Analytics makes us more human

The Merseyside Edition

So a few weeks back, I made the case for analytics as a vehicle to humanize football, or at least make it more humane. There are some interesting examples in the post-weekend news dump this morning—one negative, one positive—so I thought I’d take a look at each.

Everton’s not so Kean

The Athletic has a story on the Moise Kean debacle this past weekend, after the 19-year-old former Juventus prospect was ignominiously subbed off after being subbed on for a total of 18 minutes in the second half, a dreaded outcome for any player but horrible for a striker.

It’s of course never easy for a player to leave the biggest and most successful club in Italy to play for Merseyside’s relegation-battling also-ran, and Kean’s numbers have dropped off since he arrived. And the author treats us to I guess what we’re meant to read as a scathing assessment of his 18 minutes?

In his brief time on the pitch on Sunday, the 19-year-old won neither of his two aerial duels, made no successful tackles, completed two out of four passes and mustered nine touches. Shortly before being withdrawn, he was seen wandering around the midfield forlornly after ceding possession to an opponent.

I mean, we’re going with an 18-minute sample here. Obviously, Kean wasn’t playing very well, but this isn’t the way to ‘prove it’ one way or another. This is just a bunch of counting stats with zero context on how they compare to 90-minute averages for similar strikers (and even then, these stats wouldn’t be very useful).

We’re also treated to the view of ‘sources’—we don’t know how well connected they are or how many of them there are—who say he isn’t giving his all in training (he has indeed shown up late for a few training sessions).

I’m not just ragging on the author here for all this; he’s just filling in the blanks as best he can. But this is precisely the problem with struggling prospects like Kean, players for whom we have at least some evidence were capable of playing at an elite level.

Knowing why Kean isn’t hitting his Juventus-form, something good analytics work could, in theory, help out with, is critical to ensuring the team can support Kean the best they can. Kean is, of course, not the first player to struggle to acclimatize to the Premier League, particularly after playing well in Italy, so there may be some other ‘softer’ issues Everton isn’t doing well in, like making sure he’s properly socialized, is being checked up on at home, isn’t homesick etc.

But some decent analytics work could make some comparisons between his preferred position (Duncan Ferguson for his part said the sub was purely tactical and defended the player for having to adjust to an unfamiliar position), his style, teammates he might have relied on for production, etc.

Instead, in this article, we get sources trying to hint at the fact they think Kean has an attitude problem and isn’t up to snuff and should, therefore, be shipped on. This may be true! But it would surely be worth Everton doing all they can, in every area, to ensure they support their investment in a promising player beyond running him hard and punishing him for slovenly behaviour. An analytics-based approach, combined with some basic HR work in ensuring young promising players who are living away from home are taken care of and checked on, is a win-win for both club and players alike.

Liverpool’s thinking about the future

Succession-planning in football seemed for a long time to be one of those ideas that consultancies loved to talk about but so few clubs seemed to actually do. Which was, I think, for a lot of fans, infuriating, because lack of these plans might turn a one or two-year transition period into a prolonged slump, and even relegation.

It’s not as if clubs didn’t think long term; the problem was these long-terms plans almost always involved something like “our five-year plan to get into the Premier League/Champions League, or win the PL or whatever.” These were always best-case scenarios, not fully functional contingency plans. Nor did these plans ever seem to span more than one manager. They didn’t take into account replacing ageing squads, or ensuring institutional knowledge could be passed down from manager to manager, DoF to DoF.

Now, Ornstein’s coverage of Klopp’s recent contract extension opens a window onto a team that might finally be thinking of the future.

Among the German’s key motivations for committing until 2024 was a desire to lay the foundations for “Liverpool after Klopp” and that will include a gradual restructure of the team.

Owners FSG are understood to have outlined a vision of the current squad winning the Premier League title this season and next before the rebuild gathers pace, during which time Klopp’s men finishing far lower down the table would be acceptable so long as the ultimate project remains on course.

By then, a lot of Liverpool’s key players will be moving into the latter part of their careers and already there are young talents and new signings being identified as potential successors.

So here you have a team willing to tolerate a protracted period of struggle to ensure follow-through on a difficult future rebuild, and making this decision in the middle of a season where it looks Liverpool have their best chance in years to secure their first league title since 1990.

If ownership has the cojones to stick with this vision, it will provide the kind of security, trust and patience that never seem to get a foothold in the modern game, where people still believe clubs are just a hire away from getting their very own Sir Alex Ferguson—a manager will make all their problems go away and be able to seamlessly adapt to bounce back after a poor season or two.

That trust will have a knock-down effect not just on Klopp’s eventual successor *cough Stevie G cough cough*, but also the young prospects expected to take up the mantle of the current first team. That will be a Liverpool that wouldn’t give up on a 19-year star because he’s struggling to adjust but will have already put in place a series of checks to monitor progress and assist in player development.

Or not, I’m just spit-balling here.

Though succession plans are, of course, not strictly reliant on analytics, the two go in hand, because to sustain a team you have to have meaningful measures or KPIs in place. You have to be able to tell the difference between a team that is suffering some poor luck and a team that has some major structural problems.