The Ship Be Sinking

Mouth Almighty

Oh No! Who Could Have Forseen This!

Three examples of a larger point today, about injuries and outcomes:

Let’s start with Arsenal. Fans have been saying this entire season that Arsenal have been hobbled badly by injuries, and just wait until all of our hurt players come back- we’ll be a force then. An example from today at Arseblog: “…imagine this formation with Arshavin and Walcott in place of Eboue and Bendtner and with Cesc in the centre of midfield.” I don’t mean to pick on Arseblog here, this is simply the most recent example of this I’ve seen.

I don’t buy it, and never have. The premise is that Arsenal’s injury woes are somehow uncommon or unexpected, a freak occurrence which is unlikely to be repeated and thus has little predictive value for the future performance of the team. There’s two major problems with this line of reasoning. The first is that while this year’s team have had a few freak injuries which were largely unpredictable in that they happened to usually durable players (like Cesc, or Kolo’s malaria), there have also been players like Robin van Persie with very long histories of frequent injury who have been almost totally healthy all year long. The uncommon and possibly unrepeatable outcomes have been on both sides of the ledger, so it’s difficult to say that Arsenal’s luck has been exceptionally bad from the standpoint of reasonable expectations of player health.

The second major problem is the assumption that injuries are usually non-predictive events. Arsenal are, in fact, a team full of fragile players: Diaby is rarely good for more than 2 or 3 games at a stretch, Eduardo, sadly, has suffered minor issues after the majority of games in his comeback after a year away, Rosicky’s problems are well known, RVP is usually injured for substantial stretches, Gallas has suffered his share of pulls and muscle problems and such (and is now over 30), Walcott has natural issues with his shoulders which have put him out more than once for extended periods- the list here is substantial and no doubt you can add to it. In all of these cases, there’s a track record of repeated injuries of a similar character which arise from the normal conduct of play, the sort of things which a player is simply going to have to do in order to perform. A non-predictive injury would be one like Cesc’s, where there’s no history of similar injury, or one like the tackle which put Eduardo out for a year which was exceptionally vicious. When Diaby, for example, misses several weeks with a muscle strain, that’s simply to be expected based on prior performance; if that sort of thing was in fact not repeated next year, THAT would be a freak occurrence because of how far it would depart from reasonable expectations.  Assuming that next year’s Arsenal will have a better track record of health than this year’s is an expression of hope, not reason.

So what, you may ask? That’s our second example. Take this article as an example of what happens when people fail to take into account the overall team-wrecking effects of having star players with major injury histories which are ignored.

(Aside: I just wiki’ed Eric Lindros, whose page as of right now has been edited to read “Eric “Biggest Waste of a Hockey Player Ever” Lindros (born 28 February 1973 in London, Ontario, Canada) is a retired professional ice hockey player.” Lol, as the kids say.)

The article attempts to defend the acquisition of Lindros on the grounds that what was traded for him ended up being no big deal, and that Lindros was ok. This is foolish analysis on the face of things, since Pavel Brendl’s eventual flameout was not known at the time, and therefore his perceived value could have been used to acquire many other players than Lindros; the trade shouldn’t be evaluated just on the basis of what each player did, but on the potential value of what each asset in the deal could have been used for at the time.

That niggle aside, the real problem with Lindros wasn’t the trade, but what Lindros did when he was with the Rangers. Wiki has the stats; you’ll note a good first year, healthy, followed by a healthy second year with a major statistical collapse, and then a third injury ravaged year before he left the team. If you’d been watching regularly (I had season tickets at the time), you’d have seen a player who was playing in a totally different fashion than he had previously in order to shield himself from further injury (Lindros had a history of concussions which ended his career, for those who don’t know), who had given up his physical style of play and was avoiding certain positions in order to minimize his risk.

The problem for the Rangers then was twofold: firstly they had ended up trading for a player who, even when healthy, was not at all the player he had been before his injuries became a constant risk, and the similarities for Arsenal fans to the situations with Eduardo, Risicky, etc. should be obvious; the second was a much more insidious cost, which is usually still not recognized. Read that article again, and focus on this sentence: “Losing puck-mover Kim Johnsson was somewhat more painful, though not the franchise-magnitude sacrifice Bondy would have you believe compared to the potential upside of Big E. “

That’s everything wrong with this mentality in one line. By the time he reached the Rangers, Lindros had suffered seven reported major concussions. SEVEN. Each concussion suffered predisposes the sufferer to a greater risk of another- was it at all reasonable to assume that a player who had suffered 7 major concussions in 8 NHL seasons was unlikely to suffer any more? But because Lindros was big and strong and had a good touch, people continued to fall in love with his “upside”- even the author of this article, 5 years after Lindros washed out of New York due to…a concussion.

Lindros was at all successful with the Rangers because he changed styles to minimize his risk, becoming less effective but more durable; this was entirely predictable based on his history, because the only other option for him was to continue to suffer debilitating injuries which were likely to be career-enders. While he was still somewhat effective, the unjustifiable assumptions (which were not his fault) that any day now he would play just like he did in Philadelphia ended up hamstringing the team for years on end, as they budgeted and made player moves around the idea that big number 88 was going to give them far more than he was physically able to do. The result was three years of failure in which the team finished 11th, 9th, and 13th in the conference, with no playoff appearances. The implications for Arsenal here should again be fairly obvious.

A final example to illustrate probability and outcomes. Take the fighting world, where these distinctions are usually clearest: at UFC 95 recently, Paulo Thiago fought Josh Koscheck. Kosckeck, a striker and wrestler who had previously fought the current champion, was a heavy favorite in that fight and considered a top-5 welterweight in the world; Thiago was considered a less experienced and less well-known UFC debutante reliant on his jiu jitsu. So of course Thiago knocked Koscheck clean out in the first round with a two punch combo.

And yet, anyone who gave the matter much thought would realize that if these two were to rematch, the smart money would all be on Koscheck. He has a far greater skill level, a stronger track record, more options in a fight; that Thiago knocked him out despite a demonstrably lesser skill level is what is colloquially called a lucky punch, but should be understood more specifically as an unlikely but possible outcome. Let’s say you could get those guys to fight 100 times under identical conditions- odds are, Thiago knocks Koscheck out, say, 8-10 times out of a 100; maybe he submits him another few times, ekes out a few decisions, etc. Kosckeck still probably wins 60-80% of those fights based on greater demonstrated ability. That doesn’t make Thiago’s victory hollow, or unimportant; it makes it an uncommon outcome which was always possible, but which it would be foolish to bet on happening more than a small fraction of the time.

An even better if less recent example would be the two Georges St. Pierre vs. Matt Serra fights. GSP, clearly the more talented of the two, got knocked out in the first round in their initial encounter; in their second, he humbled Serra badly and destroyed him in the second round, and not once since then has there been any talk of a third meeting- for many reasons, one of the biggest of which is that UFC matchmakers and most fans instinctively understand the concept of possible but unlikely outcomes, which because they’re so infrequent, have little predictive value for the future.

To summerize: Arsenal have been badly hit by injuries this year, it’s true; but they have not been more badly hit than might have been expected given the histories of their players, and if you’re banking on them being substantially more healthy next year, you’re either assuming they’ll have substantially different personnel, or else you’re counting on a miracle.

Advertisements

February 25, 2009 - Posted by | MMA, The Arsenal, The NY Rangers | , , , , ,

No comments yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: