Tim Wakefield: an 80-pitch pitcher.

A walk and one flat pitch is nothing close to resembling evidence.

(because the numbers support it), but not the application to yesterday’s game. A pitcher who had been as effective as Wakefield had been through six innings, 83 pitches (only one base hit besides the bunt single, one walk, one HBP) should not be pulled, especially when the bullpen is routinely overtaxed. And there’s no manager in the league that would pull him, even with full knowledge of the numbers you cited.

Yesterday’s game was lost because the 1-6 hitters in the lineup were 0 for 20, not because Francona and his staff failed to follow an 80-pitch rule for Wakefield. And I would be shocked if they were not aware of his increased ineffectiveness late in games. Do you really think that you are paying more attention than they are?

Worth noting that I don’t see the same pattern in Tim’s 2010 splits, but it does show up in 2009 (OPS allowed by 25-pitch chunks: .677 pitches 1-25, .678 26-50, .770 51-75, .899 76-100). So it could be a blip this year, since last year didn’t show anything, but he showed the same pattern in 2009, and a much starker difference this year.

This is in no way a detailed statistical study, but I do think there’s something here.

Wow, who knew? I wonder if it’s the same principal that doesn’t allow me and my fat ass to run 40 miles every day, fatigue. Not exactly a mind blowing conclusion that after a certain point a pitchers OPS against will climb, I bet that just about every pitcher in baseball has the same telling stats.

However, last night while technically fitting into the pattern, doesn’t really fit the theory. Yes, he gave up that HR, but he was still pitching just as well to everyone else, the HR wasn’t caused by the fatigue that says Wake shouldn’t throw more than 75 pitches.

So I ran the numbers without those two games (7/18 vs. the Orioles and 7/24 vs. the Mariners) to see how deep an impact these starts alone made on the figures. I also added the seven PAs at pitches 101+ for the season (0 H, 2 BB, 3 SO), as those are also relevant to the discussion. So for pitches 76+ on the season, other than those two games:

PA.. BB.. SO… OBPA… SLGA…. OPS
58.. 7….. 5….. .349…… .531…… .875

So the pattern holds (cf. the numbers above for pitches 51-75: .330 OBPA/.465 SLGA/.795 OPS) that says the aging Wakefield gets worse late in pitch counts, but the difference is more gradual and much less drastic (>.275’s difference in OPS). Of course those numbers treat those two bad games as outliers, yet we who follow Wakefield are on our toes waiting for them to happen at any time. But it is worth noting the circumstances of those two games. In the one he was trying to eat the 7th inning with an eight-run lead when he gave up a grand slam on his 100th pitch. In the other he was trying to get through the 5th having been given a 6-2 advantage when he gave up the lead on his 93rd pitch. In both cases he was left in a little longer in hopes of sparing the bullpen, and in both cases the game was managed effectively enough as both were won by the Sox despite Wakefield’s shortcomings.

In both of Wake’s starts since his two stinkers, he has been kept in the game in the 7th inning long enough to give up the lead. I expected it to happen both times and these numbers support my intuition. But on both occasions he was giving up only his third run of the night after a very well-pitched game. Francona, I suppose, was willing to bank on either Wakefield following through with a less than three-run outing (a not very good chance) or the bats carrying the team (a pretty good chance), and that seems to me a reasonable enough way to manage a struggling rotation that taxes the bullpen but which is supported by an historically good offense.*

*This assessment of last night’s game hangs on the fact, which Francona has demonstrated well enough for me, that the way to beat the Indians in late innings is with hard-throwing lefty relievers—even Franklin Morales and Randy Williams will do. (We saw what happened even to Bard for the second time in a row, though Papelbon seems to know what to do.) Hence it wasn’t the worst of plans after the sixth to try to use Wake vs. the bottom of the Indians’ order plus Williams starting with Kipnis to build a bridge to Paps to close the game. Either Wakefield would get three outs between Chisenhall, Kearns, Marson, and Carrera, or someone on this ridiculously good offense would step up and in three innings’ time produce one more RBI.

Nice post. It’s true that his pitchers do tend to “go flat” at about that point (80 pitches/6 innings), and a 68mph flat knuckle ball (essentially a very slow change-up w/o the threat of a 90mph fastball coming) isn’t that hard to hit.

Wake is the man, a real blue-collar guy. He’d throw 15 innings if Tito left him out there. But you’re right, Tito needs to exercise more restraint.

I haven’t looked them up, but what are the numbers on Wake’s relief appearances throughout his career? It’d be interesting to see, if when coming in as relief, he needs a certain # of pitches to “warm up.” Obviously he doesn’t throw fast and put an immense strain on his arm, but it must take a certain # of warm-ups pitches to get that knuckle ball going right. We all remember that one relief pitch. I’m not focusing on that one, but rather his career in relief. It’s a finesse pitch and takes a certain touch. Most likely, like with other starters, it’s difficult to switch to “relief mode.”

Personally, I like Wake as a starter. I feel that within the first 20 pitches you know which Wake you have for the evening, and you can plan accordingly. That’s why it’s great to have sometime starters like Aceves in the pen. If it’s bad Wake, you bring in Aceves in the third. If it’s good Wake, you hope he hands the ball off to Bard.

Good post!