MLB DFS Strategy Guide
I dive into the data of the last three seasons to give tips on MLB DFS strategy.
I am not as big of a DFS player as I used to be, but the DFS game was the original reason I wanted to make daily projections, and I did win $50K one time playing the game, so it has a special place for me.
In this post, I will go over what the data tells us about the best way to attack MLB DFS. This will all be centered around the DraftKings scoring system, so if you’re a FanDuel player, some of this won’t be as relevant, but plenty of it will be.
If you’re a paid subscriber here, you get my daily projections which project each player’s DraftKings points total for that day (as well as projecting all of the other stats that go into it along the way). Sign up today to get those projections and come along for the 2024 ride with us.
Pitchers
Here’s the scoring system DraftKings uses:
Wins have a heavier impact on FanDuel, and they give points for quality starts as well, so things are much different over there. We aren’t talking about that here though, so keep that in mind. Most of this doesn’t really apply outside of the DraftKings scoring system.
Bonuses
There’s a huge bonus for the rare instances of a complete game or no-hitter. Over the last three years, we have seen about 30 complete games per year, which makes the thing so rare that we shouldn’t even really consider it in our pitcher choices. Only one pitcher had more than two complete game last year (Sandy Alcantara), and he’s not even in the picture for 2024.
The win bonus is significant. Here is how SP scoring breaks down:
25th Percentile: 5.8 points
50th Percentile: 13.3 points
75th Percentile: 21.2 points
So the win bonus is worth about 30% of the average SP outing. Four points don’t seem like a ton when a lineup scores 100-150 points, but it’s a game of decimals, especially when you’re playing the tournament-style formats, so it matters. We can get a pretty good feel of how likely an SP is to get a win by looking at the money lines, so we should be favoring SPs who are heavy favorites. This is all pretty obvious, so I want to move on.
Strikeouts Are King
Here is how each individual stat correlated with DraftKings points since 2021:
So strikeouts have the highest correlation with overall points. Earned runs is right there with it, but we know that a pitcher has much more control over their strikeouts as compared to their earned runs.
The 10 highest SP point scorers (by DraftKings points scored per start) since 2021:
There is only one pitcher there with a strikeout rate below 25%. There are exceptions all of the time, you don’t have to only play high strikeout rate pitchers (sometimes salary doesn’t even allow for it), but if there’s one rule for picking pitchers, that’s it. Target strikeouts.
Hitters
There is more to analyze with hitters since we can start talking about lineup spots and stacking and the like. Here’s the DraftKing scoring system:
No bonuses, no penalties.
Category Correlations
In the same fashion as we did with pitchers:
Unsurprising result here. We want runs, homers, RBIs, and hits. This is a little different than pitching since these categories are all highly correlated among themselves. A home run always adds a run and at least one RBI, so you can see the HR+RBI correlation is very high (.654).
Since a home run encompasses two other things that give you points, it’s fair enough to just target home run hitters. If you get a homer from a guy, you’ve already exceeded your target from them.
Home run hitters often have lower floors since they tend to strike out more, but we don’t get hurt on DraftKings for strikeouts so it doesn’t hurt quite as bad. It’s a really low correlation with steals, although it did come up a bit if we just look at 2023 data (.302).
You should just trust the projections rely on the stacking stuff we’ll talk about soon when picking hitters, but if you’re in between choosing a home run hitter and a steals guy, the edge goes to the power bat.
Walks are a weird one. You do get two points for a walk, but it does prevent you from a higher-point event like an extra-base hit. When a hitter walks at least once in a game, they average 8.76 points, when they don’t walk in a game, they average 6.1. So the difference there is greater than the two points you get for the walk, and that comes from stealing bases and scoring runs subsequent to the walk. A high BB% hitter is better for the floor than for the ceiling. If we’re fishing for a 30+ point outcome, taking a walk isn’t a great way to get there. But two points is good (and a walk is worth a bit more than two points since you can continue to score points after that), so we’re fine with walks.
Lineup Spot
First, average plate appearances per game by lineup spot (this is the average for the player that begins the game in that lineup spot, pinch hitters are not considered). I used 2022 and 2023 here since the universal DH started in 2022:
1: 4.48
2: 4.41
3: 4.30
4: 4.20
5: 4.06
6: 3.95
7: 3.81
8: 3.63
9: 3.47
If you’d rather see that over a 162-game season:
1: 726
2: 714
3: 697
4: 680
5: 658
6: 640
7: 617
8: 588
9: 562
So yes, the projection changes substantially if a guy you expect to lead off ends up at the bottom of the order.
So we know that the higher a player is in the lineup, the better. But how valuable is each lineup spot for fantasy points on a per-PA basis? We need to control for hitter skill here since typically your 1-4 hitters are much better than your 6-9, so we can balance it out by looking at fantasy points scored per total base. How much more valuable is a homer when it’s hit from the clean-up spot as compared to the lead-off spot? That’s the kind of question we’re trying to answer. Here is it:
DraftKings Points Per Total Base
1: 3.63
2: 3.60
3: 3.59
4: 3.61
5: 3.59
6: 3.57
7: 3.58
8: 3.63
9: 3.67
To plot that to see the trend more clearly:
It might be a bit surprising to see the nine-hole on top like that. But remember, this is per total base, so it’s not considered at all how many plate appearances the spot gets per game. Getting on base turns out to be quite valuable for run scoring when you have the top of the lineup following you. So we see the 8-9-1 spots leading the way here because there are a lot of very good hitters behind those spots, typically.
Considering everything all together, we can just say that the higher in the lineup you are, the better. The most important thing in scoring fantasy points as a hitter is getting plate appearances.
One easy way to take advantage of this is to find those “surprise” lead-off hitters or guys that hit at the top of the lineup when that’s not the norm for them. Their DFS price tag will be influenced by their lineup spot, so a guy that usually hits 9th moving up to 1st for a game or two makes a big difference, and they will almost automatically project to be a good value pick that day because of that.
Home/Road Advantage?
Baseball is weird in that the road team gets more chances to hit, since they will hit in every inning of the game, while the home team might not hit in the bottom of the 9th if they have the lead. Since 2021 the road team has 51.6% of the plate appearances.
However, in most of the games where the home team doesn’t bat in the ninth, that is because they have already scored more fantasy points than the road team (by virtue of them being in the lead), so the two things mostly even out. The road team scores 50.7% of the fantasy points offensively. That’s still more, but so slightly so that we shouldn’t consider it in lineup decisions (and take note that this is baked into projections anyway).
Lineup Spot Correlations
Stacking is a key part of MLB DFS. In a game with a running clock, points is a zero-sum game. If you score a touchdown in football, that means time came off the clock during that drive and nobody else scored a touchdown during that time. You took away a potential touchdown from someone else. Same with basketball, every basket takes some time away from the total clock, so if a player is not involved in a basket their team scores, that event hurts them for that game.
In baseball, however, scoring fantasy points does not take points away from your teammates. If you hit a single, you are no closer to the end of the game after that event as compared to before it. There is a positive correlation between players in the same lineup. If a lead-off hitter hits a double, that raises the expected fantasy points for the hitter that comes up next, since there is more opportunity for an RBI (+2 points).
So we build a lot of stacks in MLB DFS. We can only play five hitters from the same lineup at maximum, so it’s important to drill down the best lineup spot combinations. We can make a huge correlation matrix here.
What you have here is the correlation between every pairing of lineup spots. The higher the number or the greener the background, the higher the correlation.
So the winner is the the correlation between the #3 and #4 hitters. These are where most home runs come from, and I believe a lot of this comes out of the first inning. If the #4 hitter bats in the first inning, that means someone ahead of him got on base. That raises expected fantasy points in that PA. If the clean-up man homers in the first inning, it must be at least a two-run homer, and one-third of the time or so that will be driving in the #3 hitter. That doesn’t fully explain it, but it gets us in the arena.
The greenest boxes are all down the diagonal, meaning that hitters hitting consecutively in the lineup have the strongest correlations.
So if we’re stacking, we want to focus on those consecutive spots in the lineup. A 1-5 or 2-6 stack makes the most sense if you’re doing a full five-man. If you’re going to mini-stacks, I would lean towards 2-4, 3-4, 1-3 in general terms.
We shouldn’t completely ignore #9 hitters either. They are often really cheap, and they do tend to feed off of big games that come from the top of the lineup. It can make sense to include them in a top-of-the-lineup stack if the price is right (you will also get lower ownership on those #9 hitters as well).
To get more on this, we can go through each lineup spot and isolate when those spots had “big games” (I defined that as more than 15 DraftKings points). We can zoom in on those games and see how the other lineup spots did in those situations.
When the lead-off hitter scored 15+ DraftKings points, here is how the lineup looked:
So when the lead-off hitter scored at least 15 points, they averaged 20.05 points. That pulled the entire lineup up at least one point, with the biggest differentials coming from the #3, #2, #8, and #9 hitters in that order. The #5 was affected the least. So the further away you got from the #1 hitter, the less impact the lead-off man’s big game had.
When the #2 hitter had the big game:
It benefited the #1 hitter the most, the #3 hitter the second most, and then it helped the #5 hitter a bit more than the #4. That doesn’t seem to make sense at first, but we should remember that hitting a home run hurts the guy behind you. When you hit a homer, that next guy comes up with the bases empty.
This isn’t me making a case to not play consecutive hitters, because a home run helps the guy in front of you more than it hurts the guy after you. So I want to stack a 3-4 hitters because the #4 guy’s extra-base hits tend to add points to the #3 guy, which outdoes the fact that the #4 guy might get hurt if the #3 guy goes deep.
Let’s do the #3 hitter:
Here we see the biggest help being again to the guy right before the #3 hitter in the lineup, but the player that was second-most benefited is the next guy up, the #4 hitter. So much of this has to do with the first inning, I think. If a #3 guy has a big game, he probably scored points in the first inning, which means the #4 hitter got to come up in the first inning, quite likely with men on base.
Let’s do one more and skip to the bottom of the lineup, here’s what happens when the #9 hitter has the big game:
The biggest boosts go to the #7 and #1 hitters in these cases.
No matter what spot we pick, we see every hitter in the lineup going up by at least one point when someone there has a big game.
So yes, stacking makes a ton of sense, and you should do it in your lineups - especially your tournament lineups. When you’re trying to beat thousands of other people, it’s a good idea to leverage this correlation. Some of the advantage there is offset by the fact that this is well-known, and most lineups in a tournament will be stacks of some sort, but just because most of your competition is playing smart doesn’t mean you should play dumb.
I don’t have the data to back this up, but the best stacks would seem to be 1-4 hitters from road teams in lower-ownership spots. This doesn’t mean stacking the Rockies in Los Angeles against Tyler Glasnow because nobody else will, but maybe it does mean stacking the Pirates in Cincinnati against Hunter Greene.
We will get into individual stack ideas when the season gets rolling and I start writing up some DFS pieces, and we’ll certainly keep all of this in mind when we do it.
I think that will do it for this post. To do a recap:
Target high strikeout rate pitchers
Target home run hitters
Stack offenses
The bigger the better if you’re going for an outlier ceiling score
Keep your hitters close together in the lineup
I don’t think I broke any new ground here, but I was able to put some numbers on things, and I think this will be useful for the more novice players out there.
This was all for free, but become a paid sub today to get my full daily projections and DFS slate write-ups when the season gets rolling!