This is going to be like choosing AWB from 800 right backs.
United appoint Dominic Jordan as director of data science
- Thread starter Tom Van Persie
- Start date
Haha, no worries!
Can he play in a midfield three?
Ok this seems suspiciously like progress ?
Just wondered how you know this for a fact, are you something to do with the club?
We could have got a cheaper data scientist from Europe.
Pointless appointment. Get a top class coach and take it from there.
You'd have had to be blind not to see how many chances Dortmund were squandering in Klopps last season.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.
It doesn't have to be either/or.
The analysis wasn't restricted to chances alone. But anyway.
Data science is often used in the advertising industry. Are we certain that this appointment is not mainly for the improvement of the sales department?
Its nice and all but at this moment you can hire Michael Owen or someone from RedCafe to point out obvious stuffs like we need better midfielders and if we sort that out, we will be better. You dont need Deep Neural Networks to know if Fred has a pair of chicken legs that crumble under pressure from opponents.
Yes because this is exactly what we need.
Nevermind actual good football coaches or managers. The scientist will sort the tactics, substitutions and team selections out.
Maybe it takes a scientist to show Ole et al how to rotate a squad successfully or get Fred benched.
I'm calling it now. This changes nothing.
Nevermind actual good football coaches or managers. The scientist will sort the tactics, substitutions and team selections out.
Maybe it takes a scientist to show Ole et al how to rotate a squad successfully or get Fred benched.
I'm calling it now. This changes nothing.
Some people will really take any chance to sh*t on the club, won't they?! Sounds a useful hire to me...I'm failing to see the issue. Yep, we haven't signed a CM but why should that mean that this hire is a bad one?
It’s either sack Ole/bench Fred or nothing!!
Data Scientist - Fred is rubbish.
Ole - Fred starts every Match.
Ole - Fred starts every Match.
Not pointless at all, the top clubs all had data analyst specialists except for us. It's good that we're modernising at last, but yes the next step is upgrading the coaching setup.
Concerning we didn't have one tbh.
The bottom line is this (along with the changes in our structure, scouting processes and recruitment processes) can only be positive steps. Though it does underline just how antiquated a lot of how the club was run was even a couple of years ago given all those areas of the club had to be revamped.
It doesn't change our managerial issues but not everything about the club has to be centred on the manager or see the manager dragged into the conversation. It can be a positive in its own right without all the braindead "you don't need data to know Fred is bad" comments.
Though if you did want us to move on from Solskjaer or in particular take issue with the way we're coached under him compared to the way some of our rivals like Klopp, Pep and Tuchel operate then this is still the sort of modernising change you want to see taking place in the background. Because a more data driven, analytical approach directly influences the tactics those coaches employ and their ability to instil them in the team. Even if you don't think it will change what Ole does, it still gives the next manager a better platform to do so.
It doesn't change our managerial issues but not everything about the club has to be centred on the manager or see the manager dragged into the conversation. It can be a positive in its own right without all the braindead "you don't need data to know Fred is bad" comments.
Though if you did want us to move on from Solskjaer or in particular take issue with the way we're coached under him compared to the way some of our rivals like Klopp, Pep and Tuchel operate then this is still the sort of modernising change you want to see taking place in the background. Because a more data driven, analytical approach directly influences the tactics those coaches employ and their ability to instil them in the team. Even if you don't think it will change what Ole does, it still gives the next manager a better platform to do so.
It is not how much data you have, it’s how you use it. How can you think that a tactical overhaul will not benfit from better ouput from the data?We need a tactical overhaul, not more data..
That said, if we suddenly start playing good football and actually creating chances, no doubt he will be credited
The ones that win have top class coaching and a top class fitness department. That's what we need. Fit players who are coaches well. These nerds arent going to help with much. Clubs with these guys still make terrible signings.
One of Liverpool's data scientists going through some of their modelling for pitch control.
You clearly dont understand much about this «data science» do you?
May I suggest you go on youtube and search «deep machine learning» and then go from there..
How long has it been since we've unearthed hidden gem ourselves? Last one I remember was Martial and unfortunately he hasn't become what he first shown he could.
I'd love if we bought an unknown player who becomes fans favourite, hopefully acquisitions like tthis one make it possible soon.
I'd love if we bought an unknown player who becomes fans favourite, hopefully acquisitions like tthis one make it possible soon.
I hear he's a real nice guy.
I'm calling it now, we won't have a clue especially in the short term whether this backroom appointment changes anything good or bad. And it's really not worth another rant about the manager over.
Well. It’s not really about players to buy. It’s more like to figure out how to get an cutting edge when playing. Machine learning can be to football tactics, what the car did to the horse. No point in training horses when you should drive and build cars. In other words maybe we are approching the end of human tactics?
Which brings up a kinda interesting question. What are the limits of computer assisted help the manager is allowed during a game? Does Klopp and Pep run advanced AI-driven real time analytics during matches to help them understand the game correct and do the most effective changes? And in case yes, should it be allowed?
Last edited:
We need a tactical overhaul, not more data..
That said, if we suddenly start playing good football and actually creating chances, no doubt he will be credited
Really? You are actually in a position to know whether what he brings on data is something we need or not?
My thoughts
Baseball's inherently an easier sport to apply statistics to because of the pitcher vs. hitter nature of the sport and because baseball has always fetishized stats to some degree. Also, to be fair, the technology that allows for good movement tracking on the pitch didn't even come about till the early-mid 2010's IIRC. At this point, most high-performing baseball teams are run by quants, which I think we're decades away from seeing in football given its slow adoption and general aversion to statistics-based decisioning.
As it relates to United specifically, being behind the curve with these things is just par for the course really.
Data science is must these days. Glad finally club is taking it seriously.
Need capable hands to implement the suggestions though. Lets see.
Need capable hands to implement the suggestions though. Lets see.
When your eyes can tell you we drastically need to improve our tactical approach and patterns of play I don’t think a data analyst is the hero we need currently but as others have said we are playing catch up with clubs that have these in place already.
To me it’s obvious our coaches aren’t up to it but we know that they won’t be replaced so have to hope that appointments like these improve what we are seeing.
No...behaviour is.
Interesting take. People rating Fletcher as a player v Fletcher as a recruiter.
I
United have employed a range a data scientists for some time now.
Apparently Dom has been brought in as a Director to oversee all activities.
In addition, he is a long time United fan.
Why anyone would see this as a negative move is beyond me.
Wow that's kinda deep use of such technology, I never thought about it that way.
Although it does look inevitable i guess every team will have to do the same to stay competitive, so I'm guessing it will be allowed
Data is amazing, data helps drive decisions, i am not against this appointment.
Where i do worry is more data in the hands of Ole, i think he is suffering from analysis paralysis. In recent press conferences and post match interviews, i hear him say a lot that he has not checked the stats or the stats tell him a reason why he made a decision. It appears to me that we have someone already providing this information, for a club of our size, i would be really surprised if this is a new function, it may just be a new leader to draw freah ideas.
Unfortunately, i dont think Dominic will help Ole become tactically better, make better subs or stop playing McFred. We now live in a world where data is way more accessible, it gives us great tools to understand certain things, many people across the world make decisions based on facts and figures too. However, for it to work it needs to be balanced with the right qualitative information.
I think this signing is completely independent to what Ole does now or in 6 months. An extra chart here and there will not save his job.
Where i do worry is more data in the hands of Ole, i think he is suffering from analysis paralysis. In recent press conferences and post match interviews, i hear him say a lot that he has not checked the stats or the stats tell him a reason why he made a decision. It appears to me that we have someone already providing this information, for a club of our size, i would be really surprised if this is a new function, it may just be a new leader to draw freah ideas.
Unfortunately, i dont think Dominic will help Ole become tactically better, make better subs or stop playing McFred. We now live in a world where data is way more accessible, it gives us great tools to understand certain things, many people across the world make decisions based on facts and figures too. However, for it to work it needs to be balanced with the right qualitative information.
I think this signing is completely independent to what Ole does now or in 6 months. An extra chart here and there will not save his job.
I do hope that this system will come out with good players that don't always cost 50-80m.
Last edited:
Solskjær's Law
As a discussion on Redcafe grows longer, the likelihood of a person making an Ole Out post increases
Yes. It is really interesting. All this football analytics is now done after game, which really is’not that helpful. As in any industry, what you want is actionable insight in real time (while the game is on) so you can gain something from it right there and then. I do think some clubs already do this on a semi-automated level, but this is nothing compared to how it will affect the game in the years to come.
Even do chess is a lot more simple game than football (fixed postions and variables), it is still extremly complex. Super computers that are trained for chess is now a lot better than the world champion — who is probably the greatest chess player of all time.
Now imagine what that will do to football when all clubs get real time decision making by super computers, that is a lot better than lets say Pep Guardiola. A super computer can run 10.000 simulations based on a defined outcome (goal) based on historical data and in-game metrics in seconds. So, the basis for formation changes and subs will change alot. I think it will improve the game, but it might give some clubs an advantage we are not used to
Will this make the players better? Maybe in the sense that they will be utilized better, but noone will run faster, shoot harder or tackle better because of this.
Just on the bold part, an article I read said that Liverpool currently have a seven minute lag on match data and the "live" data is less accurate than it is post-match. With Liverpool being one of the leaders in this regard, it gives some idea where the technical capabilities are atm.
That is really interesting. Wonder why it’s s 7 min .. such a strange delay. With the bandwidth we get with 5G.. I think it’s just a matter of time before every byte is live. I really thought they were a bit more advanced by now, but I'm not sure they are telling the complete truth here.
Are clubs allowed to put their own equipment into the black west most players use now?
Being a norwegian, it comes to mind, back in ‘94 when we held the winter olympics — we had developed a special glider for cross country skiis that gave up to 5% advantage. We won a lot of gold in that era. Even those who make the rules was not aware. It took many years before this «technology» was available for everyone competing. Yes, we are sneaky bastards.
My point is, if a club has some sort of an advantage here, I don't think they would tell
I think the article meant that there was a seven minute delay in computing/applying the analysis to real time events on the pitch. The analysis gets completed after a seven minutes of the event happening on the pitch and has nothing to do with the speed of your internet connection if i understood it right. Might be wrong too.