Game 4 Starter...

...if the Giants are up 2-1, i.e. they don't sweep and aren't down 2-1.

The potential game 4 starter for the NLDS between the Giants and Braves has roster implications. Why? Madison Bumgarner is likely to make the roster regardless of whether or not he gets the nod from Bochy. On the other hand, if Barry Zito does not get the nod, the likelihood he makes the 25-man is far less assured. With the weapons Atlanta has – Heyward, McCann and Hinske – a pen with a few lefties seems like a swell idea. But something tells me that Affeldt and Runzler would be far more effective in relief situations than Zito.

So who should get the start? I think most, if not every single fan, would say that it should be the kid: Bumgarner. But with the money Zito’s making there are some unfair biases, probably, and certainly some unjust hatred – it’s not his fault Sabean or McGowan thought he was an ace.

The Breakdown (with an assist on stats from Fangraphs):

Barry Zito logged about 198 innings as a starter. During those innings, he accumulated a WAR of 2.1 or 2.1 per 200 innings. He struck out 6.77 per 9/IP and walked 3.79 (ratio 1.79). His final WHIP was 1.34 and he yielded .90 HR/9. His .285 BABIP was near his career average (.276). He induced a ground ball only 36.1% of the time. His ERA was a respectable 4.15 and his FIP finished at 4.25. He was able to do all of this with an 85.7 MPH fastball – which was declining throughout the season – and averaging almost exactly 6 innings per start. All in all, Zito was a serviceable major league starter if stored at the back end of a rotation.

After being promoted, Madison Bumgarner logged 111 innings, during which he collected a WAR of 2.0 or at a rate of 3.6 per 200 innings. Bumgarner punched out batters at a pretty insignificantly higher rate than Zito, 6.97 (i.e. +.20 per 9 IP). But – and this is a huge but – he walked only 2.11 per 9 IP (K/BB ratio of 3.31). This is significant, especially when you consider a huge strength of the Braves, i.e. their ability to walk at a high clip – over 10% of the time. His WHIP is slightly better at 1.31, and also his HR/9 (.89). Other factors which I believe bode well for him, and legitimize his worthiness for the nod, are his BABIP and ground ball rate. Madison has given up a .322 average on balls in play, significantly higher than that of Zito. We don’t have enough data to understand where his true talent level is in terms of BABIP, but I’d venture to say it will probably end up closer to league average or around .300. Perhaps he’s been just a bit unlucky. He’s also been significantly better at getting the ground ball (45.1%). His final ERA was a really nice 3.00 and his FIP 3.66, all while averaging 91.3 MPH fastball velocity and at increasing rates as the season neared its end. Lastly, he averaged 6.17 innings per start (just ½ a batter more than Zito).

In this case, the fans have it exactly right. The way in which they arrived at their decision for the Game 4 starter was more than likely decidedly different, but no matter.

One small consideration may be the added innings to Bumgarner’s young arm. I don’t think this will be a factor in their decision, though. We’re talking about maybe 3 starts if the Giants go to the World Series, and thus a maximum of eighteen (18) innings or so. If they were going to back off the kid – whom is throwing better and harder now than at any point this season, I should point out – they would have already. As management knows, fans know, and players – especially Aubrey Huff – postseason doesn’t necessarily come around all that often, and when it does, you need to strike when the iron is hot.

If I’m Bochy, I’m taking Bum’s ability to throw strikes over Zito’s prior playoff experience.*

* Without Zito, all four starters would be making their postseason debuts.

UPDATE: Bruce Bochy acknowledged that the game four starter would either be Bum or Lincecum, depending on whether the Giants were up or down in the series.  Phew!