Lionel Messi and Soccer Equilibrium Outcomes

So another World Cup has come and gone. Enough water had passed under the bridge that I no longer resented Argentina for their dismal performance in 2002 when I wagered on them. I was vaguely hoping for an Argentine win, just because I would have liked to see Lionel Messi win a cup.

'Twas not to be, of course.

A very good starting point for understanding Messi is this excellent post by Nate Silver going through a whole lot of metrics of soccer success and showing that Messi is not only an outlier, he's such an outlier that his data point is visibly distinct from the rest even in simple plots. Like this one:


(image credit)

Seriously, go read the whole thing. If you're apt to be swayed by hard data, it's a pretty darn convincing case.

So what happened in the World Cup? Why didn't he seem nearly this dominant when you watched him play?

The popular narrative is that there's some inability to perform under pressure - in the big situations when it really counts, he doesn't come through with the goods. He's a choker, in other words.

This is hard to disprove exactly, but one thing that should give you pause is that with Messi on the team, Barcelona has won two FIFA Club World Cups and three UEFA championships. This at least suggests that the choking hypothesis seems more specific to World Cups.

So one explanation consistent with the choking hypothesis is that the World Cup is much higher stakes than the rest, hence the choking is only visible in that setting. It's possible, and hard to rule out.

But another possibility is that the difference comes from the way that opposing teams play against Messi in each setting.

Remember, a player's performance is an equilibrium outcome. It's determined by how skilfully the person plays that day (which everyone thinks about), but also by how many opposing resources are focused on the person (which very few people think about).

Let's take the limiting case, since it's easiest. Suppose I take a team comprised of Lionel Messi and ten guys from a really good high school team, and pit them against a mid-range club team. My guess is that Messi wouldn't perform that well there, and not just because he wouldn't have as many other good people to pass to. Rather, the opposing team is going to devote about 4 defenders just to covering Messi, since it's obvious that this is where the threat is. Throw enough semi-competent defense players on someone, and you can make their performance seem much less impressive.

Have a look at the pictures from the Daily Mail coverage of the game against the Netherlands. In one, Messi is surrounded by four Dutch defenders. In another, he's surrounded by three. The guy is good, but that's a pretty darn big ask of anyone.

In other words, Messi may be better than the rest of the Argentine players by a large enough margin that opposing teams will throw lots of resources into covering him, making it harder for him to shine. In soccer, like in martial arts reality (as opposed to martial arts movies), numbers matter. Jet Li may beat up 12 bad guys at a time, but it you try that in real life, you're on your way to the emergency room or the morgue, almost regardless of your martial arts skill.

The last piece of the puzzle for this hypothesis is the question of why this doesn't happen when Messi plays at Barcelona.

I'm a real newb at soccer (evidenced by me referring to it as 'soccer' - you can take the boy out of Australia, etc.), but my soccer-following friends can tell me if I'm right here or not.

My guess is that the rest of the Barcelona team is much closer to Messi's level of skill than the rest of the Argentine team. This means that if opposing teams try to triple mark Messi in a Barcelona game, the rest of the attackers will be sufficiently unguarded that they'll manage to score and the result will be the same or even worse than if Messi were totally covered. As a result, Messi goes less covered and scores more.

There's a reason that the sabremetricians (who tend to be among the most sophisticated of sports analysers) talk about wins above replacement. You need to think about the counterfactual of if the person wasn't there, not the direct effect of what they did or didn't do in equilibrium.

Of course, the skeptics will point out the cases where great stars did manage to indivdiually play a big role in lifting their national teams to great success. What about Maradona, they say?

This is a fair question. Sometimes you really can get it past five defenders to win a world cup. Maybe that's what a true champion would have done yesterday.

Or maybe the English just weren't marking as well as the Dutch were.

Or maybe, even more pertinent, the rest of the Argentine team in 86 was sufficiently better in relative terms that England couldn't afford to mark Maradona as hard. The effect of this, if true, would be for Maradona's performance to look more spectacular relative to the rest of his team - having a good team means less defenders on you means more heroics. And when that happens, you look individually more brilliant, leading to you getting all the credit and making it look like you won the game single-handedly. If you really were that much better than everybody else, you would be less likely to deliver a performance that showed this fact to a novice observer.

Not many people think in equilibrium terms. This is why we analyse data.

The data case, however, is clear. Viva Messi!

Out of Sample Predictions About World Cup Rioting

So Brazil gets humiliatingly crushed in the World Cup by Germany, 7-1. While there is much to be said about this, mostly in the way of cruel mockery, it has already been done by folks much more learned on the subject than me. As a side note, while watching my streaming of dubious legal status, I did reflect on how the ideal commentators for a complete drubbing are the BBC ones, since they just ooze dry and scathing humour. It's full of great adjectives like 'shambolic' and 'appalling', and they managed to get in some classic digs (quoted from memory):

'This has been the worst 45 minutes of football in Brazilian history'.

'Without Neymar, could this be the worst team to make a World Cup Semi-Final?'

and my favourite of all:

'And Oscar scores the most pointless of World Cup goals...'

So since it would be mean to pile on more, let me focus instead on something where I can add more value. Given that Brazil has been crushed and humiliated, will this defeat lead to rioting? Plenty of people seem to think it will - this CBC story in the Google cache version has the sentence 'Brazil riots feared as home team routed by Germany', but this has now been scrubbed. For a prediction, let's turn to my favorite author on violence, Randall Collins (I've written about him here and here). In his excellent book, 'Violence: A Micro-Sociological Theory' (pdf of the first chapter available here), he makes the following observations (p312):

'During the 2002 World Cup, Russian soccer fans, who were watching the game with Japan on a big screen in a central Moscow square, rioted after Japan scored the one goal of the game...

The 2002 Moscow riot is both a political riot and a defeat riot, the counterpart to a victory celebration riot. As we will see, celebration riots can be just as destructive as defeat riots; and celebration riots are much more common. Losing a game is generally emotionally deflating, and the crowd lacks the ebullience and the traditional rituals (such as tearing down goal posts), which can segue from a victory celebration into a destructive riot. Defeat riots require an additional mechanism. One clue is that defeat riots seem to be more common in international competition than domestically, and where sports rivalries are highly politicized. Defeat riots depend more on features extraneous to the game, since the emotional flow of the game itself will generally de-energize the defeated and energize the victors. 

So while this is an international competition, I'd say that the thrust of the Collins prediction is that, contra the predictions of many, there won't be rioting.

And the verdict?

Brazil Riots in World Cup? Nope; Bogus Photos Spread After Germany Beats Brazil 7-1 in Soccer Semi-Finals; Fake Demonstration-Protest Tweets in Belo Horizonte Trending

1-0 in Russia might have been enough to get people angry, but 7-1 just produces dejection. People don't burn buildings while dejected.

It's still too early to tell, and I'll continue to see if I (and more importantly, Mr Collins) are wrong, but my guess is that there won't be any rioting.

Seriously, if you didn't last time I talked about it a few years ago, go and read the first chapter of Collins here. I am no apologist for the general predictive power of sociology, but the man knows his stuff.

How to make Canadian Football Awesome

When I was in Toronto about a month ago, a friend of mine who lives there told me how there's some suggestion that the Buffalo Bills NFL team may be planning to move to Toronto.

While I don't care a whit about the NFL, there's one surefire way to bolster local support immediately. If the team moves to Toronto, immediately rename them as the Toronto Loyalists.

Firstly, Canadians resolutely love clinging to anything that separates them from America, no matter how anachronistic (the Queen is one thing, but Quebec? Really? They're like a permanent grievance lobby designed to extract rents from the functioning rest of the country. De-annex them, I say). Stoking up vague anti-American sentiment, but in the politest of historical contexts, would be a surefire crowd pleaser.

Secondly, it would immediately create a super popular grudge match whenever they played the Patriots. And since the Patriots tend to be rather good, Americans would love it too because they'd still get to win, just like last time.

When this genius proposal is implemented, I expect fat royalty cheques to be forthcoming. Or a free Toronto Loyalists jersey, which I'd settle for as well.

One and a Half Cheers for MMA

I find myself somewhat conflicted on the subject of mixed martial arts, like the Ultimate Fighting Championships.

Far and away the best thing about them is that they've proven incredibly useful as a vast experiment in the most effective hand-to-hand combat techniques. Previously, all you had was a bunch of different martial arts - boxing, karate, jiu jitsu, what have you - and you'd just pick whatever one seemed cool to you. You'd spend ages developing techniques in that style, and learn how to counter the attacks of someone else coming at you with the same set of moves.

But this left almost completely unanswered the far more important question of what the inherent weaknesses of the style were. In other words, suppose you perfected the techniques of that particular style. What weaknesses would that leave you open to if you were attacked by someone who wasn't limiting themselves to attacking you in the same way that you were planning to attack them?

Hand-to-hand combat instructors, including places like the military, have been interested in this question for ages, and indeed had developed training that was a synthesis of a number of different styles. But UFC really caused this exploration process to explode. By providing a television spectacle and large cash prizes, it gave big incentives for the best fighters in the world to actually explore and come up with different combinations of techniques. The range of styles currently used covers a bunch of principal components (if you will) of martial arts space: ground-and-pound, submission grappling, sprawl-and-brawl (apparently rhyming names have proved popular), etc. These may not have a rich pedigree of historical tradition, but is there really any doubt that learning any one of these would prove vastly more effective than just perfecting a single traditional style?

One of the big lessons that came out of the early UFC rounds is that a lot of traditional martial arts (boxing, karate, muay thai) work great when you're both standing on your feet, but are virtually useless if the guy takes you to the ground. Which, if he's doing Brazilian Jiu Jitsu, he will. And then your training will have very little to say about what you should do. MMA has injected a lot of life into the intellectual question of fighting styles, and forced a bunch of fossilised martial arts to consider honestly what their strengths and weaknesses are.

So that's cheer #1.

Cheer #2, which is really half a cheer, is related to cheer #1. As MMA has become more popular, people have started to learn MMA directly, rather than studying other fighting styles. To the extent that I think that these MMA synthesis styles are better for self-defense, this is a good thing. If you're going to have to defend yourself in a bar, you want to have the most effective way possible. And giving people knowledge that they think will help them defend themselves can actually make them worse, if it causes them to get in more fights because they think (incorrectly) that they'll win.

Did you ever notice that Bruce Lee isn't often seen fighting his way up from the ground in movies, or dealing with guys holding him in grappling moves? Do you wonder why that is? It's not that it's not possible to keep standing up. It's that you're in a lot of trouble if your fighting style relies on both people standing up and being at a distance from each other, and you don't know how to stop the other guy taking you to the ground or getting you in a clinch hold. You're in even more trouble if you're a guy who's learned karate and gets in a fight without having given this some thought in advance. MMA thus ensures that you know better what you'll actually be up against.

The only slight hitch here is that I think that MMA practitioners don't think fully about the implicit restrictions that MMA places on fights which a bar fight does not. To a lesser extent, this is particular moves like eye gouging, small joint manipulation, groin attacks etc. But the much bigger one is the ability to deal with multiple attackers at once. Skills like taking the other guy to ground in a choke hold are immensely useful in a one-on-one fight. They are disastrous if the guy has three friends around who will kick you in the head as you perform the choke hold on the ground. The more people are attacking you, the more 'stay on your feet at all costs' becomes a crucial principle.

That's okay though. People can figure that out. Overall, I'm a pretty big fan of MMA in the abstract, and am interested in what it reveals about fighting styles. So what's the issue?

The problem is that I just find it rather gross and distasteful to actually watch. Just to check, I went over to the UFC website. The current headline was "Free fight - Shogun stomps his way to a pride title", where the photo showed the guy in question stomping on his opponent, who was on the ground. The bout ended, as it usually does, with one guy on the ground being punched in the face over and over until the referee calls it off. And I just can't help but find this barbarous and unpleasant to look at. Every now and again, one of those guys on the ground is actually being beaten to death. Sure, it's rare, but it's still troubling that during every one of those deaths, the crowd was cheering the guy on doing the beating.

In other words, what I find the most gross about these events is the crowd. I personally don't like watching guys beat the hell out of each other. But lots of other people apparently do. You can dress it up in fancy terms like watching the skill and the spectacle, but at its heart, the appeal is the same as that of the circus maximus, nature documentaries where one animal hunts and kills another, and every other kind of blood porn. People find it exhilarating to watch one creature attack and kill another.

The guys in the ring are professional athletes. They know the risks, and they're paid handsomely for what they do. That's fine. It's their job.

The guys in the audience, on the other hand, are there because they like watching people hurt each other. And try as I might, I can see nothing at all to celebrate in their behaviour. People are of course free to exercise their liberty however they want. But it takes a particularly obtuse sort of libertarian to not consider the possibility that a society where more people exercise their freedom to watch the ballet might have more to recommend than one where people exercise their freedom to watch a boxing match.

This may be human nature, but it's a particularly dark side of human nature, and not one I think ought to be celebrated.

Random Thoughts on the Olympics

- It's always good when you're watching a group of runners lined up on the track without hearing the earlier announcements, and you can tell the event purely by the competitors involved. Hmm, Kenyan, Kenyan, Ethiopian, Kenyan, Ethiopian ... it's starting inside the track, so it's not the marathon, meaning it's got to be the 10,000m. Sure enough, it is. Correlations, man - is there anything they can't tell you?

-It was grimly hilarious a few days ago to watch the Australian Olympic officials trying to put on a brave face after winning Sweet F. A. when the swimming was all done.

http://au.oztips.yahoo.com/news/article/-/14433008/aussies-not-panicking-over-low-medal-count/

But fear not, says Australia's deputy chef de mission Kitty Chiller.

"Very early days, we're only just starting the second quarter," she said.

"We've got rowing, we've got track cycling, we've got sailing, genuine gold medal hopes - three in each of those events.

"We're certainly not panicking. There's a still a very positive feel amongst management and the athletes.

"Sure, we maybe have missed out on a few medals that we thought we could've one but we've also won others - 4x100m freestyle wasn't a gold medal favourite.

"There's certainly no fear at the moment that we've failed, that we're not on track.

"We still believe we can genuinely finish in the top five overall."

Translation: the tanks are descending on Berlin from both the east and the west, but the German Army is about to fight a glorious rearguard action!

Sporting Overconfidence, Part 2

The second effect of overconfidence on the sports field is that people over-invest in the sport.

In other words, when you think you have high skill, the rewards to training are higher, because you could go on to be a superstar. And in truth, the extra training will be useful, as training always is. You will improve because you train heaps.

But the margin on which you'll make a mistake is that you'll overinvest in the sport relative to what else you could be doing with those hours - hanging out with your friends, learning Russian, snorting meth, whatever your chosen avocation is.

Unfortunately, this creates even worse effects when everyone else is overconfident too. When you know that all the other teams are likely to overinvest because they're overconfident, it means that you'll need to train that much harder in order to beat them. In other words, even if you aren't overconfident yourself, the only way to beat the other teams is to act as if you were overconfident.

Once again, behavioural economics comes to the rescue, with the sage of advice of 'Can't win, don't try, spend your time enjoying life instead'. Not exactly the stuff of inspirational speeches.

But sod it! There's more to life than winning on your six-a-side soccer tournament. How about just enjoying yourself?

The second obstacle to this is the team structure. The team captain is usually among the most psyched up about the team's chance. So you often get conversations like the following:

Captain: I was thinking we'd train three times a week. You guys agree, right?

Everyone else: *shuffle feet, don't want to be seen as the lazy one*.

It takes an unusually bold person to demand that everyone train less because they personally are lazy. But then the consensus answer is always more training, even if that's not what most people want.

If I were running things, I'd start out the first meeting with the following:

'Okay, I want everyone to write down on a piece of paper the number of hours per week they'd like this team to train, ranging from zero to five. We'll put all the pieces in a hat, then draw them out, and whatever is the median answer will be how much we'll train.'

And my team will probably get our butts kicked! But it won't matter, because we'll be doing other fun stuff and not viewing training as being a chore.

Did I mention I'd make a rotten team captain?

"We're better than those guys

Statisticians rarely make good members of sports teams.

I found this out the hard way when I used to be on a frisbee team. Most people run on overconfidence. I've had numerous arguments with people over whether this makes sense or not. The general view is that if I psych myself up that we're going to win, I'm going to try harder to make it happen. If I believe I'll fail, I'll be demoralised and not try hard.

The idea is thus that belief in success and failure has a self-fulfilling component. Only a component, mind you - if I really truly believe I can beat Kobe Bryant to the net in a game of one-on-one, I will fail. But I'll still have a better chance than if I don't believe in myself.

Frankly, I was always a bit skeptical of this argument, as it reeks of a second-best solution. In other words, if you're being rational, better answers are unlikely to come from deliberately feeding in faulty inputs. Including your chances of victory. This only works if it's the workaround to some other faulty process - one bias (inability to try hard in the face of failure) is offset by another bias (convincing yourself that you won't fail).

But I remain committed to the belief that the first-best solution is always to eliminate the biases - in this case, figure out how to try hard even if you do think you'll lose.

Since this is what I aim at, I want to know the true probability, and work from there. It's a fair bet that most other team members (if they're non-economists or non-statisticians) won't feel that way. They'll view you as a negative nancy.

I remember this came to its zenith when we were down at half time. The captain of the team was trying to get us fired up. He said 'hands up who thinks we're going to win this game'. About half the team put up their hands. He responded, 'Right, you guys are on the field'. Personally, I thought this was absurd, but that's probably part of the reason I never got made captain.

The net effect of all this is that you end up with the absurd result that on any given sports field, at least 70% of the players think they're going to win. They think that they're better than the other team. Talk about the Lake Wobegon Soccer team effect.

It also leads to a hilarious misconception of what it means to be 'better' than the other team. For most people, if they lose on a knife-edge, they'll be bitterly disappointed.

But the statistician sees it differently.

If we play against a really rubbish team, we'll win about 95% of the time. Then we'll advance higher, and play a better team, that we'll beat 70% of the time. We'll advance higher still, until we're playing a team that we have an edge over, but it's tough - we might win 60% of the time. 

And eventually, we'll get to a point where we're playing against a team that's very evenly matched. We'll have a 50% chance of winning. And we might just end up in a 16-16 game to 17. And someone drops the disc, and the other team scores. And we lose.

The non-statistician weeps.

The statistician is sanguine. In expectation, we got exactly where we should have. We bet on a fair coin, and it  came up tails. This time we lost. Next time we'll win.

But there's no disappointment just because the coin landed on tails.

Funnily enough, that might make for a reasonable consolation speech afterwards. It would certainly have a better likely effect relative to the 'we're probably not going to win, but I plan to try jolly hard anyway' speech.

On the other hand, I'd would be much more inspired by the speech that talked about the true probabilities.

After all, not everybody who's willing to face up to true probabilities is necessarily a coward. The best response to likely defeat is to stare the truth in the face, and give it the finger.

Faulty Logic on the Soccer Pitch

Let me explain briefly some logic that may seem superficially appealing, but is in fact catastrophically wrong:

Nobody wants to play goalie on our team. I have good hand-eye co-ordination, but I'm not that accurate at kicking. I'm probably going to be the worst player on the field if I play a normal position, whereas I'm probably at least decent at just catching the ball. I'll give goalie a go.

Yeeeaah. Let's call this 'Shylock logic, circa last Sunday at 11:50am'. At about 12:10pm, after getting scored on multiple times by a team that was clearly better than us, let me enlighten you on just how much you can learn in ten scant minutes about where your thinking went astray.

1. When you're crap at some team endeavour, it is less important to actually be good at something than it is to not be blamed for actively being bad at something in a way that is harming the whole team. Goalie is the worst possible position for this. Every screwup is directly attributable to you and you alone. In addition, there's going to be lots of people that remember your screwups and have it in for you.

2. Being goalie is like writing put options, in that your payoffs have crazy negative skewness. When you stop a ball that you should stop, you get a small amount of praise. But should you miss a ball that you should have been able to stop, you just went broke. You know who should be writing put options? Really highly rated financial institutions with deep pockets and lots of experience. You know who shouldn't? Mum and Pop.

3. You know how you can tell that a team is likely to be crap and full of amateurs? When nobody wants to volunteer to be goalie. You know when is the worst possible time to be goalie? When your defence is crap and full of amateurs. Because that's when you're going to get lots of shots on goal, and when you inevitably let some in, people will still blame you more than the crap defence. Talk about a winner's curse problem.

The only good news of the day was that when I subbed out, the goals kept coming, albiet at a slightly slower rate. I am quite certain I was the only person on our whole team to be significantly pleased by this development.

Moneyball Part 2

Steve Sailer's take. I really liked this anecdote:

When my son was ten, his baseball coach—inspired by Michael Lewis’s bestseller Moneyball: The Art of Winning an Unfair Game—came up with a statistically brilliant team strategy: Don’t swing. Ever.

Because few ten-year-olds can throw more strikes than balls, his team won the pennant by letting the little boy on the mound walk them around the bases until he dissolved into tears and had to be replaced by another doomed lad.

 Ha! That's fantastic.

The next spring, the parents got together and decided not to let that coach return.

Boo-urns.

Mixed Strategies

John Nash would be proud.

Today is clearly a double-dose of Yank Sports + Economics.

In other news, penalties for 'excessive celebration' are seriously pissweak. When even curmudgeons like me who don't care at all about American football think the rule is just needlessly squashing the joy in sports, that's a pretty lame achievement.