Sunday, May 20, 2012

Economics and Championships

Writing at his Forbes blog Stefan Syzmanski notes that the big game yesterday was not the Champions League final, but the promotion playoff between West Ham and Blackpool:
In one of these games the difference between winning and losing is €70 million, in the other it is €3 million. Amazingly, it’s the Premier League playoff game which involves the big bucks. UEFA pay €9 million to the winner and €5.6 to the loser of the Champions League Final (although you could argue that the winner will cash in via merchandising and an enhanced international reputation).

The winner of the Premier League playoff will get a minimum share of next year’s broadcast income worth about £40 million, and even if relegated immediately, receives parachute payments worth £48 million over four seasons. Since both West Ham and Blackpool are already entitled to another £32 million of parachute payments after being relegated last season, the net gain for each is £56 million, which is about €70 million at today’s exchange rates.
With respect to that other big game, Olaf Storbeck, economics correspondent for Handelsblatt, had predicted a Bayern victory based on a commonly used metric of skill oft discussed on this blog -- the simple economics of football squads:
[B]ased on my professional background I’m rather confident that Bayern Munich will beat Chelsea this weekend.

This forecast is not based on the fact that Bayern will play on its home turf. It is rationally derived from purely economic arguments: Judged by the current market value of the individual playes, Bayern will have the stronger and more powerful squad on the ground.

On Friday, the German football magazine “kicker” published the most probable formations of Bayern and Chelsea. Judged by the current market value of the player – estimated by the football website Transfermarkt – the German XI will be about 30 percent stronger than the English one.

The Bayern XI will be worth 290.5 million Euro while Chelsea’s players are just worth 202.5 million Euro. Even if the ageing forward Didier Drogba would be replaced by the expensive (and mostly useless) Fernando Torres, Chelsea’s value would only climb to 235 million Euro.
Well, that prediction did not verify. Even though transfer market value of squads presents a high bar of skill to overcome in prediction (see, e.g., this paper in PDF), it obviously does not explain everything.  Otherwise, we wouldn't pay much attention to sport.

In an evaluation of his prediction Storbeck explains that economics, by contrast, does have a post hoc explanation for every possible outcome:
However, the beauty of economics is its versatility. The dismal science not only explains why Bayern should have won yesterday. At the same time, the discipline is also able to point out why the team actually lost.

Potentially, the Germans faltered in the penalty shoot-out because they were playing on their home ground. This at least suggests an analysis by Thomas Dohmen, a German economist at the Univesity of Maastricht in the Netherlands.

In 2005, Dohmen analysed the performance of professional football with regards to penalties. He came to the conclusion that players of the home team have a significantly higher failure rate.

In his paper entitled “Do Professionals Choke under Pressure?”, Dohmen analysed all 3610 penalties that were awarded between 1963 and 2003 in the German Bundesliga.

The descriptive statistics of Dohmen’s data show how egregiously Bayern Munich failed yesterday. The German XI squandered 3 of their 6 penalties (this includes 5 in the shoot-out and one in the extra time). This failure rate of 50 % is twice as large as it is usually the case. As a rule, only about 26 percent of all penalty kickers do not score (19 percent of the kicks are saved by the goal keeper, 7 percent miss the target completely).

However, home teams have a worse track record than away teams: “Players are more likely to choke on a penalty kick when the action takes place at the home turf”, writes Dohmen.
The manner in which sport resolves uncertainties of the real world appears to be far more appealing than the approach taken in academic research.

1 comment:

  1. I went and looked at the underlying 'choking' paper.

    3,619 penalties, 252 were missed by choking (clear miss or woodwork) which is 6.96%

    2,560 were home, 193 missed by choking, that's 7.54%
    1,059 were away, 59 missed by choking, that's 5.57%

    He claims that as a robust finding (home chokes robustly more likely than away chokes)

    Personally, as a layman, if I made a 'finding' like that in financial data like the stock market I wouldn't consider it very robust.

    Am I being simplistic?