Amidst all the hoopla and hilarity of the season's first managerial sacking in the Premiership - Steve Bruce by Sunderland FC - it's easy to lose sight of the fact that it typically doesn't help to sack the manager. To be sure: there often seems to be a bounce-back in results for teams that appoint a new manager; but statistically speaking, the bounce-back is likely to occur in any case, with or without sacking the gaffer.
But never mind the statistical evidence. Sunderland, like the many football clubs before them, do seem to go about hiring and firing without worrying too much about what the numbers actually tell them. To be fair: aside from success on the pitch, there can be good reasons to terminate the manager. As the Guardian's Barney Ronay points out, it's often good politics and a way to manage supporters' discontent. The manager is a well paid patsy, and owners and boards need the buffer of a manager to protect themselves from criticism by the fans and media while being seen as "doing something" to improve a football club's fortunes. Or football clubs may see something in the coach that's just not working (and about which we, as outside observers, may have no information) and that needs fixing. Or, despite goodwill on all sides, the football club may in fact have enough data to make the judgment that it is time to panic - as Sunderland have evidently done.
All well and good.
But the interesting question is not "Why do they fire the manager?" Instead it is this: "What should a football club do next?" Or more specifically, what should Sunderland, this year's case in point, do next, now that they have appointed Martin O'Neill?
My advice would be this: do nothing. You heard me right. No transfers in the January window; no expensive short-term fixes (aside from the £6 million you are shelling out for O'Neill); no need to hire lots of additional coaching staff or paint the canteen a nice shade of red and white. At this point, all the doing that needs done by the owner is done. Simply let Martin O'Neill get on with the job and see what happens.
Sunderland fans may disagree. Clearly, the team is not performing, so surely it needs to be fixed by buying in a few new faces in January. And, as you might expect, there is in fact talk of a £15-20 mio transfer budget owner Ellis Short has allegedly promised O'Neill. How can you not see the logic in that?
There are two main reasons I think the smart thing Ellis Short can do at this point is to do nothing.
First, doing nothing is much cheaper than doing anything else; second, and more importantly, it will allow him to properly diagnose the problem Sunderland have at the moment.
First, doing nothing is much cheaper than doing anything else; second, and more importantly, it will allow him to properly diagnose the problem Sunderland have at the moment.
To properly pinpoint a team's weaknesses and deal with them, you need data about your manager and your squad. Good, reliable performance data, and more is typically better than less. A good starting point is what Sunderland should be achieving to be considered a success. Given the high correlation between transfer spending and on-pitch success, we have a rough way of putting a number on Sunderland's performance potential. Their squad is currently valued at £95 mio. This is the equivalent of a mid-table club, putting them in the 10th spot out of 20 Premier League clubs, so right around where Everton sit right now. Instead of 10th, Sunderland are currently 17th with 11 points, compared to Everton's 16. In terms of standing and points, Sunderland are underperforming by 7 places in the table and at least 5 points (perhaps a little more).
So even if O'Neill spends the entire £ 20 million in January, Sunderland's relative position in the football club budget table won't change (other than breaking the tie with Fulham); this means that even an expenditure of that magnitude would only minimally lower the probability of relegation. Still, of course, as is typical for most football clubs and supporters, that marginally lowered probability is really focal and visible and thus important, while the knowledge revealed by unconfounded data is seemingly ignorable and ethereal.
So even if O'Neill spends the entire £ 20 million in January, Sunderland's relative position in the football club budget table won't change (other than breaking the tie with Fulham); this means that even an expenditure of that magnitude would only minimally lower the probability of relegation. Still, of course, as is typical for most football clubs and supporters, that marginally lowered probability is really focal and visible and thus important, while the knowledge revealed by unconfounded data is seemingly ignorable and ethereal.
Let's turn to the second point then: how to diagnose Sunderland's performance. We can surmise that the club's underperformance is either due to a sub-par manager or a sub-par squad. As evidenced by his decision to fire Bruce and hire O'Neill, the owner has bet it's the former. But we want to know for sure; we also want to make sure it's not also the latter. From an analytical perspective, bringing in a new manager has one real upside in figuring out which it is. By changing the manager, players no longer have an easy alibi for underperforming ("it's the gaffer, not me"). The bad news is that winning a few matches between now and New Year's won't prove that the old manager was really the problem in the first place.
To see why that's the case, it's important to remember that teams' fortunes are similar to the business cycle - there is variation around the overall performance level the team is capable of, and times of playing well are followed by matches that aren't going so well (think of Chelsea last year experiencing a boom early in the season and later on experiencing a recession, so to speak). And since managers are typically fired during a recession, the tendency back toward a team's mean, or "real" level of performance can easily fall into the first few weeks of a new manager's tenure. But of course, this may have nothing to do with the manager's work, and everything with good timing.
So, to evaluate Sunderland's performance, we need at least a couple of months to see how good a new manager really is. New managers do well for a little while, so we need more data on Martin O'Neill. Once we know how good O'Neill is, given the squad he inherited, we still can't rule out the possibility that it's really a sub-par squad that's responsible for the football club's fortunes. For all that, we need more data than we currently have. And there is just not enough time between now and January to figure it out.
Assuming that Sunderland's new manager Martin O'Neill is a good coach - and given his track record, there is little reason to assume he is not - then it would be a mistake to go out and buy players in January. No need to cough up £20 million. Instead, I would let O'Neill do his job, earn his £2 mio. this year and go through the rest of the season to evaluate what does and does not need to be fixed in the squad. This will give you two critical pieces of information: performance under Steve Bruce and performance of the same squad under Martin O'Neill.
Among other things, this will answer these questions:
- How good a coach is Martin O'Neill really?
- How good a squad do Sunderland have?
- What do they need help with?
Put the other way around, going out and buying players now doesn't just add to the cost of firing Steve Bruce. It also makes it virtually impossible to figure out what got Sunderland in trouble in in the first place or whether the money Sunderland are spending on O'Neill is money well spent. If you were Ellis Short, wouldn't you want to know?
