Free Pedro

Yesterday, I posted the average career arc of MLB players seeking to help explain the ramifications of age on player development and how it impacts signings and trades. One key critique was that there was nothing comparative in the analysis. Today, I’ll give an example of how this analysis can be used for comparative purposes by looking at a trade of Clay Buchholz and Roy Halladay. Clearly, this trade would never happen one-for-one. The trade of Clay Buchholz straight up for Roy Halladay would be a no brainer, but I think it is important to consider this hypothetical exchange in terms of the impact of age on the way that we think about prospects. In the graph below, I have projected the ERA+ (explained here) of each pitcher going forward using the same analysis technique that I wrote about yesterday.

For each player, I started with the current season’s ERA+. There are two things of special importance to consider: (1) In the regression analysis I based my analysis on, the r-squared was only .28, which means that a good deal of the variance of ERA+ is not explained by age alone. Of course this is the case, but what it means for the analysis below is that I am only showing the effect that we can expect age to have on the future performance of the pitchers. We are leaving out everything else (defense, injury, coaching, learning a new pitch, etc.). (2) We can not be sure, especially in the case of Buchholz, what each pitcher’s true ERA+ is. As such, I have included error bars. Halladay has error bars at 6%, which includes his variance of the last two seasons. Buchholz are much larger (10%), which means he ranges in talent from about as good as Lester was last year, to about as good as John Lackey was last year. Now, he may be much worse, he’s probably not much better, but it gives us a jumping-off point.

The graph above below suggests that Clay Buchholz will eventually become a better pitcher than Roy Halladay due to Buchholz’s growth and Halladay’s aging, but this won’t happen until 2018, when Roy Halladay is 41! It also says that, even if Buchholz is much better than we expect and Halladay is much worse than we think, chances are they won’t be comparable as pitchers until 2011, when their error bars cross. I’d go so far as to analyze this graph as saying that it is unlikely that there will be a time in the next 5 years where we’ll consider Buchholz a better pitcher than Halladay.

So, the relative ages of our pitchers in question does not make me any less likely to trade for Halladay. At this point, we can be fairly certain that Halladay will be a better pitcher than Buchholz for a long time. Knowing that the Sox will get the best player in the deal, and that he will remain so for a good duration, does not necessarilly mean they should pull the trigger. Obviously, how highly the Sox value the other prospects in play, who are much harder to project than Buchholz because they are younger and lack Major League experience, will make or break this deal.