How to buy an IPL cricketer

In a paper to be soon published in the Journal of Sports Economics, Ajit Karnik, a professor of economics at the Middlesex University campus in Dubai, placed the IPL player auction of 2008 under the scrutiny of an econometric instrument known as hedonic regression. The result was an equation to help determine a cricketer’s value; the more remarkable by-product, however, was its accurate picks of IPL 2008’s four semi-finalists and its eventual winner.

Hedonic regression is not, as its name implies, a slide into the quest of unalloyed pleasure. “Say you’re buying a car. There are various characteristics of the car, such as engine power, mileage, looks and so on," Karnik says. “But when you pay a lump sum, you don’t say: ‘I’m paying so much for the engine, so much for the looks.’ So in hedonics, you try to infer the value of the various characteristics for this kind of product."

The “product", in this case, is the sportsman, and hedonic regression has, in fact, been long used in the study of American sports. In 1974, for instance, the influential sports economist Gerald Scully wrote a paper titled Pay and Performance in Major League Baseball to measure the economic value of a baseball player.

When he updated that paper in 1989, Scully was able to state: “Four factors are crucial to the determination of player salaries: the overall quality of player performance; the weight or fraction of the player’s contribution to team performance; the experience of the player; and, the popularity or recognizability of the player to the fans."

But the opportunity to deploy hedonics in cricket came only with IPL, Karnik points out, because that was the first instance where players operated in a free market, to be bought or sold like free agents in baseball or like footballers in the English Premier League. “I was looking at the auction in February 2008, and people were quoting astronomical figures for players," he says. “So I figured there must be a reason they’re doing this, and I wanted to try to get the players’ characteristics."

Here’s where it helped to be a cricket nut as well as an economist. The basic price equation would have to include runs and wickets, and Karnik knew enough about the non-stop nature of Twenty20 to fold age into that mix as well. For data, he used the players’ previous one-day international records, since many of the cricketers in the auction hadn’t played even a single Twenty20 game.

Then Karnik stress-tested a number of other additional variables against each other, to figure out which would prove significant and which wouldn’t. That told him to include strike rate, and it also introduced a premium if the player was Indian or Australian—“Indian because the IPL was being held in India," he says, “and Australian just because of their sheer performance over the last decade".

After what he calls “this very painful process", involving long hours on ESPNcricinfo, Karnik had his equation. Speaking strictly from the econometric point of view, the IPL franchises value every run at $34 (Rs1,581) and every wicket (“like gold dust in cricket", Karnik remarks) at $679. Every year a player ages shaves $20,027 off his market value, and every one-unit increase in strike rate is worth $1,226.

Plugging into the equation the cricketers’ eventual performance in IPL 2008, and comparing that with the prices paid for him at the auction, Karnik was able to work out the rates of return for each of the franchises. The four franchises with the highest rates of return turned out to be the four semi-finalists, and the franchise with the very highest rate—getting a 58.5% return on its investment—was the Rajasthan Royals.

“I was quite pleased to see that, although I don’t think the hedonics model usually picks the winner," Karnik observes. That first orgy of player-buying now in the past, he thinks that the bidding will become less idiosyncratic and more shrewd. “The best part of the study, of course, is that after having been an armchair cricketer for years, I could finally work with the game I love."

samanth.s@livemint.com