DFS Strategy through Backwards Design (Fantasy Football)
Recently, Matthew Betz and I discussed some DFS 101 Strategy on the Fantasy Footballers’ DFS podcast including a term I’ve been toying with for a while. “Backwards design” is an educational theory I adopted from my years as a high school teacher. In grad school, our professors taught us how to make our lesson plans with this process in mind simply beginning with the desired results, moving towards evidence that would reveal that result, and finally planning the actual experiences and instruction. In other words, it’s planning with the “end in mind” rather than just starting with the activities. Sounds elementary enough, right?
Perhaps you’ve heard of the term “reverse engineering”. It’s the process of deconstructing a product to see how the designs were put together. In fantasy football, we live often in the world of hindsight analysis. The entire offseason is spent analyzing the statistics and film of what happened the previous year and learning how to interpret that information. The problem with hindsight analysis is it creeps into the season and actually slows down your reaction time. Now that we’ve had our pedagogical and psychological refreshers, let’s move on to how this could actually affect our process for building DFS lineups.
With DFS, the turn around for making shifts is much shorter. In fact, it’s week-to-week. The room for adjusting and engineering a new strategy each week can lead some to light their entry fees on fire. The simple backwards design process for DFS is a three-fold stepping stone to gaining confidence:
- Calculate a Narrative
- Identify the Inefficiencies
- Stack & Mix
To get the full picture of our weekly DFS articles, tools, and research, get on-board with the Fantasy Footballers’ DFS Pass right now for the lowest price of the year.
Calculate a Narrative
As a writer, it’s the narrative that compels me to cover a story or project what I think might happen with a player or a team for the upcoming season. For DFS, the task is to “storyboard” what you think might happen in a specific game. The Vegas lines are the starting point for analyzing the slate and beginning to decipher game flow, the pace of play, and most importantly, how the scoring will unfold.
The word calculate is the keyword here as anyone can spout off reasons why they might like certain players that week. The process of working backward is about seeing how the numbers tell a story. Median projections, which much of the fantasy world operates in, give us the blandest version of a narrative. For instance, Stefon Diggs‘ floor is easily the highest among WRs leading the league in targets and receptions last year and having a low bust rate according to our consistency metrics found in the Ultimate Draft Kit. Let’s say his median projection is 16.5 DK points in Week 1 versus Pittsburgh. That is impressive and makes him a locked-in, viable cash gameplay each week. He’s Stefon Diggs. But for tournament-winning purposes, the median projection does not give us the narrative we need because DFS mostly operates at the extremes.
One of the key ingredients for a week-winning narrative is finding the players that double, triple, and even quadruple their value. You can calculate this by taking their salary and finding out how many fantasy points they need to return value. Let’s keep the Bills WR at the forefront of our discussion as his Week 1 salary is $7,600 on DraftKings. In order for Diggs to 2x in value, he needs 15.2 DK points. In other words, it’s not hard envisioning him hit that mark given his high floor. Nevertheless, you must weigh the opportunity cost of Diggs’ WR1 price tag versus the return in value. Can he go 4 or 5x this week and hit 35+ points? Or is it worth spending on players at a lower price point (say $6,500) who can hit 2 or 3x so you use that salary elsewhere like at RB? It’s simple math but calculating the narrative takes our assumptions and hopes for the week from the ethereal to a concrete storyline.
But beyond individual player projections, the most important part of ‘calculating a narrative’ is how your entire roster flows together. For example, let’s take the LV-BAL game on Monday night Week 1. The Raiders games hit the over in 13 of the 16 games hit the “over” on Vegas lines. That was by far the highest total in the league. If the Raiders and Ravens have one of their classic shootouts in Week 1, what is the best way to attack that game? Do you automatically stack Derek Carr with Darren Waller or do you go the non-conventional route by plugging in newly acquired Bryan Edwards or Henry Ruggs III?
It’s easy to go down the rabbit hole of scenarios but, remember, this is the starting place. Check the lines, get out a sheet of paper or spreadsheet, and identify early in the week the games you feel like you have a beat on. Each week on the DFS podcast we cover the four or five biggest matchups we think will win people tournaments and help cash-in for head-to-heads.
Identify the Inefficiencies
As J.J. Zachariason points out often on his LateRound Podcast, fantasy football is another example of how a market functions. There are ebbs and flows and in the DFS world, we want to find out where the market is either slow to react or downright wrong in terms of pricing for players.
Every single week I publish our First Look Pricing article which functions as a “sneak peek” for the upcoming week. We give you the low-down on DraftKings and FanDuel pricing first thing Tuesday morning. Consider it your news briefing or your morning paper for DFS every Tuesday as a chance to get the lay of the land for the main slate. Tuesday signifies the reset button for fantasy and an opportunity to turn the page and look forward to new matchups and pricing inefficiencies. For example, perhaps you’ll see the Denver backfield shift after Week 1 where Melvin Gordon may have less of a role and rookie Javonte Williams skyrockets in price.
Matthew Stockman/Getty Images
Because the pricing for players comes out on Monday evenings, we don’t have a perfect picture of the effects of injuries, backups slated to receive a bigger workload, or the aftermath of the Monday night football game. Those are the three areas to take advantage of inefficiencies for the following reasons:
Injury Discount Double-Check– The injury carnage and the news blurbs we receive on a Monday is really just an initial briefing. Our injury expert Matthew Betz puts out his Injury Recap article every Tuesday and then updates during the week as more information comes out about players and their situations. The pricing reflects this ambiguousness as Monday night decisions have to be made for DraftKings and FanDuel on pricing. Be ready to notice the inefficiencies when a player who is presumed to be in concussion protocol or left a game early is available for the next week.
The Backup Supreme Special– But the most obvious way to take advantage of the timing of pricing is finding the cheap backup. Opportunity is king especially in DFS where players are rostered as a one-week rental. Any RB in the NFL is viable if they are given the workload (15+ touches) and able to fall into the end-zone. If that backup RB has a pass-catching pedigree, they become one of the true “locks” in DFS and a chance to save you a ton of salary. Tony Pollard, Latavius Murray, and Alexander Mattison are just a couple of examples of recent backup RBs who stepped in for their injured teammates at a cheaper price and came through in a big way for DFS. These RBs become the “chalk talk” of DFS for the week.
The MNF Glitch– The pricing for both DK & FD is released Monday evening while the Monday night football game(s) are going on. This creates a weird scenario where the pricing algorithm for players in that game only can generate pricing based on the previous week’s data. In other words, if an RB were to get injured or a WR goes off during the MNF game, the prices for the following week will not reflect those performances. Use this to your advantage not only based on the prices but the emotional swing a Monday night “island game” can impart. If a player has an off game, be willing to go right back to them the next week in DFS while others might fade them due to recency bias.
Our DFS Pass also comes loaded with the Buzz Report, the roster percentage projections based on FanShare’s proprietary tool. It takes into account the number of tags online and combines them to give you a list of who the most talked-about players are at each position for that week. In other words, instead of going in completely blind, peruse the roster projector as an aid to see which players will be chalk and who is flying under the radar. Maybe you think Antonio Gibson is a great play this week no-one else is on but quickly you find out he’s the 4th highest tagged RB this week because of how cheap his salary is on FanDuel. The give-and-take is developing your process independently without the expense of being myopic and observing which players are the most talked about that week in terms of Twitter, podcasts, and DFS articles.
Stack & Mix
After putting in the time, researching the Vegas lines, identifying the price discrepancies, and reading everything we have available on the DFS Pass each week, here is when we actually start plugging in lineups. This is the good stuff. But before you start firing away like a kid with a shotgun, the starting place for making lineups is with high-floor RBs and stacking with your QBs. RBs are the cornerstone of building a successful DFS lineup because of their workloads and propensity to fall into the endzone. I won’t spend much time on that as Ben Cummins has you covered in How to Attack the RB Position in DFS. The more pressing matter is stacking.
I tracked every single winning lineup from 2020 for DraftKings’ flagship tournament, the Milly Maker. It’s plain and simple: every single one of them stacked their QB with at least one option from their same team except for Taysom Hill’s fluky Week 11.
Why stacking? The correlation is obvious as a QB who scores big will likely have a pass-catcher on their team who shows out well. If Dak Prescott hits 300 yards and three TDs, if you paired him correctly you are outpacing your opponent. Now whether that is Amari Cooper or Michael Gallup or… Ceedee Lamb or Blake Jarwin or Ezekiel Elliott…. you get the point that mixing in the right duo or trio of choice is paramount. For more information on the practice of stacking, check out Mitch Carl’s DFS 101: Stacking article.
“Mixing” refers to who occupies the player pool that you select from. In large-field tournaments, there’s no way you can cover all your bases even by max entering 150 lineups. You need exposure to some players who have a low hit rate and low projected roster percentage, especially at the WR position. Not only do you need to fill three of these positions, but the performances week-to-week are so volatile that you need to build that into your process. In Week 1, it was Sammy Watkins going off for three TDs against the Jaguars. The very next week as everyone was stacking Patrick Mahomes with the Lizard King, it was his teammate Demarcus Robinson rostered at 0.6% going off for 38.2 DK points and almost 10x his price tag of $3,500.
Let’s say you built 25 lineups with Mahomes as your quarterback. Tyreek Hill, Travis Kelce, and Clyde Edwards-Helaire would be the obvious choices to stack him with for good reason. Mixing in a Watkins in a couple of those doesn’t hurt considering his monster performances as well as the general disdain DFS players felt after he never came through again after that Week 1 explosion. Mixing gives room for error acknowledging the fact outlier performances in GPPs ultimately are what win out. For cash, keep your core group of players much tighter knowing you aren’t trying to shoot for Mars… the moon will do.
For the Milly Maker data, the stacking didn’t just end there. “Running it back” or “bringing it back” refers to pairing your stack with a player or two from the opposing team in hopes of a shootout. While it is not imperative to stack in this manner, it is a way to out-pace your opponents and differentiate yourself from the field.
Conclusion
The process of winning consistently in DFS is for the long-haul. Constructing your process through backwards design is not about hindsight analysis but rather having the end in mind before you begin plugging in names. This is more of a mindset to gain for beginning players but it is one I’ve found to routinely refocus my attention in the midst of so much noise, content, and choices in DFS. The scenarios are endless but the process can be simplified.
Let me know on Twitter (@kyle_borg) if this explanation was helpful and I’d love to continue the conversation into Week 1!