RotoAcademy Preview: Non-Traditional Stacking and Ignoring the Obvious
Last year on DraftKings, it became evident that the six-player stack (Note: DraftKings has now changed it to maximum five players) was becoming more and more popular as the year went on. A hot topic seemed to be if stacking is still a worthy strategy with everyone doing it. That question could easily be translated into “is there still an ‘edge’ to be gained by employing this strategy?” I think the answer is “yes,” because of how much upside it can provide. With that said, I do think that we should break away from the traditional form of stacking as the strategy becomes more popular.
During this NFL season, the term “double stack” became one that I heard thrown around more often on podcasts and in articles than I had in the past. There seemed to be an uptick in the amount of lineups that would roster a QB and two of his pass catchers. Taking it a step further, we also saw people stacking offenses and taking a QB, WR, and RB from the same team. They were absorbing the positive correlation between the QB/WR to help balance the negative correlation between a RB/WR or QB/WR.
What I mean when I say the “the traditional stack” is when we roster the four to five players in the top five spots of the batting order. This strategy has been growing in popularity in the years past. When we are tasked with attempting to win a GPP, we want to access upside while still differentiating ourselves from the pack. Without throwing the stacking strategy out of the window, there are still some ways that we can accomplish this. Some of these ways began to show themselves last year.
Editor’s Note: This is one of the many valuable DFS lessons that can be found over at RotoAcademy. Click here to browse through all of our free/premium offerings and improve as a daily fantasy sports player!
Breaking Away from the “Traditional Stack”
There are a few different things that we can do to break away from the traditional five-man stack. The first and probably most obvious thing would be to abandon the stacking idea entirely. I think that is too extreme and unnecessary on most slates. It would be very difficult to construct a good lineup that has no two players from the same team. Also, I think that it limits our upside a bit too much to consider purposefully avoid using any stack.
One way is to abandon the five-man stack is to construct a lineup with two mini-stacks. We could roster four players from one team, three from another, and one from another. That would look something like this:
C – L.A. Angels
1B – L.A. Angels
2B – L.A. Angels
3B – NY Yankees
SS – NY Yankees
OF – NY Yankees
OF – NY Yankees
OF – NY Mets
There are several different ways we could do this. We could do three from one team, three from another, and two from another. We are decreasing our probability of hitting on those offenses when we compare the probability to stacking one team completely, but we are allow ourselves some separation from the field.
That is one way to differentiate from the five-man stack, but there are going to be times where we want full access to the upside that a particular offense has on a given night. There are still ways to utilize a five-man stack and gain some separation from the other lineups that will be rostering five hitters from that team.
For good reason, the most popular batters to stack are the ones batting in the one through five spots in the order. As seen below, those hitters perform better than the final four spots in the order. One through five are the only batters who on average perform above their salary-based expectation.
The most popular teams to stack will be the most popular for good reasons, like being in a favorable matchup or playing in a favorable ballpark. Those things are factored into the Vegas totals. Take a look at how each spot in the order does when the team has an implied Vegas total over 4.5 runs.
Still, the top of the order is where we will find the most upside, and the most often. The bottom half of the order sees a bump, as well.
If I am targeting a team with an implied run total of 4.5 runs, I am not targeting them with the hope that they score four or five times. I am playing those batters with the hope that they crush their implied team total. If that situation plays out than we are going to see additional at-bats for those players in the bottom half of the order.
I propose that we break the stack. Rather than rostering the hitters in the one through five spot, what if we rostered something like this?
Batter in spot 1
Batter in spot 3
Batter in spot 4
Batter in spot 5
Batter in spot 6
Or maybe,
Batter in spot 1
Batter in spot 4
Batter in spot 5
Batter in spot 7
Batter in spot 8
I could go on forever, but I think that we all get the point. If the public is going to stack a team that we also want exposure to, I think it makes sense to stagger the stack. We are taking the risk of fading some of the better batters in the order in lieu of some riskier hitters, but I am not worried about that. Again, I stack teams with the hopes that they go off and absolutely crush their team totals. If that happens, the batters in the bottom of the order are going to benefit from extra plate appearances.
The typical correlation would be that if the lead-off batter gets a hit, we also want to have the guy that will be driving him in. The ninth hitter will not be driving the lead-off guy in, but he will benefit from the hit in the sense that it is getting him one batter closer to an extra at-bat. To give you an NFL example, it is very unlikely that a wide receiver is going to throw a touchdown to a running back. You can feel safe to bet that we won’t see any positive correlation between those two positions, in that regard. However, if a wide receiver catches a ball for 60 yards and is tackled at the 10 yard line, he did just give our running back the possibility of gaining some red zone work. That won’t show up in a correlation chart, but the wide receiver did just assist our running back. This works even more so in MLB. Any positive performance by a hitter is aiding every other hitter in that order by getting them one place closer to seeing an additional at-bat.
This is much easier to do when rolling out multiple lineups, but what if you are rolling with a single bullet? I think we take the chance and break up our stacks. I will agree that rostering the top of the order is probably the most optimal stack, but I will ask you this question: How many people need to be utilizing the most optimal method before it becomes sub-optimal?
Ignoring the Obvious in the Vegas Totals
As previously mentioned, I am not targeting an implied team total of four runs for my hitters with the hopes that the team scores exactly that amount of four runs. Also, I am not targeting a pitcher who is facing off against a team with an implied run total of three with the hopes that he allows three runs. I am using these as a measure of upside. I want my offense with an implied run total of four to score 10 runs and I want my pitcher facing the team with an implied run total of three to allow zero runs.
Following the advice of the first article, you are probably going to begin your daily research by looking at the information given to us from Vegas. Naturally, some things are going to stick out to you. Those things probably stuck out to me, as well. They probably stuck out to every person who is using information from Vegas to help build lineups. Similar to stacking, we are not being unique or sneaky by utilizing Vegas. How can we still use the information they are giving us without falling into a trap of letting that information lead us into the same roster construction as everyone else.
For example, let’s say it is an eight-game slate and two of the teams have an implied run total of 4.8. The next highest run total on the night is a team with an implied team total of 4.0.
As seen above, there is only a 0.28 difference of how hitters have performed in relation to their salary between an implied run total of 4.0 and one at 4.8. When utilizing the totals as a measuring block of upside rather than a repeatable starting point, we can still use them to our advantage without falling into being a prisoner of the obvious.
We can begin to differentiate ourselves by simply realizing that the most obvious signs that a team is in a good spot are going to be the most obvious signs to everybody. Vegas is a great tool, but everyone else is using that tool as well. We need to find ways to utilize it while making sure it isn’t the only tool we bring to the job.
I suggest that you still use Vegas, but now begin to use it last. Look into other statistics, break down the matchup, and rank the offenses on your own. Then, take a look at Vegas and see where your favorite offenses match up with what the lines might be telling you. See if there are any discrepancies that you could take advantage of. You may think that the “edge” in using Vegas is lost if everyone is using it, but that it only the case if everyone is using it in the same way, with the same level of comprehension as us.
In JayCabay’s Where is The Edge in MLB DFS you’ll learn:
• What is the “edge” in DFS and why it’s important to realize
• Whether stacking is optimal or even valuable anymore in MLB DFS
• How to use stacking differently to create unique lineups
• How to find your own personal edge
To read the rest of JayCabay’s ‘Where is The Edge in MLB DFS,’ you must purchase the course!
