I think so. Less keep ball at the back. Less "recycling possession". More attempts at crosses, dribbles, through balls etc.

But fundamentally quite similar in the technical requirements required. Carrickball but with more forward passes and hopefully a higher tempo.

Some people were expecting us to recruit a load of big lads and play like Luton in their promotion season.

 

[jonny_greenings_sock]

Funny you say that about Kante, I was talking to someone recently who was a coach within a pro club but now holds coaching seminars etc and he told me one of the first stats they look at for players right across the pitch is "ball recoveries". I.e. when they are the first one to touch the ball and secure possession for the team after it touched the opposition. It includes tackles, interceptions, 2nd balls etc. It's a very useful stat as it does catch some of the more subjective data like positioning, ability to read the game, reaction speeds etc.

Even players like Messi, who I would imagine has a fairly low tackling output, will score well on ball recoveries because he positions himself well and he reacts to loose balls and picks them up second balls extremely well. Pep's teams are rammed with players with high "ball recovery" stats and they have an emphasis on winning the ball back quickly when they lose it.

I know Pep has always been big on their collective ball recovery time as well. I think it's something Edwards will focus on a lot more than Carrick did.

Fascinating. Just googled it out of curiosity:

During his time with Caen, N'Golo Kanté was known for his exceptional ball recovery skills. In the 2014-2015 season, his final season with Caen, he recovered the ball more times than any other player in Europe. This was a standout statistic for the season and highlighted his key role in Caen's midfield.

 

[Marvanelli]

Xg haters are coming for you

 

[Sheriff_John_Bunnell_ret]

Billy beane was successful for a season or two. Then everyone copied what he was doing. You need to be ahead of the game to get an advantage.

 

[goalscrounger]

One of the interesting things about that article was when he said that one of the things the best management do is not always fill a weakness.

If the opportunity comes for long term value in a strength then you take that.

It is all about taking value and trying to improve over many years rather than overspending to constantly plug gaps if there is not value available there.

Yes, I thought that too, although I do wonder whether this is as relevant given the lack of depth for each position in a football squad compared to baseball one. It was all part of the message of "build for the long term, not the short" (which is what Scott has been trying to do to a certain extent to be fair).

 

[B_G]

One of the interesting things about that article was when he said that one of the things the best management do is not always fill a weakness.

If the opportunity comes for long term value in a strength then you take that.

It is all about taking value and trying to improve over many years rather than overspending to constantly plug gaps if there is not value available there.

I think this has been very evident under KS.
We see glaring holes not filled while we buy another inverted winger.
It feels like there has been more priority on finding good value players rather than addressing squad deficiencies. Either that or we aren't very good at recognising what we need.

It will be interesting to see if that changes under Edwards

 

[Ravshoutsatbeck]

I think this has been very evident under KS.
We see glaring holes not filled while we buy another inverted winger.
It feels like there has been more priority on finding good value players rather than addressing squad deficiencies. Either that or we aren't very good at recognising what we need.

It will be interesting to see if that changes under Edwards

I think this is the case.

You do fill some holes, but if someone of value comes up in a position of strength - long term that still might make the most sense.

For example, getting Conway last summer. Striker was arguably one of our strongest positions but getting a 21 year old we felt could grow for a cut price in the last year of his contract made sense.

Same with looking at Sulc despite Azaz, whittaker and Mcgree all being top championship players.

 

[atypical_boro]

The thing with xG is that it is partially subjective, and its use isn't really clearly defined as its a measure of what, arguably, 'should' have happened versus what happened.

Middlesbrough came second in xG last season in the Championship, and finish tenth. What does this tell us beyond what we already knew? We didn't score enough goals?

It is indicative rather that conclusive, and has limited value.

 

[Ravshoutsatbeck]

The thing with xG is that it is partially subjective, and its use isn't really clearly defined as its a measure of what, arguably, 'should' have happened versus what happened.

Middlesbrough came second in xG last season in the Championship, and finish tenth. What does this tell us beyond what we already knew? We didn't score enough goals?

It is indicative rather that conclusive, and has limited value.

We created more chances than the opposition... for most of the season.

I think that is true.

But it got worse as the season wore on.

And we had a lot of low level chances, but gave up a small number of high xg chances.

 

[atypical_boro]

We created more chances than the opposition... for most of the season.

I think that is true.

But it got worse as the season wore on.

And we had a lot of low level chances, but gave up a small number of high xg chances.

Yes, but what does that really mean? We created lots of decent chances and didn't score enough of them.

What have we learnt from that that weren't detectable from other stats? How do you address that specific problem, in a way that is informed by that stat above others?

We didn't score enough goals to challenge for promotion, regardless of the quality of the chances.

 

[Marvanelli]

The thing with xG is that it is partially subjective, and its use isn't really clearly defined as its a measure of what, arguably, 'should' have happened versus what happened.

Middlesbrough came second in xG last season in the Championship, and finish tenth. What does this tell us beyond what we already knew? We didn't score enough goals?

It is indicative rather that conclusive, and has limited value.

It is still relatively small in the size of data. 1 season still isn't a huge amount of data in reality.

But what it does show, that based on the small sample size, our finishing was below what is expected.

If you offered us at the start of the season we would finish up with the 2nd best xG....then I'd reckon every single club in the league would snap your hands off at that.

Of course there are caveats. How close is 3rd on the list? And what's the difference between 1st and 2nd, in xG and most importantly, what's the xGC like? It wouldn't be great if we were 2nd in xG but very bottom in xGC...

There's so much more than xG to calculate and analyse a season.

But xG is still useful.

 

[atypical_boro]

But xG is still useful.

I'd go as far as to say its interesting, I'm not actually sure its useful.

 

[Marvanelli]

I'd go as far as to say its interesting, I'm not actually sure its useful.

It is almost definitely useful.

It is used by nearly every top team in the sport. And in fact 'xG' is used in all other sports.

Even the likes of Golf. Look at DeChambeau. He relies heavily on data and analytics to help improve his game strategy and decision-making.....which has helped him win big tournaments, including the US Open.

Stats are there to be used.

xG is one of many stats that are very very useful.

 

[newusername]

Billy beane was successful for a season or two. Then everyone copied what he was doing. You need to be ahead of the game to get an advantage.

'twas ever thus.

Whether it was Scots in 1880s, black players in 1970s, Swedes in 80s or French in 90s, a manager sees an opportunity to acquire under valued players, has a short period of success and then the big clubs pile in and snaffle up the best talent while driving the prices up for all.

 

[Brian Marwood]

It is almost definitely useful.

It is used by nearly every top team in the sport. And in fact 'xG' is used in all other sports.

Even the likes of Golf. Look at DeChambeau. He relies heavily on data and analytics to help improve his game strategy and decision-making.....which has helped him win big tournaments, including the US Open.

Stats are there to be used.

xG is one of many stats that are very very useful.

Depends on the stat, if DeChambeau is told from his stat people that he is 1.4% more likely to get a birdie in a Thursday, what use is that?

 

[morrissey]

I found really interesting the bit about not necessarily concentrating on the positions that are deemed 'weak', but strengthen the area where you are already strong

 

[newusername]

I found really interesting the bit about not necessarily concentrating on the positions that are deemed 'weak', but strengthen the area where you are already strong

Only if there is more value in doing that.

 

[Moreno]

Baseball and statistics go hand in hand. It’s much more difficult in a football environment.

 

[changingman]

It is almost definitely useful.

It is used by nearly every top team in the sport. And in fact 'xG' is used in all other sports.

Even the likes of Golf. Look at DeChambeau. He relies heavily on data and analytics to help improve his game strategy and decision-making.....which has helped him win big tournaments, including the US Open.

Stats are there to be used.

xG is one of many stats that are very very useful.

Agree, and I'd add that one of the true values in any dataset is when it's used comparatively. We as fans (and the club staff) will labour over every single detail in every single match of ours but are unlikely to do so for the other 23 teams in the league. Stats are a quick route to understand competitors strengths and weaknesses and understanding, relatively, how we should be performing within the overall league.

Likewise the xG point above - by understanding our issue we can look to others historically with similar issues to understand how they addressed it, what worked, what didn't etc.

 

[Brian Marwood]

Agree, and I'd add that one of the true values in any dataset is when it's used comparatively. We as fans (and the club staff) will labour over every single detail in every single match of ours but are unlikely to do so for the other 23 teams in the league. Stats are a quick route to understand competitors strengths and weaknesses and understanding, relatively, how we should be performing within the overall league.

Likewise the xG point above - by understanding our issue we can look to others historically with similar issues to understand how they addressed it, what worked, what didn't etc.

I don't, I watch a game and decide whether we got what we deserved or not. And then move on.