Echo News Southend United boss Kevin Maher on his search for new signings

[KiwiRob]

It wouldn’t shock me if we were keeping an eye on George Alexander at Eastbourne.

Or 21 year old Charlie Caton at Chester...

 

[Medway Blue]

I don’t believe all our signings will be “data driven” as such.

It’s another tool in the toolbox to assist. The data will be related to on the pitch stuff, so the usual goals, assists, clean sheets, how they perform against their XG etc as well as stuff like pre-assists, intensity runs, distance covered, crosses, successful crosses, passes etc etc.

It’s another tool to bring players to the table. There is still a place for John Still to watch a game and come away and say wow what a player, we scout him further, and then decide to go for him.

As far as I understand, we are using this in conjunction with the old school approach, and I believe Kev will have the final say. I might be wrong, but that’s what I’m led to believe.

So, for example, we need a Harry Cardwell replacement. John Still will put forward a case for what he has been out and scene, the data team will put forward a case (I believe people will still go out to watch these “data players”) for their guys, but the ultimate decision will be down to Kev, assuming he fits into budget etc etc.

That’s always been the case that Kev has had the final say. He knocked back some of Stan’s suggestions before, he has knocked back some of John Still’s suggestions and others that have been put forward by agents and so forth.

Be interesting to see how this all develops. Brighton are an excellent example, along with Brentford, with this “moneyball” type recruitment.

BTW, if you’ve not read the book or seen the film “Moneyball” I recommend it.

I think that probably about sums it up.

I don't think COSU have any intention of choosing players, or of signing anybody Maher doesn't want. He'll always have the final say.

There is a drive for data to start being considered alongside scouting and video analysis, though, but I gather it's to support/supplement and not to replace. Going forward, I get the impression that the idea is that neither gets used without the other.

If they get alerted to a player whose data is excellent and suggests he'll fit into the football dept's style, but then watching him in person suggests that data lies, then I guess he won't be signed.

If a player has been seen in person first but data doesn't fully support the notion that he'd suit our style, then maybe the football department might have a bit more convincing to do?

I don't know how it'll work exactly, just throwing thoughts out there.

I guess there's a huge acknowledgement that last summer badly hindered our transfer business, but perhaps it's felt that considering data too might allow the building of a more robust database of players who would suit our identity, so that there are more options and so that losing out on some targets like we did this season doesn't almost completely collapse our transfer business.

 

[BriggBlue]

I hope this new approach isn't the first time we've used data, Football managers data has been good enough for a while that some major clubs use it to scout players.

Whilst it is easy to dismiss it as a game, they have a live database of every single player going down to step 2.

The data is available, as always the challenge is using and presenting the data to help shape a decision and that's where the consortium should be showing their expertise

 

[Mad Cyril]

Have any teams moneyballed themselves out of The National League in the past few seasons?

 

[IloveShrimp]

Have any teams moneyballed themselves out of The National League in the past few seasons?

Wrexham springs to mind.

 

[Mad Cyril]

Wrexham springs to mind.

Disagree

 

[Holy Joe]

Wrexham springs to mind.

Moneyball is about using data to find unliked players that can be picked up cheaply relative to what their stats suggest.

Almost the opposite of the Wrexham model which was to attract players way too good for the level by overpaying them for the level they would be playing at.

 

[Mad Cyril]

Moneyball is about using data to find unliked players that can be picked up cheaply relative to what their stats suggest.

Almost the opposite of the Wrexham model which was to attract players way too good for the level by overpaying them for the level they would be playing at.

Yep. Wrexham were the Man City of the National League. They aren’t the first team to get out of the league because they had a wideboy/sugar daddy owner.

 

[Chris Powell's mate]

Wrexham springs to mind.

Moneyball, named after the movie with Jonah Hill as the analyst, was tried at Fulham with an analytics chap called Craig Kline.
Slavisa Jokanovic won't mind me telling you he was apoplectic when targets got turned down because of a decimal point or two!
When things came to a head, Kline had to be escorted from the FFC training ground with a call to police next on the agenda.

 

[PIERS]

Moneyball v Moneybags

 

[kentblues]

Moneyball is about using data to find unliked players that can be picked up cheaply relative to what their stats suggest.

Almost the opposite of the Wrexham model which was to attract players way too good for the level by overpaying them for the level they would be playing at.

Yes - problem is when other clubs also use the same approach the data serves to make the unliked liked….regrettably….the original moneyball worked so well because no-one else was using it. That all said data as part of the overall analysis is no bad thing, but should not stop signings of players that the management team really want based on their combined decades of experience. The data can point to potential questions to be answered.

 

[Holy Joe]

Yes - problem is when other clubs also use the same approach the data serves to make the unliked liked….regrettably….the original moneyball worked so well because no-one else was using it. That all said data as part of the overall analysis is no bad thing, but should not stop signings of players that the management team really want based on their combined decades of experience. The data can point to potential questions to be answered.

Yeah I've speculated prior on if the plan is to create data on players I lower leagues than us that presumably doesn't exist in much detail (at least below NLS/N)?

Alternatively doing something with data that does exist than no one else does. If you're not doing either of those then there is no competitive advantage to be gained.

Unless of course our current/prior use of data was so poor that we were at a disadvantage and we're looking to catch up?

 

[Liam AAS]

Moneyball/data driven recruitment works in baseball because every single action is a set-piece with players performing a single action and so the data is enormous, accurate and valuable.

You cannot apply that approach to an open-play team game otherwise you end up spending £35m on Andy Carroll because he won 85% of his aerial duals.