keyboard_arrow_uptop

There was a time, not too long ago, when my bench spots were all occupied by pitchers. The benefit to that was obvious—I could stick lower-end starting pitchers there to take advantage of quality matchups and stash relief pitchers that are good handcuffs to potentially shaky closers. However, with the popularization of deeper benches and daily transactions/lineup changes, my thinking on this has shifted slightly. There is now no shortage of leagues out there with benches that go five players deep or larger, and the changes in format haven’t led to a different way of thinking about how best to deploy a fantasy bench. Until now.

The exercises below show the benefits of using one of your bench spots specifically to platoon one of your final offensive players, using the 2012 season as the example. The guidelines of the exercises are simple. For each scenario, I took two players who were drafted outside the top 200 last pre-season and set a fixed schedule of when one would be in the lineup over the other—leaving no room for subjectivity. The only exceptions to this were when one of the players was not in the starting lineup (out) or one of the players was participating in a doubleheader (in). Then, I went back through the 2012 game logs to determine the actual statistics and value earned out of this “alternative arrangement.” But before we dive in, we have to set a baseline of value for the roster spots.

Let’s assume, for the purpose of this demonstration, that this owner would have gone with a more standard strategy of taking one full-time offensive player and one flier on a starting pitcher with his/her two final draft picks. Let’s also assume that we’re dealing with a 12-14 team mixed league with five bench spots. The 200th player drafted in a league this size yields a positive return on investment if he earns $4 over the course of the season. Let’s even say that the owner is particularly prescient and is able to squeeze $6 of value out of the starting pitcher by sitting him for a few harsh matchups; even this brings us to a $10 overall value for the two picks.

Now it’s time to see how the arrangements stack up. We’ll start first with a platoon based on home ballparks, and then move to one based on actual platoon splits.

The Ballpark

Let’s dive into the arrangement with a lower level of difficulty: a platoon of advantageous home ballparks. There are a couple of parks to that exaggerate the statistics of the players that call them home. For this exercise, I chose one player who plays half of his games at Coors Field and one who plays half of his games at Chase Field. Dexter Fowler went into 2012 with a decent amount of buzz, but still saw his ADP sitting at 214 overall. Jason Kubel was generating similar sleeper hype, as he was going from a park that stifled left-handed power to one of the gentlest environments for it. Kubel’s ADP was just a few picks ahead of Fowler’s, at 210 overall.

Now, before we go any further into this part, let me clarify one thing. In choosing players for the home/road split, I put much more weight into park factors and team performance than individual performance, as there tends to be more noise in an individual player season. So, while Fowler didn’t have a huge split in 2011 (811 OPS at Coors versus 782 OPS at home), I would expect his split to be in line with the team split (OPS 113 points higher at Coors) going forward. Finding the right player, unless his skill set doesn’t translate to being elevated by an high-offense home park, is less important than finding the right park.

Based on expected performance, I set the pre-season schedule at the following:

  1. If Fowler is at home, he plays.
  2. If Fowler is on the road and Kubel is at home, Kubel plays.
  3. If both are on the road, Kubel plays.

When you use this formula to determine playing time, the final result of the lineup position over the course of the 2012 season was a .290 average with 31 homers, 98 RBI, 107 runs scored and 6 steals in 582 at bats. Here’s how it broke down by each player involved:

 

AB

R

HR

RBI

SB

AVG

Dexter Fowler

284

52

10

41

5

.320

Jason Kubel

298

55

21

57

1

.262

Combined

582

107

31

98

6

.290

Comp (Matt Holliday)

599

95

27

102

4

.295

Matt Holliday ended the 2012 season as the 9th best OF for fantasy and earned approximately $18 in a league of this size. The combination of Fowler and Kubel (each of whom finished outside the top 40 outfielders in 2012) provided $19 in value.

Net profit: $9

Conclusion: Overall, this was certainly a success, but not a blow-away success. And, you’d expect this because it’s both an easier strategy to identify and because players who play in advantageous home parks are generally overpriced on draft day.

Possibilities for 2013: The nice element of this strategy is that there are always new players that have favorable home parks when the ones who have good seasons become too expensive. If I’m trying to pull off this arrangement in 2013, I’d be looking at players like Tyler Colvin, Lance Berkman, Ryan Ludwick,, or Cody Ross. And, if you’re looking to try it in a deeper league, there’s always Nolan Reimold and, dare I say, Travis Hafner.

The Platoon

The second exercise, which takes advantage of statistically relevant player splits between facing right-handed and left-handed pitching, requires a little more attention to detail. While it’s easy to set your lineup for days at a time using the ballpark split, when you’re dealing with opposing starting pitchers, the scheduling can get a little fickler. What I looked for here were players who had a career-long track record of hitting significantly better when given the platoon advantage.

In this case, I found two players who were both being drafted outside the top 250 in 2012. Garrett Jones has been so terrible against southpaws that it has overshadowed his impressive ability to hit right-handers. Case in point, his career 852 OPS against righties is higher than the 2012 marks for Paul Goldschmidt, Jay Bruc,e and Adam Jones, while his 590 OPS against lefties is lower than Rey Ordonez’s career mark. Cody Ross seemed like the perfect platoon partner, as his 926 OPS versus left-handers for his career was more than 200 points better than his mark against righties. Last March, Jones had an ADP of 279 overall, while Ross was being taken a couple of rounds later at 312 overall.

Based on their previous history of performance, I set the pre-season schedule at the following*:

  1. When Ross faces a LHP, he plays.
  2. When both Ross and Jones face a RHP, Jones plays.
  3. When Ross faces a RHP and Jones faces a LHP, Ross plays.

*The only wrinkle in this plan was that Ross went on the disabled list in mid-May and missed a month of the season. To account for this, I replaced Ross with another lefty-masher who was widely available on waiver wires at the time: Scott Hairston. Hairston played the Ross role exactly, with no derivation from the original plan.

When you use this formula to determine playing time, the final result of the lineup position over the course of the 2012 season was a .277 average with 43 homers, 128 RBI, 99 runs scored and two steals in 617 at-bats. Here’s how it broke down by each player involved:

 

AB

R

HR

RBI

SB

AVG

Cody Ross

279

47

19

58

0

.280

Scott Hairston

32

7

3

7

1

.344

Garrett Jones

306

45

18

61

1

.268

Combined

617

99

40

126

2

.277

Comp (Josh Hamilton)

562

103

43

128

7

.285

As you can tell, this one is a little more jarring. Hamilton ended the 2012 season as the fourth-best outfielder for fantasy and earned approximately $30 in a league of this size. The combination of Ross, Hairston, and Jones (each of whom also finished outside the top 40 outfielders in 2012) provided $27 in value.

Net profit: $17

Conclusion: Now this I would consider a huge success. In the end, not only was it cheaper to use platoon splits to form a “super outfielder” than the ballpark splits, but the strategy also provided more than double the profit. And, the best part is that this strategy is so underutilized that even these exact two players are available at similar ADPs this spring as they were last spring.

Possibilities for 2013: If you want, you can do the same Ross/Jones platoon without paying a premium for it. But, if you’re looking for other names to go with, try Jonny Gomes or Scott Hairston against southpaws. And, for the strong side of the platoon, while no one’s quite as perfect for the role as Jones, there’s always Adam Lind or Matt Joyce—who will likely cost less.

You need to be logged in to comment. Login or Subscribe
captnamerca
2/21
I like the way this article lays it out. Very nice. I think I'll give the Garrett Gomes monster a try at the end of a shallow league draft.
bmmillsy
2/21
I had this ready to go this past year with my $9 Reddick-Gomes-Smith combo and then Reddick just decided to mash everything so I wasn't willing to take him out most of the year. Still might have been a decent rule for rotating these guys that outpaced Reddick alone, though (particularly since it has both AVG and OBP as categories). (8x8, 20-team, 30 man roster, 3 DL slot, with 3 to 8 minor league slots separate from the 30, so Ryan Braun is valued at about $85 here).
fabiopao
2/21
Great "moneyball" article, definitely worth a try this year
TheArtfulDodger
2/21
Love this strategy. Great article
jimcal
2/21
Thank you for doing this work quantifying the effect. Platoon strategy has been discuss for long, but never (I yet have) seen someone was able to quantify this. The follow up question is, Folwer's home stats seems short. Dexter Folwer's 2012 Home split was an astounding .332/.431/.553 in 244 AB. If default Folwer home plays, he should have the same line as his home split. Did I miss something?
bretsayre
2/21
Thanks for pointing that out. Originally, I had thought it was just poor performance for Fowler on the days he was in the lineup on the road, but it turns out that my spreadsheet wasn't pulling in his double-header from May 28th. That day, he had 7 hits, 1 homer, 3 RBI, 5 runs and a steal -- so I was actually understating my point. With that day properly pulling in, the overall stat line for the platoon jumps to a .290 average, 31 homers, 98 RBI, 107 runs scored and 6 steals. I'm going to have this fixed in the post (as it also bumps the profit up).
cgraham73
2/21
Great insight on this Bret. During your research or from your knowledge base can you think of any other good candidates for this type of split? I have LONG used my bench spots to stream starters through and am ready to try a different approach this year so looking for some new ideas...This is a good one! Thanks, Chris
bretsayre
2/21
The best names to keep an eye on for this are the ones mentioned in the Possibilities for 2013 section. For home ballparks, that could be Tyler Colvin, Lance Berkman, Ryan Ludwick or Nolan Reimold. For the L/R splits, it could be Jonny Gomes, Scott Hairston, Adam Lind or Matt Joyce.
lyricalkiller
2/21
It is fixed
hankfiddich
2/21
Not sure how much this would effect the numbers but I'll throw it out there anyway. Do the L/R splits include PAs against relief pitchers or pinch hits and/or substitutions. Although I agree that this strategy would work, I wonder if the actual production a fantasy manager could get would be reduced because Jones/Ross/Hairston are mid-game replacements or are facing those pitchers due to pitching changes. Not sure how granular you could get on that.
bretsayre
2/21
Any performance against same side pitching within those games is already factored in to the calculation. You'd expect to find slightly larger deviation between true L/R splits and ones between games started by LH/RH pitchers, but there's no way to use that specific data for fantasy. But in case you're curious to see what it looks like, here are the differences for each of the L/R split players between the two: Cody Ross vs all LHP: .295/.373/.636 Cody Ross in games started by a LHP: .286/.357/.596 Garrett Jones vs all RHP: .289/.332/.556 Garrett Jones in games started by a RHP: .276/.316/.528 But for fantasy, this really only affects batting average -- the raw stats either remain just as strong, or get slightly stronger due to additional plate appearances.
jcutiger
2/21
So I just went through this last night with those exact players. Col/Ari have a great schedule setup this year for this strategy. Approximately 55% of the time only one of them is at home. They are both on the road 14% of the time. Ari/Cin is has almost the same split.
davescottofakron
2/21
When both guys are going against a same-side pitcher, isn't it worth considering the bench for both of them. Even a blank spot for a day might be better. And please define terms for us dummies. Took me a while to figure out what ADP meant. Sorry for the nit!
bretsayre
2/21
You could certainly do that, but I wouldn't advise it. Playing time is playing time, even if it's not an ideal matchup -- and I'd almost never sit a hitter for an empty roster spot.
jcutiger
2/21
I agree. You may be helping in 1 catergory but hurting yourself in up to 4 categories.
jonjacoby
2/21
ok, both of my leagues allow only for weekly transactions and wondering if this is workable for that I guess the Home/road is easier given Lefty/righty there is a greater chance for last minute pitching changes. Using the home/ road wonder how I would work it if each team is home for part of the week and on the road for the ther. would the answer be to start the player whose road series is at the better pitcher's park?
bretsayre
2/21
I don't think something like this is workable in a weekly transaction league because it gets too subjective and you miss too much value. Even the home/road platoon would struggle to give you the same type of results, as you'd likely be missing about 30-40 games of the advantage on average by being locked in for the entire scoring period. The fact that you don't miss a choice matchup really drives up the value of the exercise.
serviceoutrage
2/24
One thing to note is that it may actually be more advantageous to get two lefty-hitting, righty-mashers for a platoon, as you are not running a real baseball team and there are more opportunities to carry the platoon advantage against a starting pitcher for a lefty.
bretsayre
2/24
The issue with that is that lefty mashers are generally better from a per game perspective than righty mashers (or at least the ones that are available late enough for the advantage to be optimized).