Counterpoints to Starting Burnett tonight

Let me start by saying that I support the Yankees’ decision to start A.J. Burnett this evening. That was the plan, and the Yankees are sticking to it. The circumstances might seem dire, but the Yankees knew this scenario was a possibility. They also know that no matter how well or poorly Burnett pitches, they’ll play tomorrow and have their ace on the mound. That sounds a bit better than having Burnett pitching a potential elimination game.

Still, I can’t help but examine the other side of the issue. Going down 3-1 and having to face the Rangers’ three best pitchers doesn’t exactly seem like an ideal scenario. CC Sabathia undoubtedly gives the Yankees a better chance to even the series, so there has to be something to the argument that favors him taking the ball on short rest. Let’s examine that case.

This morning Craig Calcaterra laid out the argument for starting Burnett tonight. Within he makes two main points and one side point. The first is that starting Sabathia tonight would only delay the inevitable, since Burnett has to pitch in this series one way or another. The second is that Burnett matches up better against Tommy Hunter than he does against C.J. Wilson. The side point is that pitchers going on short rest generally have poorer numbers than their normally rested counterparts.

To counter:

1) Yes, Burnett will pitch in this series no matter what, but that doesn’t mean that pitching him in Game 4 is the same as pitching him in Game 5. In Game 4 the Yankees will either enter an elimination scenario or they will avoid it. No one wants to enter that elimination scenario, and so I can understand going to the ace in order to keep the team alive for at least two more games. If things go according to plan and Sabathia wins tonight, Burnett could still pitch the Yankees into an elimination game. But being down 3-2 is quite different than being down 3-1. Of course, CC is guaranteed nothing and could lose the game tonight. I’d still rather lose with my best on the mound than with my fourth best.

2) Burnett matches up well against Hunter, apparently, because Hunter is not only the least of the Rangers’ pitchers, but he has fared poorly against the Yankees. Yet we know that Hunter’s regular season performance against the Yankees means exactly zero right now. This is one game, and anything can happen. If we’ve learned anything from the endless previews for each playoff game, it’s that there is no way to get a good idea of what will happen based on past performance.

What we do know is that the Yankees’ offense has been horrible this series. Can we expect them to turn it around against Hunter? Maybe. Again, anything can happen in any given game. If the offense doesn’t show up I’d trust CC to hold down the Rangers far more than I would trust Burnett.

3) Looking at the data from the past five years, yes, pitchers throwing on three days’ rest fare worse than those throwing on four, five, or six-plus days. The problem is that those starts on three days’ rest count for just 1.2 percent of all starts. In addition, a number of those starts on three days’ rest don’t follow other starts. For example, Javier Vazquez has one start on three days’ rest this season; it came after he faced one batter in relief, which certainly calls into question some of the data.

For his part, Sabathia has been superb when pitching on three days’ rest. He has done it during the regular season four times in his career, accumulating 26.2 innings and allowing just seven runs, three earned. That also comes with 26 strikeouts to just six walks. In last year’s postseason he added another two starts and 14.2 innings with three days’ rest. His overall line:

41.1 IP, 28 H, 11 R, 7 ER, 11 BB, 37 K

The walks is the most interesting part. The overall numbers for pitchers going on three days’ rest indicates that control is the biggest problem. On four days’ rest pitchers have walked 7.8 percent of hitters. On three days’ rest they have walked 9.4 percent. I’m not sure how much of that is the noise of a comparatively small sample, but it does appear to be a significant difference. Home run rates do favor four days’ rest pitchers, but not nearly to the degree of walks. Strikeout rates are close. Sabathia’s walk rate on three days’ rest is 6.8 percent, while his rate on four days’ rest (not counting the postseason) is 7.1 percent.

How would the Yankees manage the rotation if they went with Sabathia on three days’ rest? That would push Burnett to Game 5, Phil Hughes to Game 6, and then either Andy Pettitte on normal rest or Sabathia on short rest in Game 7. I’d go with Pettitte, having Sabathia ready in the bullpen. It’s almost exactly the same as the current scenario, except Sabathia would have on extra day of rest for a relief appearance in a potential Game 7.

The overall point of using Sabathia tonight is to give the team the best chance of prolonging the series. No matter the match-up, Sabathia gives the team a better chance to win this evening. That means at least two more games. Since the Yankees are down two games to one, prolonging the series should be their foremost thought.

Again, I’m not sure if I buy this argument myself. I do think Burnett matches up best with Hunter, and I do trust Sabathia in an elimination game against Wilson. The Yankees have said that they’re not starting Sabathia on short rest, so I assume that they have reasons for not doing so. But there is certainly something to the argument for starting Sabathia tonight and Burnett tomorrow. Thankfully, in both scenarios there will be a tomorrow.