Data Analysis on Players.

[Strykr]

I get that data is important, but it's not the be all and end all. Take a LWB who the stats show is pinging in crosses in the NLS because his current manager expects him to cross at every opportunity. We sign him up and we won't get the same player because he'll be told to retain possession first and cross only when there's a definite opportunity or no other option. I might have expressed that badly, but you get the general gist - a players stats can change massively depending on how the team he's playing in is set up, and what his manager expects of him.

Also massively dependant on the type of players around him. Most of these one or two season scoring wonders have got lucky and completely clicked with the players around them supplying them.

 

[amsemp]

I’m all for it. I’ve had an interest in the stats based approach for a long time and there’s some really good books out there which give huge insight into how it’s transforming the way people analyse football and the transfer market. The Expected Goals Philosophy and xGenius by James Tippett are both great reads if anyone’s interested.

The stats tell you what the eye can’t. Passing the eye test is important, which is why we’re still watching targets. But it’s not the be all and end all. A scout watching Moncur might be impressed with some intelligent plays, neat touches and body language, but I’ll bet his stats don’t make for good reading.

Stats can and often do identify undervalued players, as well as overvalued players. I don’t see why it can’t be replicated at our level to a certain extent. Dare I say it, we were all pretty underwhelmed at signing Pepple, but look how he turned out. I’m confident the stats based approach will definitely help us identify more Bim Pepple’s in future.

 

[blues exile]

I’m all for it. I’ve had an interest in the stats based approach for a long time and there’s some really good books out there which give huge insight into how it’s transforming the way people analyse football and the transfer market. The Expected Goals Philosophy and xGenius by James Tippett are both great reads if anyone’s interested.

The stats tell you what the eye can’t. Passing the eye test is important, which is why we’re still watching targets. But it’s not the be all and end all. A scout watching Moncur might be impressed with some intelligent plays, neat touches and body language, but I’ll bet his stats don’t make for good reading.

Stats can and often do identify undervalued players, as well as overvalued players. I don’t see why it can’t be replicated at our level to a certain extent. Dare I say it, we were all pretty underwhelmed at signing Pepple, but look how he turned out. I’m confident the stats based approach will definitely help us identify more Bim Pepple’s in future.

As a matter of interest, do we know that data was a factor in signing Pepple or was it down to scouting?

 

[amsemp]

As a matter of interest, do we know that data was a factor in signing Pepple or was it down to scouting?

Bim came before Shivaas, so I wouldn’t have thought so.

 

[Yorkshire Blue]

Possibly, but if it were me I'd think about what value I could add to the process, as a non football expert, I'd suggest I could add very little to the individual decision making of signing player x or player y. Where I could add value, or protect my investment, is by setting criteria/boundaries within which football experts could operate with autonomy

I assume the transfer committee is the body that sets those criteria.

COSU are experienced business men who have run businesses, sat on boards etc. They’ll have dealt with say legal and accounting reports whilst buying Southend without being lawyers (other than TL) and accountants for example. They’ll have asked questions and challenged things of the experts and I expect the transfer committee would work similarly.

Transfers involve more about footballers it’s also about it contracts and strategy. Do you go for a short term loan to plug a gap or take your third choice target on a two year deal? These are as much business decisions as football decisions, especially when COSU are funding them.

Nathan Ralph has the most complete long passes per 90 at 3.4, just ahead of Gubbins at 3.1 and Crowther at 2.8.

Here's a Jack Bridge heatmap for the season if you're interested:

View attachment 35234

And that's just on a free to access, public app.

How does that compare to last year’s?

 

[Holy Joe]

I assume the transfer committee is the body that sets those criteria.

COSU are experienced business men who have run businesses, sat on boards etc. They’ll have dealt with say legal and accounting reports whilst buying Southend without being lawyers (other than TL) and accountants for example. They’ll have asked questions and challenged things of the experts and I expect the transfer committee would work similarly.

Transfers involve more about footballers it’s also about it contracts and strategy. Do you go for a short term loan to plug a gap or take your third choice target on a two year deal? These are as much business decisions as football decisions, especially when COSU are funding them.

Quite agree - which surprises me somewhat if, as others have suggested, they're getting very involved with individual player selection. Other than say TL getting involved with negotiation and legals. Is Gulati for example getting involved with individual player selection or getting involved with the technicals/process of data collection and analysis?

 

[Mad Cyril]

Nathan Ralph has the most complete long passes per 90 at 3.4, just ahead of Gubbins at 3.1 and Crowther at 2.8.

Here's a Jack Bridge heatmap for the season if you're interested:

View attachment 35234

And that's just on a free to access, public app.

This looks like my lawn.

 

[manor15]

How does that compare to last year’s?

Less involved in the box, and slightly less defensively.

 

[Yorkshire Blue]

Quite agree - which surprises me somewhat if, as others have suggested, they're getting very involved with individual player selection. Other than say TL getting involved with negotiation and legals. Is Gulati for example getting involved with individual player selection or getting involved with the technicals/process of data collection and analysis?

I’d guess it’s as much strategy. It’s agreeing what the budget is, do we plug gaps with loans or wait for our targets to become available? What if our fourth target is available but has other teams interested so needs an answer - do we pull the trigger or wait on our long sought after first choice target whose team has maybe just sacked their manager, second choice whose EFL team would want to bring someone in in his position before they’ll let him go, third choice whose contract is up in the summer and is available now but will cost a fee?

 

[BriggBlue]

View attachment 35238
Less involved in the box, and slightly less defensively.

Which app / website are you using to get that info?

 

[manor15]

Which app / website are you using to get that info?

Sofascore app.

 

[londonblue]

We were refused access to Wyscout because Ron didn't pay the bill lol

There's a ****ing surprise. Especially as that is exactly the type of thing he would have wanted.

 

[blues exile]

Bim came before Shivaas, so I wouldn’t have thought so.

That's what I thought. The data available might well have painted an adverse picture and halted the process of signing him, in view of his previously uninspiring stats.

 

[SUFCGoals]

Personally, I think this data thing is being blown out of proportion considerably. I think the data will allow us to make more justified gambles on players (because every player is a gamble). This will allow us to open our pool of players massively and can help us identify players running under the radar and highlight players performing well that we may not have had coverage of beforehand. Not just regionally, nationally but within the home nations and internationally (if we have that sort of pull).

If this data does just that, that'll then prompt John Still / his scouts / the football department to watch the players in person.

Realistically how much data would there be for George Wind when he was at Billericay, as far as I understand there aren't any real data sources below Step 2. So John Still and his army of men are going to continue what they've been doing for the past few years.

Ultimately I think the data will be an aid to human work and will more so be used as an identifying tool.

 

[dloman]

We have always used data and all agreed that it was a good indicator. If we signed a striker who had played 300 games and scored 7 goals there would be a meltdown. This seems to me to be an extension to this (big extension).

On the face of it, this seems like a good thing to do. It will need a good model and a good set of processes (how to integrate with the human scouting, process to get the names on the whiteboard) etc. This would take time to set up which seems to be where we are today.

This is where data could be of benefit. Using the example above, deep diving the data could potentially show that whilst the goals to games ratio as low, the number of team goals scored that the mythical striker was involved in both directly and indirectly could have been 1 per game. It could also show that when he was not playing the team scored 0.2 less goals per game etc.

Very crude way of showing how data can be used and may otherwise have been missed, but you can see some of the potential of it in conjunction with physical scout reports.

 

[PIERS]

I’m sure it has lots of uses, but it’s a bit boring .. heat maps and assists can be used in conjunction with what a manager or scout can see . But as mentioned , how we use a player will be far different to how they have been playing ..
Brighton and Brentford are often mentioned as examples for data driven scouting. But they are both owned by serious wealth , takes the edge of it a bit. Peterborough are a more realistic aspiration, although they too have far more funds

 

[southendkid]

Brighton and Brentford are often mentioned as examples for data driven scouting. But they are both owned by serious wealth , takes the edge of it a bit.

Compared to us are Brentford rich? Yes. Compared to other premier league sides are they? No.

Since getting promoted in 2021 (and excluding clubs that have been relegated) only Everton have spent less money than Brentford in the transfer market.

 

[kentblues]

Time should be okay if you are looking at a player from National S/N, as all games are on Wyscout etc, so you could extract that quite easily.

The problem will be if they're from any lower than that level, where the info isn't available. I'm guessing you'd have to approach the club in question for permission to have a copy of the footage from games, then build your profile based on that, which will take a lot more time. We all know that a camera doesn't show the whole situation also, so there is that to add to the equation.

We have no means of developing players below NLS/NLN. No means of giving them meaningful development of game time. Unless we make them play with the u19s (if that’s allowed)

 

[kentblues]

As a matter of interest, do we know that data was a factor in signing Pepple or was it down to scouting?

I think it was due to what Luton told John Still, ex Luton manager, when he asked about the availability of anyone available who met our urgent needs….

Doubt a computer programme was consulted. And based on album previous record it would likely have said no. Which is why he was available and cost peanuts (or “value” as Kev would refer to it)

 

[PIERS]

Compared to us are Brentford rich? Yes. Compared to other premier league sides are they? No.

Since getting promoted in 2021 (and excluding clubs that have been relegated) only Everton have spent less money than Brentford in the transfer market.

The seasons before they got promoted they were losing 30 and then 50 odd million quid. Data maybe. Hard cash definitely. There was a massive property play as well.. . They will be sold or relegated within three years. If they aren’t sold they will do a Luton