Subscribe via RSS Feed

Managerial Responsibility

[ 14 ] August 9, 2010 |

Or, what can Don Wakamatsu and Martin O’Neill possibly have in common, besides either newfound unemployment or having managed teams that I give a damn about (the Mariners and Celtic, respectively)?

Aston Villa manager O’Neill shockingly resigned with immediate effect only days before the season is to commence.  Rumor has it that transfer policy this off-season sent him over the edge.  Specifically, it looked as though Villa were about to lose two of their top players, James Milner (late of the England World Cup debacle) and Ashley Young, while O’Neill was not allowed to re-invest 100% of the proceeds from the sale, nor did ownership sanction the contract demands of Stephen Ireland, coming in part trade from Manchester City in the proposed / rumored Milner move to Manchester.  O’Neill, perhaps correctly, interpreted this as a surrender of ambition, and walked away.  O’Neill is highly regarded, and considered a hero by Celtic faithful.

His timing is crap for not only Villa, but also his own; walk out a couple weeks ago, and the Liverpool job is his.

Less surprisingly, the Seattle Mariners just fired their manager, Don Wakamatsu, and several members of his coaching staff.

Wakamatsu did not deserve this.  Last year, he was regarded for his brilliance, if not for every tactical decision he made on the field, for his ability to actually manage the cast of highly paid athletes / egos under his supervision.  This year, a 42-70 could have had the effect of attenuating the perceived brilliance, and his correctly showing Ken Griffey Jr. the exit door to his career made him close to universally unpopular.  These are superficial, anecdotal pieces of evidence; the sabermetric literature (that I am familiar with, I am now a couple years behind I’m afraid, although there is some interesting stuff here) has had a difficult time establishing that the field manager of a ball club has much measurable effect at all, and is negligible at best.

If baseball managers do not have any (as of yet) measurable effect on the probability of team success, is the same true for soccer managers?  Typically, soccer managers have a dual role from an American perspective: GM and field manager.  Player acquisition / disposal, the starting lineups, and on field tactics are wholly under his (or her) control.  This is not so in baseball, but also not the entire point.  While analysis on this question is highly limited, my non-rigorous, unsystematic hunch informed by anecdotal evidence perhaps hobbled by some subconscious selection bias tells me that the manager has a measurable effect on the probability of success.  Note, I’m not suggesting that the manager is the sole determinant of success, but that it is measurable.  (That study does not quite get at my question, but it’s the most rigorous I’m aware of).

I suspect that a “name” baseball manager would have also walked if presented with the situation O’Neill faced: the classic ‘fire sale’ followed by a clear lack of ambition, because his reputation is on the line.  However, the difference in the two cases is that the reputation of O’Neil is deserved, while the reputation of Wakamatsu, be it his brilliant 2009 or his miserable 2010, is not.

Comments (14)

Trackback URL | Comments RSS Feed

  1. BillCinSD says:

    The first question is what do you mean by measurable and success. Personally, I doubt one would ever be able to separate the contribution of the manager, from the players, and from the assistants. Success with different teams and sets of players (Brian Clough, Alex Ferguson, Joe McCarthy) probably means more than success in one place or type of situation. But then that doesn’t mean it’s a measurable difference

  2. Davis X. Machina says:

    I spammed that paper all over North Stand Chat when Brighton Hove Albion was doing their managerial revolving door thing the last couple years. IIRC it has harsh things to say about mid-season sackings, which were happening annually, or more. As each new Seagulls savior was wheeled in, including a rehire of a previous manager, and the team didn’t budge in the table, people still acted surprised.

    If baseball people as a group were slow to embrace sabermetrics, they look like the geek squad compared to footie fans.

    • BillCinSD says:

      goals were good enough for Tinsley Lindley and Fatty Foulke, why should I care about completed passes or assists or tackles. Those newfangled stats are ruining the game

  3. howard says:

    for the case that managers matter, i give you jose mourinho.

    more broadly, i don’t honestly know how you measure managerial performance as such (i like the general principle of at least looking at deviations from the predicted record your runs scored/allowed would suggest), but we are talking about real, live human beings, not strat-o-matic cards.

    and when you’re talking about real, live human beings, person management – reaching and motivating and utilizing correctly your personnel – of course has some impact.

    now, of course, if we could set up a few parallel universes and try out different managers with the same personnel against the same opposition on the same schedule in the same circumstances, then we could learn something, but until then, this is one area where i’m going to trust my instincts that there are horses for courses: that some managers are better equipped for the given task at hand (coaxing cynical veterans to give their all; bringing along young talent; motivating well-paid 20-somethings to think team first; whatever it may be) than others.

    martin o’neill, for instance, has struck me over a longer period of time as a very good manager in terms of extracting the maximum the talent he had available could produce. i have no clue about wakamatsu: certainly it’s hard to manage how you do much better with an offense as bad as the mariners’ is.

    • Thlayli says:

      Another example would be Pep Guardiola. He took over a team that finished third in 2008, sold Ronaldinho, didn’t buy anyone of consequence … and won the treble.

      • howard says:

        and how did he do it? by influencing the things managers can influence: shape and tactics.

        there are a lot of reasons why messi with barcleona looks like one of the greatest players in world soccer history and messi with argentina looks like a nicely talented player, but one is that guardiola utilizes a system that allows messi to be his best.

  4. Henry Holland says:

    the manager has a measurable effect on the probability of success.

    I present for your perusal one Gene Mauch.

    1964: Decides to go to two-man rotation down the stretch, which meant that the two men in question, Bunning and Short, were gassed during two crucial series. Still one of the most spectacular collapses in American sports history.

    1982: The Angels go up 2-0 in Anaheim, then promptly lose the next 2 in Milwaukee. Up 3-2 in the 7th of the deciding game 5, Luis Sanchez is laboring with two runners in scoring position and a lefty hitter, Cecil Cooper coming up. Does Genius Gene replace the gassed Sanchez with fresh lefty Andy Hassler? Hahahaha, don’t be stupid, of course he doesn’t, Cooper singles in the two runners. Bonus Genius Gene: He brings in Hassler *AFTER* Cooper singles in what turns out to be the winning runs. My intense dislike of Gene Mauch begins.

    1986: *SIGH* One strike. That’s all that stood between the Angels and a cathartic World Series appearance. Mike Witt was a stud that was a big reason why they were there in the first place. But no! Genius Gene perhaps remembers 1982 and pulled him with 2 outs in the 9th and the Angels ahead 6-5. Lefty Gary Lucas plunks Rich Gedman and Genius Gene brings in Donnie Moore (RIP), who –wait for it– he had overused and was pitching with a sore elbow. Dave Henderson hi…..aw, fuck it, we all know what happens next.

    What a lot of people forget is that the Angels rallied to tie the score in the bottom of the 9th and had bases loaded with one out, Doug DeCinces at the plate. A simple fly ball and the Angels are in. But no! he swings at the first pitch, infield fly rule, the Sox…aw fuck it, they lost in 11 and got smoked in games 6 & 7 in Boston.

    So, yes, managers *do* matter, no matter what statheads say.

    Note: I work in a construction office and one day I get a call from….Doug DeCinces. I came *this* close to blurting out “You CUNT! Why did you swing at the first pitch with the bases loaded?!?!”. Oh how I wish I had……

    At least it was a small measure of comfort to see the Red Sucks lose in the World Series in part because of an ex-Dodgers player.

    • Dave Brockington says:

      Instead of “So, yes, managers *do* matter, no matter what statheads say.”, I would suggest that managers *can* matter, in rare, anecdotal situations.

      • JRoth says:

        So managers don’t matter, so long as you exclude situations in which they do matter? I suppose that’s one way to be right in every situation.

        I think that the lesson of DIPs (and McCracken’s original overreach) should probably be applied a bit more broadly than it is. While outliers may not change the overall distribution very much, when making personnel decisions one is not looking at the distribution: one is looking at a specific player, who may well be one of the outliers. Is Charlie Morton a guy facing a lot of bad luck, or is he a guy who doesn’t belong in the major leagues? Either way, he’s a statistical blip, but for the Pirates, it will make a significant difference.

  5. elm says:

    Saying that studies find no measurable effect is not the same as saying there is no measurable effect. In this case, we don’t know how to measure managerial effectiveness, so of course we can’t find an effect we can measure.

    Contrast this with clutch hitting, where we also don’t find a measurable, consistent effect, but where we have some good ideas of how to measure “clutch.” Thus, when study after study finds that clutch hitting doesn’t exist as a repeatable skill, I tend to believe it.

    • Dave Brockington says:

      Yeah, I get that, which is why I wrote the following: “If baseball managers do not have any (as of yet) measurable effect on the probability of team success”. The key here is “as of yet”. It’s a basic social science rule: the lack of an observed relationship does not mean that it’s not there.

      • elm says:

        Oh, I wasn’t really disagreeing with you, more making a general point inspired by both your post and Henry’s characterization above mine.

        And my point was that sometimes I’m more willing to accept the null hypothesis than others: that we can’t find clutch hitters suggests to me that clutch hitters don’t exist but that we can’t find a managerial effect does not suggest to me that managers don’t have an effect.

        • howard says:

          baseball managers can make the strategically correct decision and still have it go wrong and make the strategically incorrect decision and have it go right, which is part of the difficulty of measuring their impact.

          the other part is that, as i suggested earlier, when it comes to dealing with the actual individuals who make up the team, how management treats workers matters just as much in baseball as everywhere else in life.

          i mean, does anyone deny that some people make better bosses than others? and that makes a difference to productivity and output?

  6. [...] in baseball having a measurable, systematic effect on the probability of success of their team (something I posted about a few years ago), apparently the owners of Blackburn believe that its of paramount [...]

Leave a Reply

You must be logged in to post a comment.

  • Switch to our mobile site