How a simple checklist could save clubs millions

Football is a strange industry.

It is often said in the analytics world that the single most valuable person in running a football club would be a logical outsider to just look at what was being suggested and have the power to say "No".

No, you can't spend £30m on that 27 year old player from the Russian or Turkish league because he has had a good 3 months of form.

No, you shouldn't give your manager a new 5 year £20m deal because you've had an unsustainable amount of luck in the last 6 weeks.

In short someone to just act like a board should in a normal company. No irrational exuberance, nor irrational despair.

Some of the crazy decisions (Everton 16-18 being my personal, painful example) made in the industry lead us outsiders to look at it and arrogantly assume we'd be that rational person making the correct decisions all the time.

Is this classic Dunning-Kruger behaviour of illusionary superiority?

Maybe but the thing I like about football analytics is that, even if you disagree with the interpretations people make there is usually some thought that has been put into it. 


Managers and their impact on player stats

As a relative newcomer to this field one thing I may have missed out on is the impact of the managers preferred style of play on the statistical output of his players.

In other words do some players look better, and some worse, not because of their ability but because of the role they are being asked to play?

Being logical people you'll probably agree, but with caveats, as logical people you also have caveats. In a previous article I gave an example of Baines for Everton versus Baines for England. When denied a covering midfielder and taken off the set pieces the best attacking full back in the league looked like a poor defender.

So if we are looking at what a left backs role in our preferred system will be we will also have an idea of the type of statistical output we are expecting from the player. The xStats if you like.

Should a manager be trusted with tens of millions if they can't explain what they want to see from their team?

If the system is working as intended then our left back should be making these types of passes, delivering crosses to this type of area, be positioned here when defending the back post. These are all coachable things. It may be that you have a player capable of exceeding your expectations, they may be able to attack more often, recover position quicker. All great, but if you expect your right back to be delivering crosses and instead they are passing inside then you have an issue (and it may not be the fault of your rightback). You could of course pick this up with your eyes, and videos, but having it as data will pick up any biases. Your prodigal son, crowd favourite right back may be just as guilty as your boo-boy left back. Data helps make fairer decisions. 

So if I were that sound minded person who could say "no" to big decisions at  a club what would be on my checklist if the board came to me with a yes/no for a new signing?

Checklist to save tens of millions of pounds.

What is the style of play you want to play?
What are your expectations in data output from a player in that position?
Explain how the potential signing fits in with that style?
Who are they replacing and why is this player better suited?
What happens to the players this new signing will move further from first team action?
Do we have players already in the squad who have the skills to complement them or do we need additional new signings?
What alternative players were considered?
What is the cost of the whole deal and is it good value for money?

An average PL signing now costs £10.3m in transfer fees and £40k+ a week in wages. If a manager wants "backing" in the market at these sorts of levels he needs to have a plan for the player, and the squad as a whole.

Getting transfers wrong can cost tens of millions.

Even basic due diligence would have the potential to pay for itself many times over.

So why does this yes/no person have to be an outsider? 

Football seems to operate on a basis of relationships. This isn't always a bad thing but it always amazes me how many big name coaches end up hiring the same team they've always had wherever they go. As if whoever happened to be the goalkeeping coach and physical trainer at the second division Portugese club they got their first job at are the best in the world at that job.

This "trusted lieutenant" syndrome sees players signed on recommendations without logic being applied. As we agreed earlier, the system maketh the player.

As soon as you appoint an insider they form relationships. It is much harder to say "No" when you have developed trust.

Just applying the basic logic of the above checklist could be an important step into modernising a club.