Pedro Martinez: Duel Personalities

A classic pitching duel. One for the ages? Probably not, but it must have been a lot of fun to been a Mets fan in the stands or in front of the TV yesterday afternoon as Pedro Martinez and John Smoltz went toe-to-toe for seven innings and change.


Pedro 

After a less than stellar re-debut (re-but?) as a starter on Opening Day (six earned runs in 1.2 innings), Smoltz was masterful this Sunday, as good as he had ever been, allowing only two runs and no walks while striking out 15 in 7.1 innings. If this performance was any indication, Smoltz will be just fine returning to the rotation this year, as long as he can stay healthy. That 7.1 inning performance was the longest he'd been out there since September 1999, before he missed all of the 2000 season following his eleventeenth reconstructive elbow surgery. And though I couldn't find the data to prove it, I'm sure 115 pitches are more than he's throuwn in several years as well. Let's hope he holds up, because it sure is a lot of fun to watch him pitch.

Unfortunately for him, he picked a bad time to be awesome, because Pedro was, er..., awesomer, fanning nine while allowing only two hits and one run in a complete game, his first since last August, which had been his first complete game since September of 2003.

Pedro Martinez was once one of the greatest pitchers in the game, and he will no doubt go down in history among the greates pitchers in history. But at this point in his career, 33 years old and having just posted the highest ERA of his career, he's certainly not what he used to be.

His reputation at this point is that of a 6-inning pitcher, or, if you will, a 100-pitch pitcher. The reality isn't quite that bad, though there is some evidence that his effectiveness drops off considerably after the 100-pitch mark:

2003-04 Splits for Pedro Martinez

                 BA    OBP   SLG  OPS     
Pitches 1-105:  .227  .290  .363 .653
Pitches 106-?:  .297  .350  .378 .728

Granted, lots of pitchers would love to only allow a .378 slugging percentage (you reading, Eric?), regardless of the game situation, but for Pedro, this denotes a considerable fall-off.

Two caveats should be given with these data, though. The first is that the high-pitch count data are gleaned from a very small sample size, only 74 at-bats, as compared to the first lot of data, which occurred in 3,154 at-bats. Because the sample size is so small, it's skewed severely by his 2003 numbers, when he allowed 12 hits (mostly singles) in 33 at-bats at the ends of high-pitch count games. Neither the 2002 nor the 2004 numbers show such a severe disparity, but I didn't think it was fair to look at his 2003 stats without showing his most recent work.

The second caveat is that this does not include his performances in the playoffs the last two years, when he was, shall we say, not so good late in games, especially when his pitch count exceeded 100. I didn't have a good way to evaluate those and I can't take the time to go through every pitch and do all the math. Sorry.

The other reputation, that of a six-inning starter, is one that should be taken seriously too, though. As I mentioned, it had been eight months since Pedro had hurled a complete game, and even though he tossed only 101 pitches this Sunday, I got to wondering how he tends to follow up such performances:

           IP  H  ER  HR  BB  SO  PIT   ERA
Comp Game   9  7   1   0   1  10  117  1.00
Next Game   6  6   3   1   1   7  101  4.50

Like I said in my last mini-analysis: small sample size, so take it with a grain of salt. It should also be understood that any drop in performance may have had more to do with the relatively high pitch count (117 on average) in these complete games than it had to do with the actual innings.

However, with that said, you can see that Pedro has not typically followed up these appearances all that well. He averages only 6 innings, and hit hit-, walk- and home run-rates all increase, as does the strikeout rate, ever so slightly. I only used his last four complete games and the games immediately following them for these data, but it does seem that at least for the last couple of seasons, Pedro found that he was a tough act to follow, even when he followed himself.

So, come Saturday when Pedro faces former Met Al Leiter, don't be toos urprised if he "only" produces a Quality Start, something like the 6-inning, 3-run performance he averaged above. Against Leiter, that may be enough to win anyway.