United appoint Dominic Jordan as director of data science

Stig

Full Member
Joined
Feb 26, 2019
Messages
1,473
Most exciting news I've heard for ages at the club. Other than Ronald's return.

Brentford's data analyst team used to set the odds for one of the big bookmakers.

I love it.

They will say McT or whomever loses the ball with 55% of his passes, so he is actually more useful to the opposition than us. Don't play him any more.

Martial spends an average of 45 seconds in the opponents penalty box, which is why he only scores one in a blue moon.

Then with facts you can change who you pick and how they should play.

I love it.
 

Ali Dia

Full Member
Joined
May 10, 2013
Messages
14,133
Location
Souness's Super Sub/George Weahs Talented Cousin
Billy Beane knew this in 2002 so why are we looking into this in 2021?
Because for too long we thought backing a manager with an open chequebook was the way to go in the post Fergie world. Our new managers also had their own data teams brought in. Would Fergie have gone for all these extra staff and departments? Would he have needed them? Probably not. When you’re winning leagues with patched together teams your man management and sports psychology is through the roof. He could also spot a winner and a leader a mile off. It takes a long time to build that aura and autonomy though. It’s not as simple as just get another big name manager and every player at the club will run through walls and adapt for him. it’s good to have these departments that will most likely transcend and supplement various managerial appointments. It would be great if we had long term but less well known transfer targets who statistically fit in with what we need instead of making knee jerk decisions. It’s the only way to guarantee excellent value in the long run.
 

el diablorojo

Full Member
Joined
Nov 18, 2010
Messages
841
Supports
Brentford
Just playing catch up with Brentford ;)

For this to be an impactful appointment you need a few things in place including but not limited to:
  • Firstly access to decent and (if possible) unique data sets
  • Secondly an understanding of what that data can tell you and what it can't
  • Thirdly and probably most importantly - the buy in of all involved, no point doing the hard work on data and stats if Ole is just going to dismiss it out of hand or the recruitment team are just going to sneer at any data driven transfer targets.
 

Ali Dia

Full Member
Joined
May 10, 2013
Messages
14,133
Location
Souness's Super Sub/George Weahs Talented Cousin
Just playing catch up with Brentford ;)

For this to be an impactful appointment you need a few things in place including but not limited to:
  • Firstly access to decent and (if possible) unique data sets
  • Secondly an understanding of what that data can tell you and what it can't
  • Thirdly and probably most importantly - the buy in of all involved, no point doing the hard work on data and stats if Ole is just going to dismiss it out of hand or the recruitment team are just going to sneer at any data driven transfer targets.
I saw a pretty good video on your boys again during the week. Huge emphasis on set pieces which I didn’t realise. Like many many extra hours to get the edge there. Liverpool poached your throw in coach? Liverpool are fully embracing it getting any edge they can. You guys are my second club now. I’m not even into Maths at all but it’s fascinating!
 

Cheimoon

Made of cheese
Scout
Joined
Jun 22, 2020
Messages
14,325
Location
Canada
Supports
no-one in particular
I saw a pretty good video on your boys again during the week. Huge emphasis on set pieces which I didn’t realise. Like many many extra hours to get the edge there. Liverpool poached your throw in coach? Liverpool are fully embracing it getting any edge they can. You guys are my second club now. I’m not even into Maths at all but it’s fascinating!
There was a thread about this guy a while back. If I remember correctly, he doesn't work for any club exclusively, he does sessions occasionally for a number of clubs, internationally. (I think he is Danish himself.) Though if memory serves, he did say that he doesn't work for two clubs in the same league, so maybe that's the poaching part you mentioned?
Billy Beane knew this in 2002 so why are we looking into this in 2021?
My thoughts :lol:
From what I've read, there is already a data team. This is just a new director of data. It's unclear to me if he is replacing someone, or if this is a new position to add weight and bodies to the existing team.
 

sullydnl

Ross Kemp's caf ID
Joined
Sep 13, 2012
Messages
34,063
Murtough's description of how we currently use data:

We already make extensive use of data to analyse players’ performance and physical condition, and to scout opponents and recruitment targets".
The new guy's description of how data can be used:

There is so much potential for data science to benefit the club, from assisting with player recruitment, automatically analysing patterns of play right through to using computer vision to extract information from video feeds in real time.
The difference in emphasis seems to be it being further integrated into what's actually happening on the pitch in real time.
 

Teja

Full Member
Joined
Aug 17, 2014
Messages
5,701
Just playing catch up with Brentford ;)

For this to be an impactful appointment you need a few things in place including but not limited to:
  • Firstly access to decent and (if possible) unique data sets
  • Secondly an understanding of what that data can tell you and what it can't
  • Thirdly and probably most importantly - the buy in of all involved, no point doing the hard work on data and stats if Ole is just going to dismiss it out of hand or the recruitment team are just going to sneer at any data driven transfer targets.
Brentford are my second team now as well, hope it goes well for you in the PL.

Unique data sets IMO are unnecessary - there's a lot of research going into understanding football through data, we have pressing models, possession models, expected value add through player action models etc. etc and plenty of high quality sources to get this sort of data from.

The Barca folks (Javi Fernandez @JaviOnData) have done amazing work on possession models e.g., here's how they view a single possession.



Some more cool graphics from his talk for a deeper dive, linked to 22:06 where the model comes together.


Liverpool, City have been on the cutting edge of data science as well.

What we desperately need are folks that can use publicly available models / best practices and apply them on United games.
 

Chesterlestreet

Man of the crowd
Joined
Oct 19, 2012
Messages
19,505
That would make Varane, Sancho and Ronaldo C players, then?
It only applies to non-players, duh.

Everyone knows this. Ole and Fletcher and the other dross can (money, innit) attract A players. (This is indeed very odd - but, again, money innit).

But they can't attract A data dudes (it's impossible). You need A data dudes to attract A data dudes (money innit, doesn't apply here, nobody knows why).

If your tea lady is a solid B...phew, you can forget about attracting anyone close to A to assist her: more like C or D (they'll use inferior brands, etc.).
 

ivaldo

Mediocre Horse Whisperer, s'up wid chew?
Joined
Nov 15, 2012
Messages
28,699
It only applies to non-players, duh.

Everyone knows this. Ole and Fletcher and the other dross can (money, innit) attract A players. (This is indeed very odd - but, again, money innit).

But they can't attract A data dudes (it's impossible). You need A data dudes to attract A data dudes (money innit, doesn't apply here, nobody knows why).

If your tea lady is a solid B...phew, you can forget about attracting anyone close to A to assist her: more like C or D (they'll use inferior brands, etc.).
Thank you for breaking this down for me. I was blind but now I see. I was so naive!
 

Adnan

Talent Spotter
Joined
Oct 5, 2013
Messages
29,862
Location
England
Just playing catch up with Brentford ;)

For this to be an impactful appointment you need a few things in place including but not limited to:
  • Firstly access to decent and (if possible) unique data sets
  • Secondly an understanding of what that data can tell you and what it can't
  • Thirdly and probably most importantly - the buy in of all involved, no point doing the hard work on data and stats if Ole is just going to dismiss it out of hand or the recruitment team are just going to sneer at any data driven transfer targets.
Everything is in place from the understanding of the data sets to having a department with a dedicated team of 5 already working within the division which is a team of 5 led by Mick Court. But I'm not sure there's a buy-in with the manager and I personally believe we won't take advantage of the good work John Murtough has done behind the scenes until the recruitment department and the Head Coach work in unison. And for that we need a coach who is accustomed to working in such a environment.

The likes of Amad Diallo and Hannibal (recruitment department signings) look as if they're gonna be wasting their formative years under a manager who is afraid to look outside his comfort zone IMO.
 

el diablorojo

Full Member
Joined
Nov 18, 2010
Messages
841
Supports
Brentford
Brentford are my second team now as well, hope it goes well for you in the PL.

Unique data sets IMO are unnecessary - there's a lot of research going into understanding football through data, we have pressing models, possession models, expected value add through player action models etc. etc and plenty of high quality sources to get this sort of data from.

The Barca folks (Javi Fernandez @JaviOnData) have done amazing work on possession models e.g., here's how they view a single possession.



Some more cool graphics from his talk for a deeper dive, linked to 22:06 where the model comes together.


Liverpool, City have been on the cutting edge of data science as well.

What we desperately need are folks that can use publicly available models / best practices and apply them on United games.
I think the unique data sets and unique use of the data is where you are going to make the bigger advances against the competition though. It's something our DOFs point to as an advantage we have - access to the Smartodds data that other clubs don't have.

I hope we are here to stay in the PL :)
 

el diablorojo

Full Member
Joined
Nov 18, 2010
Messages
841
Supports
Brentford
Everything is in place from the understanding of the data sets to having a department with a dedicated team of 5 already working within the division which is a team of 5 led by Mick Court. But I'm not sure there's a buy-in with the manager and I personally believe we won't take advantage of the good work John Murtough has done behind the scenes until the recruitment department and the Head Coach work in unison. And for that we need a coach who is accustomed to working in such a environment.

The likes of Amad Diallo and Hannibal (recruitment department signings) look as if they're gonna be wasting their formative years under a manager who is afraid to look outside his comfort zone IMO.
That's my worry about it working at united too tbh you really do need buy in from the whole club if it's really going to have the impact it is designed to. At Brentford Mark Warburton got us into the play offs in our first season back in the Championship however he wasn't fully bought into the way we were going about things and leaked to the media that he wasn't happy/claimed he was getting sacked so his contract wasn't renewed. Not saying Ole should be got rid of but...
 

Adnan

Talent Spotter
Joined
Oct 5, 2013
Messages
29,862
Location
England
That's my worry about it working at united too tbh you really do need buy in from the whole club if it's really going to have the impact it is designed to. At Brentford Mark Warburton got us into the play offs in our first season back in the Championship however he wasn't fully bought into the way we were going about things and leaked to the media that he wasn't happy/claimed he was getting sacked so his contract wasn't renewed. Not saying Ole should be got rid of but...
The same happened at Liverpool with Brendan Rodgers according to Liverpool correspondents. Rodgers didn't buy into the methods of the recruitment department which was led by Michael Edwards. Rodgers was eventually sacked and when Klopp came in, he immediately embraced Edwards and his team and allowed them to identify transfer targets, which has helped the club end 30 years of pain which has culminated in them winning the biggest prizes in the game.

According to reports, one of the main reasons Klopp took the job at Liverpool was after seeing the work that had been done by the likes of Edwards in the recruitment department, which he saw as something that would help him achieve the success he's achieved.
 

Cheimoon

Made of cheese
Scout
Joined
Jun 22, 2020
Messages
14,325
Location
Canada
Supports
no-one in particular
The same happened at Liverpool with Brendan Rodgers according to Liverpool correspondents. Rodgers didn't buy into the methods of the recruitment department which was led by Michael Edwards. Rodgers was eventually sacked and when Klopp came in, he immediately embraced Edwards and his team and allowed them to identify transfer targets, which has helped the club end 30 years of pain which has culminated in them winning the biggest prizes in the game.

According to reports, one of the main reasons Klopp took the job at Liverpool was after seeing the work that had been done by the likes of Edwards in the recruitment department, which he saw as something that would help him achieve the success he's achieved.
I also read an interview (I think with Graham) that described a conversation between Graham and Klopp, in which Klopp was impressed with how well Graham understood how Dortmund's poor performance in Klopp's final season there had more to do with an unlikely combination of bad luck than poor performance. Klopp was all like 'see, you saw it too when watching our matches!' But Graham hadn't seen a minute from it, he derived it all from the data.

I might be mixing up details here, but it was something like that.
 

Adnan

Talent Spotter
Joined
Oct 5, 2013
Messages
29,862
Location
England
I also read an interview (I think with Graham) that described a conversation between Graham and Klopp, in which Klopp was impressed with how well Graham understood how Dortmund's poor performance in Klopp's final season there had more to do with an unlikely combination of bad luck than poor performance. Klopp was all like 'see, you saw it too when watching our matches!' But Graham hadn't seen a minute from it, he derived it all from the data.

I might be mixing up details here, but it was something like that.
I'm not sure how it worked at Dortmund when it comes to their structure. But at United and Liverpool, it seems both clubs did the scouting, followed by collating the gathered information which then turned into a list of names which was then passed on to the guys working in the analytics departments at both clubs respectively.

So at United you have the scouts (region dependent) who would then report to Marcel Bout who would collate the gathered information, run his own rule over it before filtering the process and passing the information and list of names to Mick Court and his team who work in the analytics department, who do the number crunching .

It's also important to note that Bout has worked with some of the greats when it comes to coaching, in Jupp Heynckes and LVG who both had him as their match analyst, so his understanding of the actual intricacies of the game is at the elite level.
 
Last edited:

Bilbo

TeaBaggins
Joined
Sep 27, 2004
Messages
14,208
While I agree that it helps to have experience with sports stats, or football stats specifically, Dominic Jordan is not being dropped into a vacuum at United. There is already a team of analysts, supposedly with tools and processes and work flows in place, and Jordan is being brought in to lead that team. For that reason, his job is probably not so much to do the actual analysis, but to optimize the team's performance. He does seem to have a good background for that, having led data analytics teams previously. This is what the relevant BBC article says:

His education and experience also provide the right data/stats knowledge and skills to deeply understand and fit in with what the team is doing. So I wouldn't be worried so much about him not having the sports angle. In any case, whoever was in charge of hiring him at United will have been aware of Jordan's specific background, and will also have considered other candidates, likely including people with more experience in sports specifically - yet still chose him. That can't be by accident.

To be clear, I'm not trying to argue that the appointment will necessarily prove right or be super successful, I just think that considering the context and circumstances a bit more helps to understand why Jordan was appointed (among everyone who theoretically could have been), and to be more optimistic about it.
Thank you. Reason above why the Cafe is still worth reading. Great post amidst a bunch of agenda driven, United negative rubbish
 

devilish

Juventus fan who used to support United
Joined
Sep 5, 2002
Messages
61,525
.

They will say McT or whomever loses the ball with 55% of his passes, so he is actually more useful to the opposition than us. Don't play him any more.
The discussion would probably follow similar lines

Ole: Carricky why is this new guy saying mean things about Scotty? Everyone loves Scotty, he runs alot and he's Sir's favourite player.
Carrick: You don't have to listen to him. Never forget whose at the wheel.
Ole: Thanks Carricky. I am at the wheel. Fantastic.
 

Cheimoon

Made of cheese
Scout
Joined
Jun 22, 2020
Messages
14,325
Location
Canada
Supports
no-one in particular
I'm not sure how it worked at Dortmund when it comes to their structure. But at United and Liverpool, it seems both clubs did the scouting, followed by collating the gathered information which then turned into a list of names which was then passed on to the guys working in the analytics departments at both clubs respectively.

So at United you have the scouts (region dependent) who would then report to Marcel Bout who would collate the gathered information, run his own rule over it before filtering the process and passing the information and list of names to Mick Court and his team who work in the analytics department, who do the number crunching .

It's also important to note that Bout has worked with some of the greats when it comes to coaching, in Jupp Heynckes and LVG who both had him as their match analyst, so his understanding of the actual intricacies of the game is at the elite level.
Sorry if I wasn't clear - I meant that Ian Graham (Liverpool's Director of Research) had checked the data on those Dortmund games, and was therefore able to comment on them in a way that made it seem to Klopp like Graham had been watching those games. I looked it up, and this comes from an NYT article that I once read through a link someone posted in another thread on the forum. Here is the article, and this is the relevant bit (right at the start):
Jürgen Klopp was in his third week as Liverpool’s manager, in November 2015, when the team’s director of research, Ian Graham, arrived at his office carrying computer printouts. Graham wanted to show Klopp, whom he hadn’t yet met, what his work could do. Then he hoped to persuade Klopp to actually use it.

Graham spread out his papers on the table in front of him. He began talking about a game that Borussia Dortmund, the German club that Klopp coached before joining Liverpool, had played the previous season. He noted that Dortmund had numerous chances against the lightly regarded Mainz, a smaller club that would end up finishing in 11th place. Yet Klopp’s team lost, 2-0. Graham was starting to explain what his printouts showed when Klopp’s face lit up. “Ah, you saw that game,” he said. “It was crazy. We killed them. You saw it!”

Graham had not seen the game. But earlier that fall, as Liverpool was deciding who should replace the manager it was about to fire, Graham fed a numerical rendering of every attempted pass, shot and tackle by Dortmund’s players during Klopp’s tenure into a mathematical model he had constructed. Then he evaluated each of Dortmund’s games based on how his calculations assessed the players’ performances that day. The difference was striking. Dortmund had finished seventh during Klopp’s last season at the club, but the model determined that it should have finished second. Graham’s conclusion was that the disappointing season had nothing to do with Klopp, though his reputation had suffered because of it. He just happened to be coaching one of the unluckiest teams in recent history.

In that game against Mainz, the charts showed, Dortmund took 19 shots compared with 10 by its opponent. It controlled play nearly two-thirds of the time. It advanced the ball into the offensive zone a total of 85 times, allowing Mainz to do the same just 55 times. It worked the ball into Mainz’s penalty area on an impressive 36 occasions; Mainz managed only 17. But Dortmund lost because of two fluky errors. In the 70th minute, Dortmund missed a penalty shot. Four minutes later, it mistakenly scored in its own goal. Dortmund had played a better game than Mainz by almost any measure — except the score.
I brought this up to complement your comment on Klopp appreciating Edwards' work at Liverpool, as this conversation will Graham will have been another thing that helped Klopp see the value of the work of Graham and his team. I am sure Ole will have had similar conversations at United.

Btw, I recommend reading the full NYT article to anyone curious about the power of data in football, and about the role of a club's data science team. As you were discussing with @el diablorojo, there are great articles on Brentford's work as well.
Thank you. Reason above why the Cafe is still worth reading. Great post amidst a bunch of agenda driven, United negative rubbish
Cheers! (Also @ivaldo.) I must admit I work at a data organization (although I'm not statistician myself), so obviously I am biased towards the subject. ;)
 

micmac

Full Member
Joined
Oct 31, 2012
Messages
172
Location
Trinidad
First we wanted to do it the Bayern Way and keep everything in house….then it was the Barca way now we trying the Brentford way!
 

Adnan

Talent Spotter
Joined
Oct 5, 2013
Messages
29,862
Location
England
Sorry if I wasn't clear - I meant that Ian Graham (Liverpool's Director of Research) had checked the data on those Dortmund games, and was therefore able to comment on them in a way that made it seem to Klopp like Graham had been watching those games. I looked it up, and this comes from an NYT article that I once read through a link someone posted in another thread on the forum. Here is the article, and this is the relevant bit (right at the start):

I brought this up to complement your comment on Klopp appreciating Edwards' work at Liverpool, as this conversation will Graham will have been another thing that helped Klopp see the value of the work of Graham and his team. I am sure Ole will have had similar conversations at United.

Btw, I recommend reading the full NYT article to anyone curious about the power of data in football, and about the role of a club's data science team. As you were discussing with @el diablorojo, there are great articles on Brentford's work as well.

Cheers! (Also @ivaldo.) I must admit I work at a data organization (although I'm not statistician myself), so obviously I am biased towards the subject. ;)
My bad, I just got out of bed not long ago due to working nights. And I've read your post again after having a cup of Coffee and it makes sense.
 

Jibbs

New Member
Joined
Apr 23, 2013
Messages
2,238
Steve Jobs, the legendary Apple founder and CEO had a favorite saying that defined his hiring philosophy, for decades: “A players attract A players.B players attract C players.”

(By "player" he didn't mean a footballer, he meant any member of any organization)

When your club is run by Ed Woodward, Murtough, and Fletcher- all obviously B players at best- you should expect that they will only be able to attract people who are "C". You need leaders who are A players, not just for their own quality, but also so that they bring other excellent people.


p.s. The reason B players don't attract A players is not just that A players don't want to work for or with B players, but also: B players do not like to hire A players even if they can - they are intimidated by working with better people, whether that is conscious or subconscious defense mechanism, it is something that is very real
summed up perfectly. But this appointment seems promising to me, irrespective of Ole stays or goes (which he should)
 

Blood Mage

Full Member
Joined
Aug 25, 2019
Messages
5,822
If Ole gets cold hard data that proves McFred doesn't work and VDB should be playing until we sign a better alternative that's good.
 

giggs-beckham

Clueless
Joined
Sep 9, 2007
Messages
6,926
He might get cold hard data that proves he's the number one issue with our teams performances.