Wednesday, June 29, 2011

How Much Does a Sub-Optimal Lineup Cost the Giants

Almost everyday, without fail, when the Giants lineup is posted, there is something to complain about. Whether it is Aaron Rowand getting a start against a right handed pitcher or Miguel Tejada hitting second, there is usually something there to get under your skin.

The standard thing to say is that the day-to-day lineup isn't that big of a deal. For most teams, you are looking at about ten to 15 runs difference over the course of the season, but what about the Giants this year?

Julian Levine of Splashing Pumpkins brought this up in his defense of Bruce Bochy, so I thought that I should look a little deeper into how much the sub-optimal lineup costs the Giants.

To run this little experiment, I used the the rest of the season ZIPS projections from Fangraphs and the lineup analysis tool from Baseball Musings. I used three different lineups for my analysis; the hair pulling lineup, the okay, I guess this works lineup, and the best lineup that the tool could come up with. (I substituted in Mike Fontenot because I figure he will get the lions share of playing time at 2B when he comes back for all the lineups).

Here are the results:

Hair Pulling - 3.803 per game
Okay I Guess - 4.022 per game
Optimal - 4.130 per game

Next, we will make some assumptions and compare the results to optimal.

100 percent "Okay, I guess" lineup the rest of the way would make the team trail the optimal lineup by about nine runs when all is said and done. This is pretty much in line with what we would expect, that lineups don't make that big of a deal because over a full season, that would be 17 runs total, so not too bad.

If we change it to 80 percent "Okay" and 20 percent "Hair pulling" out, we get a difference of 12 runs. That is getting a little bit bigger. If we had done something like this over the whole season, it would come out to 25 runs. To me, this is probably the most likely scenario as Bochy tries to keep people engaged through out the remainder of the season.

Finally, lets look at the doomsday scenario where we have 65 percent "Okay" and 35 percent "Hair pulling." This mix would give us difference of 15 runs for the rest of the season and would have meant 30 runs over a full season. That would be roughly three wins, which is a pretty big deal.

This is obviously not a perfect experiment, as there are a number of assumptions and many variables not taken into account, such as defensive performance or platoon splits that effect lineup construction.

That being said there is probably a pretty decent chance that the repeatedly sub-optimal lineup penciled in by Bruce Bochy has cost the Giants a fairy significant amount of runs and that it will continue to as the season goes on.

Become a fan on Facebook and follow The Crazy Crabbers on Twitter. There is lots of good stuff going on there. Sign up for our free newsletter!

No comments:

Post a Comment