Benching a superstar is madness. Or is it?

Would you ever consider benching one of your top draft picks for 20 percent of his games?

If your initial response resembles something like “Are you nuts?,” consider how Adrian Gonzalez performed in 2008 against right-handed pitching and left-handed pitching:

Adrian Gonzalez vs. righties: 381 AB, .320 AVG, 24 HR, 81 RBI, 76 R, 0 SB
Adrian Gonzalez vs. Lefties: 235 AB, .213 AVG, 12 HR, 38 RBI, 27 R, 0 SB

As seen above, “A-Go” has a horrible time hitting for anything other than a bit of power against left-handed pitchers. He’s not alone: Grady Sizemore, Josh Hamilton, Carlos Pena, David Ortiz and Brad Hawpe are just a few left-handed stars who hit significantly better against righties than they do against lefties.

As baseball fans, we’ve grown up with a simple truth: Lefties hit righties better, and vice versa. In the major leagues, managers exploit platoons all the time to deal with this phenomenon. In some cases, fantasy managers may wish to do the same.

On my blog, I’ve explored better use of reserve spots for a fantasy bench, including the use of platoons.

In most instances, this might mean shuffling left-handed players with noticeable splits in and out of the active lineup to maximize production from a single roster position. Players like Luke Scott, Skip Schumaker and Shin-Soo Choo should come relatively cheap in drafts and auctions. Rostering two, and interchanging them in a lineup depending on a given day’s match-up, is a fairly effective way to not only increase production at one spot, but also save some auction money or a top draft pick to fill another hole. The sacrifice that a fantasy baseball manager gives for this luxury is a bench spot, some roster flexibility, and six months of getting up each morning to check on pitching match-ups.

That leads us to an interesting and almost unthinkable question: Might we try the same thing with a superstar and improve upon his production?

To do an analysis, we’ve had to make some basic assumptions after some research on the percentage of right-handed starting pitchers in baseball (about 70 percent) and the average number of times a left-handed batter could expect to see left-handed and right-handed relief pitchers in a single game.

From that, we deduced we’d be playing a star like Adrian Gonzalez approximately 80 percent of his starts (some vs. lefties out of necessity). Given that, we’d expect to retain about 94 percent of his stats vs. right-handed pitchers and about 57 percent of his stats vs. left-handed pitchers. We’d be using a cheap platoon partner like Lyle Overbay (.291 AVG, 15 HR, 61 RBI, 61 R, and a steal vs. righties in 2008) for the rest, accumulating approximately 26 percent of his stats vs. righties and about 10 percent of his stats vs. lefties.

We didn’t adjust for strength of pitching, but what’s the outcome?

Adrian Gonzalez in 2008: 616 AB, .279 AVG, 36 HR, 119 RBI, 103 R, 0 steals
Our platoon: 610 AB, .289 AVG, 34 HR, 115 RBI, 104 R, 0 steals

In other words, pretty close and unfortunately, inconclusive. Still, the strategy holds promise, especially if we see Gonzalez’ stats regress this season or if we could use the waiver wire for better replacement stats.

Performances vary year-to-year and small sample sizes can be especially tricky to navigate, unbeholden to the law of averages.

To be honest, we don’t think we’d have the guts to actually try this out for real on any player rated as a fifth-round pick and above. But for players like Carlos Pena or Brad Hawpe? No hesitation whatsoever.

The Incompleat Starting Pitcher
The end of the nine-inning start and how we got here.

Print This Post
Sort by:   newest | oldest | most voted
bsball
Guest
bsball

Two notes of caution:

1. There’s a fair amount of random variation in platoon splits for individual players, especially if you are looking at only one year’s worth of data.  It may be better to use an average for all players rather than using individual players’ rates.

2. You are not going to get the full platoon advantage using this strategy because starters don’t pitch the whole game.

Eriq Gardner
Guest
Eriq Gardner

Good points, bsball.

Yes, I’m aware of the random variation. That’s why I made the point about small sample sizes.

And, it’s true you are not going to get the full platoon advantage. That’s why I was careful to do some work on the amount of plate appearances that lefties would be seeing vs. both LHP and RHP relievers. It was factored into my calculation above.

Millsy
Guest
Millsy

Brad Hawpe’s splits last year were .282 vs. Lefties and .283 vs. Righties, with a very slight difference in HR/AB against the two.  Could be selection bias, though, in that his manager only played him against crappy LHP.  Also, it’s one year…but I just felt like making a weak point.

Eriq Gardner
Guest
Eriq Gardner

Hawpe had a smaller split in his average in 08, but
-We have a much bigger sample set to judge BH upon
-If you look at his OBP split, it’s 41 points
-you’re right that his LHP weakness probably influences the way his manager uses him. One piece of evidence is 70 RBI vs. RHP vs. 15 vs. LHP

jw
Guest
jw

Last year I platooned Adrian with Carlos Pena in the second half of the season after an owner kicked Pena to the curb. Adrian on the road against lefties was an easy call to give him the night off. I didn’t do the numbers on what kind of production I missed from Adrian and I also played Pena at U a bit, but I seemed to remember gaining some HRs out of the deal.

jw
Guest
jw

Oops…I meant at home versus lefties was the cue to sit Adrian.

bpasinko
Guest
bpasinko

Makes sense.  Having a useful bench is underrated.  You can get top production for half the cost if you have good lefty/righty platoon mashers on your bench. 

What about a bench base stealer.  If I play Michael Bourn when guys have days off, is that stupid.  Is it not worth having a sb threat take the spot of a Luke Scott, or some guy who is clearly better against one side?

Does the perfect bench have both, or is there often not room for it.

Millsy
Guest
Millsy
I like the idea of the bench base stealer, depending on the format.  I play in a H2H league with 8×8 roto categories, rather than points.  I find steals to be extremely expensive for the most part for guys that are viable starters.  However, in an league like that, variability can account cause you to tie a top team 8-8 in one session, then lose 15-1.  That 15-1, in my experience, can cost you a playoff birth.  However, if your starting lineup has guys like David Wright or Orlando Cabreras…low double digit steals…and you see that you’re doing well in… Read more »
Mark Harrison
Guest
Mark Harrison

I think one thing missing here is the fact that you will have other players on your team other than the star who may also have a poor matchup based on the splits and who would be more useful to platoon.

timbo
Guest
timbo
I play in a roto league with daily lineup changes.  I love having one speedster on my bench.  I pay attention to the SP and C that allow the most amount of SB.  Do you think Posada will throw out more than….2 people this year?  I don’t.  When the speedster is facing one of those combos, I’ll plug him in for any other normal starter that has a bad matchup.  I always seem to finish higher in the SB category than people would expect.  You can maximize the SB output of a Taveras/Gomez/Bourn/Pierre without getting hurt too much by their… Read more »
wpDiscuz