DFS Stacking: The Strategy & Thinking Behind It (Fantasy Football)
The chances are you either googled for DFS stacking or you’re looking for a way to stack like a boss in 2021. After highlighting BestBall stacking practices in June, Betz and I discussed the DFS stacking on a recent Fantasy Footballers DFS episode.
The biggest difference between stacking in DFS versus Bestball is the realization this is a weekly game with salaries and roster percentages that move our selections. Stacking in DFS comes down to correlation and betting on a combined ceiling. If you aren’t thinking through your selections each week with correlation in mind, you’re already behind the eight-ball. I’ll discuss what that looks like and how stacking increases the range of your outcomes.
Stacking in Cash Lineups
The pursuit of the perfect lineup can be an illusion. Luckily, in cash lineups, you aren’t trying to hit perfection but give yourself the opportunity for some upside. Employing stacks in cash may feel like a risky move but instead, you are betting on some simple correlation between a QB and one of his pass-catchers. The obvious answer is to find a QB & WR that give you a high floor but remember how volatile WRs can be sometimes. Every WR in the NFL had at least one week outside the top-50 for fantasy.
If correlation feels like a bit of an ambiguous term, here is a visual for those who like colors. This is the correlation of fantasy points since 2014 for between every single position per FantasyLabs’ research.
Let’s take some of this data and parse out where the field is undervaluing stacking.
- QB + WR1 is clearly the most closely correlated (0.55) among stacking options but it might surprise you to see how close the WR2 and TE1 are to that number. The field likely undervalues these other positions when it comes to stacking.
- But it’s the other QB on the other side of the field that has the highest correlation (0.59) among any pairing. Obviously, we can’t play two QBs in the same lineup but it tells us if one QB is going off, the other side of the offense is set up for success as well. Building contrarian lineups with this in mind gives you access to more players and combinations for stacks.
- Just to take this a step further, I threw out this question recently on Twitter just to see what the perception was about QB stacking. Now the sample size isn’t huge, but it’s interesting a group of people that play fantasy football (I’m assuming that’s the main target group that follows me) mostly got it wrong.
Which of these positions do you think has a stronger correlation in terms of stacking with a QB for #DFS ? 🤔
— Kyle Borgognoni (@kyle_borg) August 12, 2021
- Per our correlation chart above, the opposing team’s WR2 carries as much direct correlation (0.41) as their RB1. This isn’t to say there’s causation but rather something to take advantage of in large-field tournaments. If you’re rolling with a Cowboys stack, pairing Curtis Samuel instead of Terry McLaurin also would likely give you leverage on the field.
Six Questions to Consider When Stacking
When considering stacking, you have to ask yourself about the combination:
What is the combined salary? Just add the salaries together and ask yourself what type of percentage of your overall salary you are giving to this stack. Anything over 30 percent for two players is a high-end stack regardless of what pass-catcher you are pairing. That might be alarming but your QB is likely going to be one of your top 2-or-3 highest salaried positions each week.
What is the combined roster percentage? Use our roster percentage projections in the DFS Pass and ask yourself what is the combined roster percentage of these two players. Remember for tournaments, you should aim to get in somewhere between 100-125 percent for your TOTAL roster. For a more in-depth read on this subject, check out the recently published Roster Construction & How to Use Roster Percentages article.
What is the combined ceiling? At a minimum, if you aren’t getting 25 points per slot with your stacks, you are not getting the return you want for GPPs. Weigh this out in your stacks… can you get 60 points from these guys? I know we all paint with rose-colored glasses in DFS hoping to win millions but median projections can’t help us in this exercise.
What is the right game environment? We discuss this all the time but Vegas knows what they are doing. Team implied totals need to be part of the weekly analysis but even more than that, what does the pace of play look like? Luckily, we cover those exact metrics each week on the podcast and in the DFS Pass each week.
I looked at every single game in the 2020 season per Vegas and how the closing line look from an over/under and whether the spread was covered.
The main takeaways, at least from the 2020 data, is that games hit the under far too often on totals above 50. And the over hit a solid 57.6 percent of the time on games just below that mark. I definitely want to target this next tier of games on each slate especially if roster percentages and stacks are likely less popular from these games.
What is the right combination of players? For GPPs, this is why we try to Max Enter. There are so many possible permutations that your eyes will go cross-eyed trying to cover all your bases and get every single possible combination each week. As mentioned in the Forecasting 101 episode, there is a range of outcomes for each player and on a game-by-game basis, the volatility is unbelievable in this sport for many positions but especially WRs. Be willing to combine the WR1 + WR3 or WR2 + TE sometimes to mix things up while stacking. You’re looking for the ceiling outcome not always the most likely outcome.
What is the right amount of leverage? You want to create distance between your opponents and the way to do that is by creating leverage. We can use heavy roster percentages (25+) as our ally and find plays that contradict those lineups in our stacks. If Christian McCaffrey is the chalk RB that week, perhaps look at a Sam Darnold stack with two Panthers WRs instead. We mentioned this recently on the How to Win at DraftKings podcast and will continue to give leverage plays in the DFS Pass all season long.
Let’s Put This to the Test
Here’s an example from the 2020 season I plucked randomly. I chose Week 3 in 2020 and looked at FantasyLabs data on roster percentages and where the field was attacking this slate from a stacking perspective.
What stands out about this group?
In other words, if you had this information in front of you before the slate per our Roster Percentage projections in the DFS Pass, what would you note? The Cardinals stack would be the most popular stack on the slate while also noticing that their combined salary is roughly 30 percent of your $50,000 DraftKings salary. Who would you be running it back with on the Detroit side this pairing to achieve a ceiling outcome? The field seems unsure which Seahawks stack to go with although Dallas’ defense. Miles Sanders seems to be the popular RB that week against the lowly Bengals likely due to his price.
What actionable information could you take?
- As we’ve noted, the RB + DST stack doesn’t correlate as much as the field thinks. Plus when you combine that with over 50 percent roster percentage, Sanders feels like a fade in tournaments.
- While the Cardinals stack’s roster percentage and combined salary are not unheard of, it does make you pause knowing the most popular stack is expensive. You would need to get creative on the Lions’ side or stack someone else from the Cardinals to create leverage.
- You either must pick a Seahawks WR or go with both in this instance and roll the dice.
- The Bears were easily the cheapest of these stacks and the environment against the Falcons seems like a solid GPP play but not nearly as contrarian as people should want.
After factoring in the right game environments (hello Seattle’s porous defense) and the right combination of players, there are some sharper plays that emerged. Running the Seattle stack back with a Dallas WR clearly was an easy move to make but Michael Gallup was actually the best play over more popular teammates Amari Cooper and Ceedee Lamb. If stacking the Seahawks seems expensive and even popular, then creating leverage at other positions is crucial.
Clearly, with Miles Sanders being the most popular play on the board, you had to get different. As you move down the list, some elite RBs in the same tier emerge as Austin Ekeler was just $400 more than Sanders. Chris Carson was a way to move off Seahawks stacks so he certainly would’ve made sense. The Big Dog Derrick Henry is also always enticing and being sub 12 percent, he wasn’t as popular as other weeks.
Finishing the Drill
Now that you’ve seen some of this played out and an example from last year, let’s finish by reviewing Game Stacks and Team Stacks from a correlation standpoint one more time.
It’s not just enough to stack your quarterback with a pass catcher, you’ll typically want more players from the same game environment than just those two. The QB-WR combination (due to its popularity) doesn’t provide a large edge in GPPs by itself. There is high risk and high reward we know but stacking raises the overall correlation and upside of a lineup. I mean a WR from an entirely different game isn’t correlating my friend.
Because the game itself is a correlating environment, we can open ourselves up to more and more ceiling outcomes. Per a 4for4 article, a QB having a big game has a positive effect on nearly every other player in the game, both on the QB’s own team and on the opposing team. When a QB posts a 25-point game, there’s a 61-percent chance the opposing QB will too.
A QB-RB1 stack is also a reflection of the success of the entire offense. Think in 2020 when the Titans offense rolled, it wasn’t just Derrick Henry leading the charge. Ryan Tannehill also showed a positive correlation with his backfield mate. Although the correlations are higher for the WR2/WR3/TE1, the RB1 is another solid choice because it gives you access to the entire offensive arsenal.
Using FantasyLabs data from 2014 to 2020, we can see how fantasy points correlate with each position in stacking and where the field differs in their approach.
- Triple Stacks work- Using a QB-WR-WR + Opposing Player was underutilized and the correlation was even stronger when the total was UNDER 50. Don’t be afraid to grab four total players in a game but the field certainly isn’t. Top-10 lineups in Milly Maker over the last two years used a QB and two of his teammates 41.2% of the time while the field did it just 28.9% of the time
- We underestimate TEs. Their correlation is almost as strong as WR2 & they are usually much cheaper. If you are avoiding the elite tier of TEs, ensuring that your TE is stacked with your QB is a solid bet and one to take advantage of versus the field. QB-WR2 and QB-TE1 are tied for the second-strongest positively correlated stack. We’re not asking our TEs to go bananas but a stack such as Brady + Bucs WR + Bucs Gronk // Russell Gage is setting your apart.
- We underestimate game stacking opposing RB1. This creates leverage because normally a simple stack of QB + WR // WR is going to make up a ton of lineups. Using Matt Ryan + Calvin Ridley // Ronald Jones II is just different.
- We overestimate “pass-catching RB2s” by a large margin. We think we know game scripts and how things will go but your Nyheim Hines types mostly are wildcards.
Stacking is like getting a Verified Check on Twitter. Stacking is cool. All the cool kids do it. Which now makes it kinda normal… not as cool. You have gotta put a personal flair and mix it up.
Stacking is like a greenhouse for fantasy points. We need more than just an app to select a stack. We need the right game environments, the right combination of players, and the right amount of leverage.
Stacking is like a party to invite new friends. Game stacking unlocks correlation no one else is using. Get in on that party.