Saturday, February 2, 2008

The Long Haul

Tom Haudricourt started a discussion in his blog about which of the Brewers young players they should try to secure with long term contracts.

The collective bargaining agreement says that once a player has three years of Major League service he is eligible for salary arbitration. After six years he is eligible for free agency. J.J. Hardy was eligible for arbitration this year for the first time, and Corey Hart, Prince Fielder, and Rickie Weeks will be eligible next year for the first time. Should the Brewers try to sign these players to multi-year deals?

If a player goes to salary arbitration, it seems by definition, you wind up paying fair market value for him. True, a little bit higher or lower depending on what the arbiter decides, but all-in-all, pretty close. If you chose to, you could just keep signing players to one-year contracts throughout their arbitration years and as a result wind up paying roughly a fair market price for them. The decision to sign a player to a long term contract while he is in his arbitration years would then be dependant on whether you feel the player will become better or worse than he is right now and therefore worth more or less money. If you feel that a player is going to get better and therefore demand more money two or three years from now, it would be wise to try to sign him to a longer deal with the thinking that over the course of his contract you will wind up saving money. Strangely from a player’s perspective, it works just the opposite. If a player feels that he is going to be significantly better two or three years from now, then he will hurt his earning potential by signing a long term deal.

Now of course that’s all very simplistic. There is more to the decision than that. From the player’s perspective, there is the security of knowing that you have a guaranteed pay check for some period of time and from the team’s perspective there are things like present and future value of money, gate receipts, media revenue projections, and so on. But at its core, it seems to still boil down to figuring out which players have the best potential to improve, then signing the ones who do and not signing the ones who don’t.

Right now the three players that the Brewers have signed beyond the 2008 season are Jeff Suppan, Bill Hall, and David Riske. A handful of others have a club option for 2009, but those are the three to whom the Brewers have guaranteed money. If you looked at their roster and decided which three players you wished were signed beyond ’08, none of those three would probably be on the list. In fact they wouldn’t be in your top five or top seven. That being said, you could argue that the Brewers may have made unwise decisions by signing those three players and instead should have saved the money to lock up Fielder, Hardy, Weeks, or Hart at the risk of possibly losing in Hall’s case, or not signing in the other two cases, those players.

The history of the Brewers, and many other teams I suppose, is littered with players who signed long term contracts and then underperformed. At the time the signings took place, they seemed to make sense, but as they played out the players became a financial drain on their teams, locking up precious resources for little in return.

No one has a crystal ball to know if a 24 year old player is going to be better when he’s 27 than he is today. If you did the decisions would be easy. But there is a sense I think that most decent 24 year old players will be better when they’re 27, and therefore it makes sense to sign as many of them as you can for as long as you can. You think that in the long run it will be cheaper to sign J.J. Hardy to a three-year deal now than to sign him to a series of three one-year deals for an amount decided by an arbiter. But then why do so many of them go bad? Just what are the odds of a player being better when he’s 27 years old than when he’s 24?

Well, I thought I’d check.

I set up a study using the Sean Lahman database. The study (for now) only includes hitters. The metric I used for determining value was Bill James basic runs created formula:

RC = (H + BB) * TB / (AB + BB)

[I’m sure that the 50% of the readers of this blog who are sabermetricians are both cringing at the simplicity of this design, but I feel it serves the purpose.]

I pulled a list of players from 1950 to present who at the age of 24 (season year minus birth year) had at least 25 runs created, and who were at least 27 years old in 2007 (so I could study them three years later). In order to reach 25 runs created, you have to play about half time. Ranked by RC at the age of 24, at the top of the list were Albert Pujols, Vladimir Guererro, and Willie Mays; at the bottom Luis Terrero, Mike Heath, and Kenny Landreaux. There were 748 such players.

Then I looked at the runs created for those players when they were 27 – three years latter. I was curious how many were better and how many were worse.

For each player, I indicated whether they had gotten better or worse simply by the difference in their runs created. Of the 748 players in the study, 376, or almost exactly half had improved by the time they were 27. The other 372 had declined (including 43 players who were no longer playing at the age of 27). Isn’t that interesting. It’s a crapshoot. A 24 year old player on average has a 50/50 chance of being better in three years.

Then I divided the players into four equal sized groups based on the rank order of their 24 year old RC.

Group Players Improved Pct
Top 187 78 42%
Second 187 92 49%
Third 187 97 52%
Bottom 187 109 58%

As you can see, players in the top one-fourth of the study were far less likely to improve than players in the bottom one-fourth. There are two forces at work here. One is that players tend to improve up to about the time they are about 27-30 years old, and then decline after that. The other is that sometimes players put up stats that suggest a level of achievement above their natural ability – perhaps because of luck more than anything. Those players’ stats tend to decline the next season. This study suggests that the second force is stronger.

I repeated the study looking at 25 years olds and compared how they fared three years later at the age of 28. This time, using the same post-1950 and 25 RC criteria, there were 1,032 players in the study. (This stands to reason since there are more decent 25 year old players than there are decent 24 year old players.) Three years later, only 412 of those players, or 39.9% were better.

I again split the study into four equal groups:

Group Players Improved Pct
Top 258 82 32%
Second 258 116 45%
Third 258 111 43%
Bottom 258 103 40%

This time it was only the upper fourth of players who much less likely to improve, however there was no group that was more likely to improve than decline.

I think the moral of the story is that the players who you would be most tempted to lock up with long term contracts, the ones who are very good at a very young age, are the ones from whom you are least likely to get a return on your investment. An example is Prince Fielder. He hit 50 home runs last year. Next year he is probably more likely to hit fewer than 50 than he is to hit more than 50. That doesn’t mean he won’t hit more than 50. It just means that the odds aren’t on his side. If you sign him to a contract as if he’s going to hit 50 home runs every year, you have a good chance of being disappointed.

Jared over at Right Field Bleachers discusses the Brewers situation and eloquently echoes my feelings. The Brewers don’t need to be in any hurry to sign their players to long term deals and to do so could derail their future. They have exclusive rights to J.J. Hardy for another three years, Prince Fielder, Rickie Weeks, and Corey Hart for another four, and Ryan Braun and Yovanni Gallardo for six. There’s plenty of time to sort out who’s worth keeping and who’s not.

No comments:

Blogged Blog Directory