Does the Brett Gardner/Marcus Thames Platoon Make Sense?

Written by Greg Fertel | 10 April 2010

It's hard to make judgments just four games into the season, but it seems evident to me that the Yankees do plan on employing some sort of platoon with Brett Gardner and Marcus Thames. In two of the team's four games, Thames has gotten the start when the opposing team has a left hander on the mound. At a glance, this might seem to make sense.

Gardner is, after all, a light-hitting outfielder; there's no arguing that. Thames, on the other hand, has fared pretty well against lefties over the course of his career, hitting them to a tune of a .256/.329/.514 line(.360 wOBA) in 696 PA.

With Gardner, we have a much smaller sample to work with versus left handed pitchers. He has had just 93 PA versus lefties in his major league career. Because of this, I'm choosing to add in his MLEs from 2007-2008 to expand the sample a little bit. Like always, I'm not sure how scientific this idea is, but this brings our sample to 298 PA and a .235/.298/.332 line with a .284 wOBA. The Book tells us that left-handed hitters need 1000 PA vs left-handed pitchers to show a true talent level. Without that many plate appearances against same handed pitching, regression towards the league average platoon split is needed.

Using the same method and including MLEs from 2007-2008, I come up with a .268/.363/.359 line versus righties for Gardner, which comes to a wOBA of .330. Using this set of numbers, Gardner would have an overall wOBA of .319 and a -14.2 percent split versus lefties, while the league average is a -8.6 percent.

To factor in regression I did the following equation for Gardner's numbers:(.142*298+.086*1000) / (298+1000).

This gives us an estimated platoon skill of 9.89 percent. Now, we can use this number and CHONE's 2010 projection for Gardner to predict how he will fare against lefties this season. CHONE projects a .335 wOBA, which would mean we'd expect Gardner to produce a .302 wOBA against lefties.

Now, onto finding an estimated platoon skill for Thames. The Book says that right handed hitters need 2200 PA versus LHP for that number to be used without regression. For Thames, at least, we don't have to use any MLEs.

Thames's career wOBA is .336, .326 vs. RHP in 929 PA, and .360 vs. LHP in 696 PA — a 10.1 percent difference. But he’s a righty, so we regress toward 2200 PA of the average (6.1%): (.101*696+.0611*2200)/(696+2200) for an estimated platoon skill of 7.1 percent. Using the CHONE projection of .328 wOBA, we’d estimate Thames to put up a .351 wOBA versus LHP.

Over the course of 150 plate appearances, Thames projects to be worth 5.35 more runs than Gardner versus left handed pitching.

Of course, hitting isn't everything.

Yesterday, we witnessed first hand that Thames is not a very good defender, so the question is whether his offensive prowess outweigh Gardner's defensive advantage. Using Jeff Zimmerman's UZR projections, Thames is forecasted to be worth -11 runs over the course of 150 games. In left field, Gardner projects to be something like +15 runs over 150, given his strong center field track record.

Over 35 games(similar to the 150 PA I used earlier), Gardner stands to save 6.07 more runs in the field than than Thames.

That is a significant amount over such a short period of time, but it sounds about right to me. The difference in range between Gardner and Thames is remarkably wide. The play we saw last night was an example of one that Gardner would make, and Thames turns into a double.

Overall, this gives Gardner a 0.72 run advantage over Thames, if we are just talking about starts against left-handed pitchers.

This platoon that the Yankees are employing doesn't make much sense. Sure, Thames's power against lefties is alluring, but his offensive value is not strong enough to compensate for his defensive flaws. Add in the fact that playing sporadically could affect each player's rhythm and hitting, and it makes less sense. I can't see an argument where starting Thames against lefties is the right decision. Gardner should be playing left field every day, regardless of who is on the mound for the opposing team. Thames is the team's fifth outfielder; it's time that they start acting like it.

For projecting the platoon splits, all of the information I used was via Matt Klaassen's article at FanGraphs, Estimating Hitter Platoon Skill.

Screengrab Credit: Joseph Pawlikowski/River Ave Blues