Jonathan Bales is the author of Fantasy Baseball for Smart People—a guide designed to help you profit on DraftKings.
One of the reasons I love GPP play is because the landscape is constantly evolving. I think you can make a decent argument that there’s no true everlasting “optimal” strategy in tournaments – just what is optimal at the current moment based on what others are doing.
Of course, the most popular strategy in daily fantasy baseball tournaments right now is stacking.
It takes all of about two seconds to realize why stacking – particular large five and six-man stacks of MLB offenses – is so popular in GPPs. To win a tournament, you need to score a lot of points. You need a lot of upside. Stacking leads to that offensive firepower.
This data is pretty clear. With the strong correlation between teammates’ fantasy scoring in daily MLB, there’s a whole lot of merit to stacking if your goal is to maximize upside. In terms of pure points, you’re going to have a difficult time beating a six-man stack over the long run.
So why would I ever not stack?
With results this extreme, it might seem like there’s never a time when it’s okay to forgo stacking. If you need a lot of points to win tournaments and stacking can lead to a lot of points – more points than not stacking – then it follows that stacking should be a smart move in tournaments, right?
I, too, am a fan of stacking in many situations, but I think the idea that it’s a necessary strategy – that you can’t win without it – is wrong. There are two basic tenants behind my thinking:
1) The strategy that leads to the most GPP wins isn’t necessarily the one that’s best for you.
2) The goal isn’t to maximize points at all costs. The goal is to win.
Let’s start with the first. Most tournaments are taken down by stacks – maybe not always six-man stacks, but usually at least four or five-man groupings. I think it’s easy to see which strategy leads to the most wins and then extend that to justify using such a strategy yourself, but I think that’s a mistake that ignores baseline rates.
As an example, let’s say that 80 percent of the field is stacking in tournaments, on average, and we see that 70 percent of winning lineups have a stack. Without knowing that 80 percent baseline, it would appear as though stacking is very profitable and the best tactic to employ, when in reality non-stacks would be leading to more individual success after adjusting for how often they’re used.
Let me again mention that I think there are many instances when stacking is a very profitable strategy, and I do indeed stack in probably as many or more lineups than I don’t. However, I think the idea that stacking is so much more advantageous than not stacking, in all situations, is probably no longer correct. There’s a very good chance that it was correct perhaps even just 12 months ago, but recent shifts in public lineup construction have altered the expected value of stacking.
That brings me to the second point: the goal isn’t to maximize points, but rather to win. I think a lot of players treat these two goals as synonymous when they shouldn’t be. Yes, you need a lot of points to win a tournament, but I think it’s a better strategy to try to figure out ways you can win while scoring as few points as possible. And the way to do that is to fade the chalk – the most popular offenses that are stacked in a given night.
Now, I don’t think fading the chalk is always smart. Chalk plays are popular because they generally offer value, and maximizing value will generally maximize point-scoring. In being contrarian, however, we’re knowingly forgoing a little bit of value in order to see dramatically reduced ownership on our players, which ultimately allows us to be in position to benefit when others are wrong; when the visitor at Coors Field tanks, scores have a high likelihood of being lower than normal, which is ultimately great for those users who faded that team. This all comes down to how popular stacking is and how much lineup overlap it is creating. Are we at the point that stacking has become so popular that, although it maximizes points, it isn’t +EV due to the similarity in lineups? For the most part, no, I don’t think so, but I do believe there are perhaps ways to utilize elements of stacking and playing the chalk without running into a bunch of lineups that look just like yours.
The Future of Stacking
As DFS tournaments evolve, I think we’re going to see more and more unique strategies employed, many of which will be centered more around game theory and public psychology than player value. Due to how popular stacking has become in the past year or so, I think there are a few tactics that, even if not superior to stacking right now, are becoming more and more legitimate.
Once we reach a point at which tournaments are completely saturated with six-man stacks, not stacking at all (or using no more than, say, three players from one team) could end up being +EV. In this scenario, we’d basically be using a cash-game lineup in tournaments.
The primary downside is of course a lack of correlations built into the lineup, but not stacking does allow you to play chalk while still maintaining a unique combination of players. When you stack the Rockies or their opponent at Coors, you might be on the chalk, but you’re also creating a lineup that’s not at all unique, meaning you basically need to be perfect to win.
A similar strategy is to use two mini stacks (maybe a three-man and a four-man) of two chalk offenses. You’re acquiring some of the correlation benefits of stacking, maintaining a unique lineup, and still utilizing high-value batters. Right now, I think this is my favorite GPP strategy when I want to have exposure to offenses I know are going to be really popular.
It’s worth noting that I think this approach makes the most sense when you want to be on the chalk. If you’re going to be contrarian, there’s less incentive to forgo a larger stack because we know five and six-man stacks are the shit in terms of maximizing points. By fading the chalk and going contrarian, you’re necessarily competing with fewer similar lineups and thus don’t really need to get away from a full stack in order to differentiate yourself.
Bottom of Order
Finally, I think there might be some merit to using players who hit near the bottom of the order, but only in specific situations. This is similar to the decision to use a full or mini stack; mini stacks make sense as a way to differentiate when playing the chalk, but not as much if you’re already using a contrarian offense. Similarly, it could be smart to use a player or two at the bottom of the order on certain chalk offenses, but it’s probably not as viable if you’re using an offense you anticipate having lower ownership.
I’m not entirely sure what the future of tournament strategy will look like in daily fantasy baseball, but I do see some changes in what’s becoming “optimal.” There are a whole lot of really awesome aspects to GPP lineup construction that are in constant flux, but I think the one everlasting doctrine of daily fantasy tournaments is that the best players moving forward will be the ones who harness public opinion and are always one step ahead of the crowd.