Jump to content
IGNORED

On to the serious stuff - Swansea v Brentford


Lanterne Rouge

Recommended Posts

3 minutes ago, Kid in the Riot said:

Fair play to Brentford, if they win the final they'll be the 50th Premier League club. 

We'll remain the biggest club in English football not to get there. 

No real sign of getting there anytime soon either, sadly. Watching tonight’s game shows how far away we are.

This appointment is huge for us.

Link to comment
Share on other sites

1 minute ago, stephenkibby. said:

Lew feel the pain but am now in the whatever mode. Do you think a top man is on his way? Me neither.

Nobody really knows mate. It could be absolutely anybody at this stage, 4 weeks in!

 

Link to comment
Share on other sites

28 minutes ago, Kid in the Riot said:

Fair play to Brentford, if they win the final they'll be the 50th Premier League club. 

We'll remain the biggest club in English football not to get there. 

Even more depressing fact on a night we maybe are thinking of going for it big time by employing the crap managers assistant 

Link to comment
Share on other sites

2 minutes ago, Cov 77 said:

Even more depressing fact on a night we maybe are thinking of going for it big time by employing the crap managers assistant 

I don't want Holden but, given going for the previous manager's assistant is the exact reason Brentford are where they are, this seems an odd time to make this point. Brentford are a great advert for how promoting from within sometimes works...

Link to comment
Share on other sites

41 minutes ago, RedRock said:

Said a few seasons back, don’t bother with our scouting, just bug Brentford’s analytical team and offer a few £££ more than them for the players they identify. 

Quite depressing how far off we are from mounting a realistic challenge for the Premier League. 

They have proper data analytics.

Link to comment
Share on other sites

38 minutes ago, BritAbroad said:

I get the impression that Brentford will be like Norwich, in terms of spending, and the board won't be harsh on Frank if they come back down (assuming they go up, that is).

I think Brentford are structurally more secure a side than Norwich of 2018/19 were tbh.

That said they would need strengthening but wouldn't surprise if they looked at the Norwich and quite notably, Burnley type model if they went up.

Link to comment
Share on other sites

11 minutes ago, LondonBristolian said:

I don't want Holden but, given going for the previous manager's assistant is the exact reason Brentford are where they are, this seems an odd time to make this point. Brentford are a great advert for how promoting from within sometimes works...

....providing the previous manager was any good.

 

Link to comment
Share on other sites

10 minutes ago, LondonBristolian said:

I don't want Holden but, given going for the previous manager's assistant is the exact reason Brentford are where they are, this seems an odd time to make this point. Brentford are a great advert for how promoting from within sometimes works...

Yes...when you are already relatively successful and you have talented backroom staff, it can pay off to have continuity, and recruit from within?   But I don't personally believe that we are of the same standard of Brentfords coaching staff. IWhat can work for one club does not necessarily mean it will work at another IMHO.

Link to comment
Share on other sites

1 hour ago, JollyestOli said:

Nowhere near as good as Leeds, as evidenced when they played each other.

Had them down as the second best tbh, seemed more dominant than West Brom- yet just blew up in the last two.

Had a feeling they may drop points at Stoke, as in draw, but didn't foresee them losing both!

Leeds certainly outplayed, if not dominated Brentford over both games!

Link to comment
Share on other sites

1 hour ago, maxjak said:

I  am genuinely curious about this, could you please expand and give us the benefit of your knowledge?  What are data analytics exactly, and why don't City have them?

have you seen the film or read the book - Moneyball?  If so, you’ll see that data plays an important part in Baseball, and some very clever people understood the weaknesses in looking at the wrong stats.  Too much focus on Runs and Hits, rather than the previously under utilised On Base Percentage (OBP), e.g. if you don’t get on base you can’t score.  They also spotted that batters who were “walked” to first base by the pitcher throwing fouled balls rather than risking being hit for home runs was also undervalued.

That concept has come into football.  It is not a perfect science, because football is fluid, but the clever people have spotted the ability to use data better, both in recruitment / talent id and performance (both physical and on-pitch).

Brentford’s success is built on the foundation of Matthew Benham and Rasmus Ankersen, two massive exponents of the use of data, one in the world of gambling, the other in football, both heavily combined initially in the success of FC Midtjylland....the Danish connection to Brentford.  It is fair to say their recruitment has been exemplary.

I make no apologies for sharing this video of Ankersen....it resonates very closely with our position re Goal Difference.

So at the heart of Brentford you have considerable data knowledge and skill, and a department working there based on their fundamentals.

We have a talent Id team (data analysts) ran by Sean Gilhespy.  I can’t comment personally, some on OTIB might.  Let’s just say that our team is still evolving and Brentford’s is mature. ?

We also have Performance Analysts ran by Sam Stanton.

If you see the little bar charts I create, they come from professional scouting data / software.  So I can prove that any old muppet can access data and present stuff.  The skilled operators will be producing much more sophisticated and context rich data / analysis than I can, joining one piece of data with another to create links.  Simple stats like pass completion are taken to great depth by skilled data analysts linking it to things like:

- who they passed to

- where did they pass it from

- where did it end up

- likelihood of a chance from that pass

creating good visuals to simplify the raw numbers.  This is from a guy called Swans Analytics, who along with a small group of similarly minded folk, created a company called Mrkt Insights, providing football data consultancy.  They pitched a recruitment review to several clubs, and won a contract to work with Swansea City, amongst others.  Another guy in their company is at Peterborough.  The data I get is cheap....the data they get with x,y coordinates costs lots and lots.  But you can do so much with it.

image.thumb.png.c9715a9dab517a252eab42bec87dfb25.png

You don’t need a PhD in Football to get the message across.  The above shows the passes made by Pack and Brownhill last season.  The darker the colour, the more chance the pass creates a chance...called “passcore”.  Not dissimilar to Jens Hegeler’s “packing” stat

Im sure @Lrrrwill give you a better explanation.

Link to comment
Share on other sites

1 hour ago, maxjak said:

I  am genuinely curious about this, could you please expand and give us the benefit of your knowledge?  What are data analytics exactly, and why don't City have them?

Well to give an indication how recent its use in football is Man Utd (less than 6 months) and Liverpool (year or two) have only just set up departments for it specifically, Brentford are a LONG way ahead of the curve. Davefevs provides a good explanation tbh so I won't add to it atm partly because I'm too tired to think properly and off to bed :sleeping: But on twitter if you find the right accounts you can see some excellent work done which is used to examine players from large source groups. Any of the guys involved with Mrkt Insights as Fevs says would be a good place to start.

Link to comment
Share on other sites

8 hours ago, maxjak said:

Yes...when you are already relatively successful and you have talented backroom staff, it can pay off to have continuity, and recruit from within?   But I don't personally believe that we are of the same standard of Brentfords coaching staff. IWhat can work for one club does not necessarily mean it will work at another IMHO.

That is how Liverpool did it from Shankly onwards. 

Worked very well until they brought in, an "insider" in Souness  after a spell away.

But in their case, the first manager, Shankly, was a rip roaring success so those that followed just had to keep it going.

In our case, Holden would be following someone who I consider as one of our worst managers ever when one considers how much he was allowed to squander on transfer fees and the regular "losing for months" runs.

Is Holden experienced enough to change that into a promotion winning side? Probably not, so we must recruit one who will change us for the better.

Link to comment
Share on other sites

9 hours ago, LondonBristolian said:

I don't want Holden but, given going for the previous manager's assistant is the exact reason Brentford are where they are, this seems an odd time to make this point. Brentford are a great advert for how promoting from within sometimes works...

But about the only similarity between Dean Holden's background and Thomas Frank's is that they have been assistant managers at a club looking for a new manager. Frank managed Danish national youth sides - got the u17s to the European Champs semi-final, managed the u19s - and got Brondby to successive Europa cup qualifications. He joined the national sides in 2008. He was an amateur footballer and a specialist coach. By almost complete contrast Dean, after a long playing career, has been a coach since 2014 at Walsall and Oldham before he came here. He was manager at Oldham briefly.

Now, if we were looking for a Thomas Frank character I'd be happy!

Link to comment
Share on other sites

9 hours ago, Davefevs said:

have you seen the film or read the book - Moneyball?  If so, you’ll see that data plays an important part in Baseball, and some very clever people understood the weaknesses in looking at the wrong stats.  Too much focus on Runs and Hits, rather than the previously under utilised On Base Percentage (OBP), e.g. if you don’t get on base you can’t score.  They also spotted that batters who were “walked” to first base by the pitcher throwing fouled balls rather than risking being hit for home runs was also undervalued.

That concept has come into football.  It is not a perfect science, because football is fluid, but the clever people have spotted the ability to use data better, both in recruitment / talent id and performance (both physical and on-pitch).

Brentford’s success is built on the foundation of Matthew Benham and Rasmus Ankersen, two massive exponents of the use of data, one in the world of gambling, the other in football, both heavily combined initially in the success of FC Midtjylland....the Danish connection to Brentford.  It is fair to say their recruitment has been exemplary.

I make no apologies for sharing this video of Ankersen....it resonates very closely with our position re Goal Difference.

So at the heart of Brentford you have considerable data knowledge and skill, and a department working there based on their fundamentals.

We have a talent Id team (data analysts) ran by Sean Gilhespy.  I can’t comment personally, some on OTIB might.  Let’s just say that our team is still evolving and Brentford’s is mature. ?

We also have Performance Analysts ran by Sam Stanton.

If you see the little bar charts I create, they come from professional scouting data / software.  So I can prove that any old muppet can access data and present stuff.  The skilled operators will be producing much more sophisticated and context rich data / analysis than I can, joining one piece of data with another to create links.  Simple stats like pass completion are taken to great depth by skilled data analysts linking it to things like:

- who they passed to

- where did they pass it from

- where did it end up

- likelihood of a chance from that pass

creating good visuals to simplify the raw numbers.  This is from a guy called Swans Analytics, who along with a small group of similarly minded folk, created a company called Mrkt Insights, providing football data consultancy.  They pitched a recruitment review to several clubs, and won a contract to work with Swansea City, amongst others.  Another guy in their company is at Peterborough.  The data I get is cheap....the data they get with x,y coordinates costs lots and lots.  But you can do so much with it.

image.thumb.png.c9715a9dab517a252eab42bec87dfb25.png

You don’t need a PhD in Football to get the message across.  The above shows the passes made by Pack and Brownhill last season.  The darker the colour, the more chance the pass creates a chance...called “passcore”.  Not dissimilar to Jens Hegeler’s “packing” stat

Im sure @Lrrrwill give you a better explanation.

Thanks so much for taking the time and trouble to educate me. Yes,   I have seen Moneyball, which i enjoyed.  So we do have inhouse data analytics at City, but it is in it's infancy compared to Brentford?  And I loved the Video, I had no idea that Brentford were owned by an Oxford educated professional gambler, which explains a lot about their success.

Link to comment
Share on other sites

Talking about appointing from within..............I see that Brentford have two Danish assistant managers who work with Thomas Frank.............Lars Friijs and Brian Riemer, I have no idea if either would be interested, but it is often wise to replicate success by using the same ideas that have succeeded elsewhere.     It is a complete shot in the dark, but i wonder if JL n' MA even considered this route?

Link to comment
Share on other sites

13 hours ago, Lew-T said:

No real sign of getting there anytime soon either, sadly. Watching tonight’s game shows how far away we are.

This appointment is huge for us.

No need to worry my friend, we will shortly have the messiah Dean Holden as Manager

Link to comment
Share on other sites

12 hours ago, Davefevs said:

have you seen the film or read the book - Moneyball?  If so, you’ll see that data plays an important part in Baseball, and some very clever people understood the weaknesses in looking at the wrong stats.  Too much focus on Runs and Hits, rather than the previously under utilised On Base Percentage (OBP), e.g. if you don’t get on base you can’t score.  They also spotted that batters who were “walked” to first base by the pitcher throwing fouled balls rather than risking being hit for home runs was also undervalued.

That concept has come into football.  It is not a perfect science, because football is fluid, but the clever people have spotted the ability to use data better, both in recruitment / talent id and performance (both physical and on-pitch).

Brentford’s success is built on the foundation of Matthew Benham and Rasmus Ankersen, two massive exponents of the use of data, one in the world of gambling, the other in football, both heavily combined initially in the success of FC Midtjylland....the Danish connection to Brentford.  It is fair to say their recruitment has been exemplary.

I make no apologies for sharing this video of Ankersen....it resonates very closely with our position re Goal Difference.

So at the heart of Brentford you have considerable data knowledge and skill, and a department working there based on their fundamentals.

We have a talent Id team (data analysts) ran by Sean Gilhespy.  I can’t comment personally, some on OTIB might.  Let’s just say that our team is still evolving and Brentford’s is mature. ?

We also have Performance Analysts ran by Sam Stanton.

If you see the little bar charts I create, they come from professional scouting data / software.  So I can prove that any old muppet can access data and present stuff.  The skilled operators will be producing much more sophisticated and context rich data / analysis than I can, joining one piece of data with another to create links.  Simple stats like pass completion are taken to great depth by skilled data analysts linking it to things like:

- who they passed to

- where did they pass it from

- where did it end up

- likelihood of a chance from that pass

creating good visuals to simplify the raw numbers.  This is from a guy called Swans Analytics, who along with a small group of similarly minded folk, created a company called Mrkt Insights, providing football data consultancy.  They pitched a recruitment review to several clubs, and won a contract to work with Swansea City, amongst others.  Another guy in their company is at Peterborough.  The data I get is cheap....the data they get with x,y coordinates costs lots and lots.  But you can do so much with it.

image.thumb.png.c9715a9dab517a252eab42bec87dfb25.png

You don’t need a PhD in Football to get the message across.  The above shows the passes made by Pack and Brownhill last season.  The darker the colour, the more chance the pass creates a chance...called “passcore”.  Not dissimilar to Jens Hegeler’s “packing” stat

Im sure @Lrrrwill give you a better explanation.

It's fantastic how they've gone about recruiting and building those two sides Dave.

A couple things I've wondered...did they think of a way they wished to play and then recruited accordingly, or did they just look at the undervalue of certain players and build around that? My gut instinct says the former.

Also...how much do they value the 'DNA' of a player compared to us? Do they worry that he maybe a complete duffus, but can do the job on the pitch. Watkins comes to mind on that one...I couldn't see him being taken on for us,although we had the chance, as by all accounts he's a bit of a dick. Does this hold us back?

Brentford also seem to like to gamble more on complete 'unknowns'...they really do believe in their system.

I find it fascinating and great to follow.

Link to comment
Share on other sites

5 hours ago, maxjak said:

Thanks so much for taking the time and trouble to educate me. Yes,   I have seen Moneyball, which i enjoyed.  So we do have inhouse data analytics at City, but it is in it's infancy compared to Brentford?  And I loved the Video, I had no idea that Brentford were owned by an Oxford educated professional gambler, which explains a lot about their success.

I guess that MA tried to implement some form of data analytics in 2012 under McInnes, when he came in as a consultant.  I understand SOD didn’t really use it when he came in, and Cotts binned it, relying on his eye and his network (with Keith Burt). Lo and behold Ashton came back in...in my opinion...to do SL’s dirty work, to sack Cotts.  Database and Wyscout (or Pro7 or whatever) was re-implemented then.  But not to the degree of Brentford.  As you say, they are seriously data led.

Link to comment
Share on other sites

20 hours ago, maxjak said:

I  am genuinely curious about this, could you please expand and give us the benefit of your knowledge?  What are data analytics exactly, and why don't City have them?

Dave has explained in great detail, all I’ll add is :

Yes, we have data analytics. We likely use the same software that Brentford use. 
Our problem is the quality of the people needed to interpret that data and translate it to a team/system/identity. 

 

Our approach is scattergun rather than tailored. 

Link to comment
Share on other sites

11 minutes ago, Harry said:

Dave has explained in great detail, all I’ll add is :

Yes, we have data analytics. We likely use the same software that Brentford use. 
Our problem is the quality of the people needed to interpret that data and translate it to a team/system/identity. 

 

Our approach is scattergun rather than tailored. 

I’ll give an example of that :

You are going to sell Bobby Reid. Your head coach views him as pivotal to the system he currently wants the team to play. Bundles of energy, ability to lead the press (and the intelligence to know when), whilst on the ball being able to have composure and awareness once the ball is won to create intelligent passing angles or retain the ball for support/counter attack. 
 

You need the data analytics team to come up with a potential replacement. 
So they scour the data and think “mmm....Bobby does the most running in the team, and can play upfront or in the AM role”. 
Let’s look at Andy Weimann. 
 

The analysts would likely have compared the running stats, seen that Weimann can run as much as Bobby as well as chip in with a few goals and assists. 
 

What they wouldn’t have looked at was how he fits the existing team structure and the way the coach wants to play. Running around is all well and good, but Bobby did it with intelligence and ability whereas Weimann does it just because he has the energy, whilst his ball retention and awareness is a long way off Bobby. 
 

By making this signing, it does not replace ‘like for like’ or at least as close as you can get, which means the head coach has to adapt the playing style. 
 

Link to comment
Share on other sites

56 minutes ago, Harry said:

I’ll give an example of that :

You are going to sell Bobby Reid. Your head coach views him as pivotal to the system he currently wants the team to play. Bundles of energy, ability to lead the press (and the intelligence to know when), whilst on the ball being able to have composure and awareness once the ball is won to create intelligent passing angles or retain the ball for support/counter attack. 
 

You need the data analytics team to come up with a potential replacement. 
So they scour the data and think “mmm....Bobby does the most running in the team, and can play upfront or in the AM role”. 
Let’s look at Andy Weimann. 
 

The analysts would likely have compared the running stats, seen that Weimann can run as much as Bobby as well as chip in with a few goals and assists. 
 

What they wouldn’t have looked at was how he fits the existing team structure and the way the coach wants to play. Running around is all well and good, but Bobby did it with intelligence and ability whereas Weimann does it just because he has the energy, whilst his ball retention and awareness is a long way off Bobby. 
 

By making this signing, it does not replace ‘like for like’ or at least as close as you can get, which means the head coach has to adapt the playing style. 
 

Good example.  The problem for exacerbated by not looking critically at what worked best for  Reid too....Paterson, good technique, drifting into pockets etc.

Reid didn’t work as well for the team alongside Diedhiou, even if he still looked good individually.  The press broke down because Fam wasn’t in sync with Bobby’s triggers.

So not only were we trying to find Bobby mark 2, we were second guessing how the new signing might play with Fam.

In fairness Fam and Weimann looks a decent fit on paper.  But this is where the data experts should be earning their corn, as you say, not just comparing player for player, but system fit too.

Recruitment is tough, but any data analyst can extract data....it’s knowing the right data and why it’s the right data for the purpose you’ve extracted it that’s key.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...