Every so often a MLB slate will arise where a very important game theory situation presents itself. Last night, White Sox pitcher Derek Holland faced off against the Minnesota Twins, and Holland had a fairly good situation as he was both cheap on DraftKings and had somewhat of a good matchup against the Twins who are capable of striking out a fair bit. This led to inflated ownership of him in GPPs, since people like both cheap value and good matchups. By going Holland, this also made it easy to roster at least one team playing in Coors Field, a dream scenario for most tournament players. While this strategy was ideal for maximizing overall value per salary cap dollar, there was a very good antidote for it, and it’s one that works quite well until enough people start doing it.


Imagine a see-saw, a child’s game where two kids sit on a wooden plank and one goes up and the other goes down. In MLB DFS terms, instead of children we are using the Minnesota Twins hitters and Derek Holland. As the Twins go up on the see-saw in performance, Holland goes down a corresponding amount. Compare this to a Coors game where you have two owned offenses generally with no pitcher on the other side of the see-saw, as pitchers are rarely going to be owned in Coors.

When we roster a team like the Twins (as we would with any other MLB team), we are hoping for:

  • Moderate to low ownership
  • A high percentile game (75% or higher) from most of their hitters in our stack

It’s a given that the times the Twins hit their average run projection of 4 that we’re not going to be doing that well in terms of winning a GPP. Of course the same goes for any team, as the fact is that a stacked team must run well above expectation to win, since it’s a near certainty that one of the other stacks on the slate will run at a high percentile.

The crystal clear benefit of using the Twins as a stack is that it leverages heavily against Holland, and it is very likely that your ideal scenario of a 75+ percentile result from the Twins will correlate with a 40 percentile or lower score from Holland. Keep in mind that it’s possible for the Twins to do well while Holland has a slightly below average game, since there’s always the chance that the Twins do a lot of late inning damage vs. the bullpen.

Keep in mind as well that for a cheap pitcher like Holland, much of his value comes in his win equity, which is 4 points on DraftKings. If he’s projected to get 14 points but 2 of those points come from his expected chance of winning, in the times he does not get the win he’s 15% worse. In lots of the games he does not win, the Twins probably scored some runs, and that is effectively how the see-saw works.

The importance of using this strategy against a good, but not great pitcher is that a guy like Holland is a lot more susceptible to a bad start than an elite pitcher, and he’ll also not get the kind of patience from his manager that a Clayton Kershaw would when he starts to falter a little bit.

On today’s slate, we’d like to target some bats against pitchers who are good, but not great. Generally these will be teams with run lines of 3.8-4.5 runs and will be in the top 25-50 percentile of all the teams’ run lines. They’re good, but not great, much like the pitchers they’re facing, and this will lower their ownership but also retain the good upside you want to actually win a tournament.

Before we get to the good pitchers, we’ve got a great pitcher in Clayton Kershaw at $11,300 DraftKings @COL. This is an interesting spot where I’d be looking to see-saw leverage COL bats and Tyler Chatwood (especially paired together) if I thought that Kershaw would be owned, which he probably won’t be. On the other hand, I’d like to play Kershaw if I thought COL bats would be owned, which they absolutely will not be. This is a situation where you can comfortably avoid going contrarian either way since there isn’t much leveraging to be done against your fellow competitors.

Whenever you’re looking to go contrarian and are on the fence between going one stack vs. another, ask yourself “How much leveraging is there to be had with this stack vs. another stack?” If your stack goes off vs a 20% owned pitcher instead of a 5% one, you’ve done irreparable damage to an extra 15% of the field and this drastically increases your chances of winning the tournament over the long run.

One pitcher that I think is quite good in GPPs tomorrow is Tyler Glasnow of the Pirates. He’s young, erratic, gives up plenty of walks and stolen bases, and is pitching in a favorable hitters park on the road. If that’s not enough to sell you on him, the D’Backs should have roughly a 4.6-4.7 run line against him.

His upside comes in the sense that the D’Backs hitters will be owned, and he’s got good enough stuff to shut them down. If you don’t believe me, look at what German Marquez did to the Dodgers at Coors Field just a few days ago. However, if there were any chance that Glasnow were going to be heavily owned (he will not be), I would hammer the D’Backs offense since they could easily steal a ton of bases against him in addition to scoring quite a few runs. Isn’t the see-saw effect fun?

In terms of a hitting stack, I am looking very early at the Phillies, since they have the 2nd lowest run line on the board as of now and face Tanner Roark of the Nationals on the road. The Nats bullpen is rested but still isn’t good, and Roark is another classic example of a good, but not great pitcher that will carry some ownership. Now, I’m not suggesting going Phillies stacks in 40 of 50 of your lineups, but certainly going them in even 5% of your lineups will give you more shares of them than what the rest of the field owns. Going out of your way to take small shares of some medium-high leverage longshots is a great way to reduce risk but increase reward, and that is the essence of being a winning MLB GPP player.

Previous post

5/5 DraftKings GPP Lineup Construction

Next post

Ghost Points in GPPs