The Greatest “Cheat Sheet” In DFS
Each week in The Oracle, OWS team members will take on the key strategy questions from that week’s slate :: sharing their thoughts on how they plan to approach these critical elements from a roster-construction, game theory, and leverage perspective.
We often start out The Oracle with the question, “What makes this slate particularly unique?” Obviously, in Week 1 there are a lot of unique aspects ::
With all of those things as common and clear factors about the “uniqueness” of Week 1 NFL DFS, we will change course from the “standard” tone of this question we have on a weekly basis and instead ask a more philosophical question: Are there any ways in which you approach Week 1 differently or have different expectations than that of a “typical” DFS week? Bankroll management, contest selection, player selection, etc.
So…a programming note.
We ask this same question at the top of the Oracle in Week 1 each year. This year, Mike (who comes up with the Oracle questions) left our 2024 answers in place, in case we wanted to reference those, or just use them — given that they are fairly evergreen answers, with sharp thinking/takes behind them.
Last year in Week 1, I included my 2023 answer, as I felt it had value/relevance heading into 2024, and then also added some new thoughts for 2024.
This year, I read my 2024 writeup and thought, “Funny, I’ve been saying the same thing on podcasts today.” So as has apparently become tradition for me, you can find part of my answer from last year first, and then you can find my “2025 answer” below that ::
MY 2024 ANSWER ::
I’m answering this question on Friday morning, fresh off last night’s Ravens/Chiefs game. In that game, both offenses had to open things up, we had lots of back-and-forth momentum, and we were a shoe tip away from a game total of 53 to 55 (the Ravens were set to go for two, rather than sending the game to overtime, so the game total would have landed in one of those two buckets). And yet, Derrick Henry had only 43 yards, Travis Kelce had only 34 yards, Mark Andrews was invisible all game, Zay Flowers finished under 40 receiving yards on 10 targets, Lamar Jackson ran for 122 yards(!) and still failed to top 30 DraftKings points, and Patrick Mahomes threw only one touchdown pass. Rashee Rice went for 100 yards, but failed to score a touchdown (a solid but unspectacular 20.3 DraftKings points). Which means that in a game that featured both offenses opening things up, and some back-and-forth-momentum, and “a shoe tip away from a game total of 53 or 55,” Lamar’s strong price-considered effort and Isaiah Likely’s 29.1 DraftKings points were the only scores that would have really mattered if this game had been on the Main Slate. (And if this game had been on the Main Slate, Likely’s ownership surely would have been under 3%.)
NFL games are very complex. The way scoring and fantasy production develop and pile up are complex. And yet, we come out of a game like that, and on Friday can still hear people talking about the upcoming games this weekend as if they are all being played on a piece of paper.
Most of the expectations people have for players heading into Week 1 will be wrong!!!
MY 2025 ANSWER ::
There was more to my answer last year than that, but that first part — coming off the Thursday night game, and being reminded of just how much people tend to get wrong in one isolated game — was the same place I found myself this morning.
A.J. Brown had one catch on Thursday Night Football.
Saquon Barkley had 60 rushing yards.
George Pickens had three catches.
Javonte Williams scored two touchdowns.
And the crazy thing is: people can watch that, be surprised by the box scores, and then still think that their edge on Sunday will be “properly predicting the box scores of every game on the Main Slate.”
This is not our edge!!!!
We can absolutely have a strong understanding of who the best on-paper plays are, and of where we could expect to make the most money if we played out this slate a hundred times. And we can gain a strong edge by understanding how to properly build rosters to account for the math that we’re turning in our favor. But we should also expect to be surprised in some places…
…and in fact, we should position ourselves to benefit from places where we might be surprised.
In my 2023 answer, I mentioned that I never personally feel Week 1 is my greatest edge. I consider myself to be an upper-echelon DFS player when it comes to 1) understanding the strategy/math of roster building, and 2) knowing players, teams, and coaches well enough to find a few places each week where my calculation of the math on certain spots is better than the calculation of the math spit out by projection systems.
In Week 1, projection systems will be at their weakest — but if I put too much confidence into my own ability to be better than projection systems in certain spots, I might actually be hurting myself instead of helping.
By the time we get to Week 4/5, I’m usually at my absolute best, as I have enough information to feel confident that I’ll find a few spots each week where projections are off, and that I’ll know what these spots are and be able to add that extra edge to my strategy/roster-construction edge and position myself for high-end results over time.
In Week 1, however, I have to really challenge myself to let go of what I think I know, and more fully embrace the idea that I might have just as many blind spots and suppositional misses as the next person.
Because of all this, I often go slightly underweight my normal bankroll in Week 1 (whereas, for others — depending on their own strengths and weaknesses — it might make sense to go a bit heavier than normal in Week 1 bankroll).
I think my own answer to this question (in terms of my own personal approach) should not be read as a recipe for you to follow yourself, but should instead be read as “a look into how JM sees his own strengths and weaknesses stacking up on this week, and how I can leverage that thinking to identify my own strengths and weaknesses here.” What is your edge in DFS? And is that edge increased or decreased given the unknowns of Week 1?
We should be comfortable embracing a little more uncertainty this week.
And we should acknowledge to ourselves that we might know a little bit less than we think we do.
While my 2023 answer is still valid, I want to expand on the variance/chalk piece. I say this every year in some format, generally on a show somewhere, but I want to write it down as well: we know very, very little in Week 1. We THINK we know things, but as JM noted, there are players whose roles may be very different than we think they’re going to be. Offenses may feature players differently, they may be more pass or run heavy, and defense consistency from year to year is very low (i.e. defenses we think are attackable may be good, defenses we think are good may be attackable). We know very little!
When the field is extremely confident but we don’t actually know things, that’s the time to embrace more variance. Recognize how little we actually know and play into spots where if the variance swings our way, it benefits us in big ways. Always remember “what do you win when you win?”
I tend to go heavier in Week 1 because it’s a week where the edge is larger than normal. We have the most casual players (the players who will drain their bankrolls early in the year and then stop), we have a smattering of new DFS players in the ecosystem (even though DFS isn’t growing like it used to, there are still new players every year), and we have the most uncertainty around how teams will use their players and how good they will be (but the field sees perceived certainty in, say, games like MIA/LAC). All of that adds up to edge, and so I tend to go heavy in Week 1 to try and capture it.
I’m also more willing to embrace variance in Week 1. As the season goes on, chalk tends to become “better,” because we gain more knowledge about how teams operate. In Week 1, chalk tends to be more fragile because we’re operating off of a lot of assumptions, but assumptions are not the same as knowledge.
Oh hey answer from back in 2023/2024. Still good! Like I say every year: Week 1 is a good week to embrace variance. If we briefly take a look at a few chalk plays…we have two rookie WRs whose roles could potentially be smaller than their ownership is predicting (can also toss Ricky Pearsall into that bucket as a 2nd year guy) while we have low ownership on some capable-but-not-as-exciting veterans. At running back we have opportunities to take advantage of unknown workload splits (what if the Pats give Henderson…or even Stevenson…a full role in Week 1 instead of a split? What if the Commanders have one main back instead of a split backfield?). And that’s not even including outcomes such as a team being significantly better or worse than projected, or just a team having an unexpectedly good game (remember the Saints putting up 40+ points in back to back games to start last season before falling back to earth?).
We know a lot less than we think and that makes Week 1 a great spot to lean into the variance and play a more contrarian style.
First of all, the basics of DFS play are the basics of DFS play. How we approach this game (read: puzzle) remains the same. Too often players get tripped up early in the year because we don’t have established trends, pricing is loose, rookies enter the fold, and the dynamics of teams are all wonky due to changing personnel and coaching. Simplify things this week – find the game environments, teams, and players in the best position to return usable GPP scores and go from there.
The biggest aspect of roster building that we absolutely must keep in mind in Week 1 is that we can’t account for every outcome, something that gains increased importance when we’re overloaded with potential plays due to pricing structure and uncertainty. We don’t build rosters looking to account for every potential outcome of the weekend, we build rosters as if that roster, with those players, in those game environments, are what you need to be successful.
The final aspect of Week 1 play that deserves mention here is the historical performance early in the season. Per Fantasy Points’ Graham Barfield (@GrahamBarfield on the X machine), 60.9% of Week 1 games over the previous four seasons have hit the under. Vegas historically wants exactly 50% of the action on the under and the over – they scoop the vig on move on. What that means is that Vegas is getting more action on the over in Week 1, amounting to a psychological play on human instincts (we think teams are going to perform better than they actually do early in the season). In other words, fewer teams are going to score more points than we think in Week 1, meaning fewer players are going to see outlier performances. Capturing the teams and players that manage to outperform expectations is paramount to early-season DFS success.
The rest of the puzzle is still there – variance, DFS theory, game theory, etc. Stick to the basics and focus your efforts on specific teams, or players, that have elite ceilings within their respective ranges of outcomes.
Bankroll – I do generally go a little higher in the amount I play Week 1 for a couple of reasons. First, I play more entries of the super large field stuff (Milly Makers on both sites and the large contest on Yahoo) this week than a normal week because overlay is likely, price points of those contests are lower, and such a large chunk of rosters are basically dead from being very inexperienced players and/or rosters made days/weeks in advance. Second, there is a ton of value in being early on players/teams and Week 1 provides the best chance to get paid off for having unique thoughts and seeing things for yourself – which I believe is a strength of mine.
Contest Selection – As noted above, I adjust slightly to leverage opportunities in the larger field contests.
Player Selection – I want to get in early on guys I think are going to pop and leverage teams who have different outlooks than the last time we saw them – as most people are in a “see it to believe it” mindsets even when they know situations are different.
“Vegas lines” are the basis for so many projections, predictions, and rankings in DFS (and really in all forms of sports betting, etc.), but Week 1 of the NFL season is the time when those lines are likely to be the least efficient and have things that are way off base when we look back at them in a couple of months or the end of the season. Every team has personnel and scheme changes from one season to the next, some bigger than others, and players are changing as well – getting older, recovering from injuries, sustaining injuries, training/practicing to improve. There are just so many variables that are extremely hard to account for heading into the season, which is part of what makes it unpredictable and fun.
Last year, for example, we saw a Panthers // Saints game with a 41.5 game total where the Saints alone scored 47 points. In the next question, we will talk about being “ahead of the curve” with players….but in this question I want to talk about teams. Last year’s example ended up being an indicator for what would be one of the league’s worst defenses in Carolina. With that in mind, we have six games on the main slate this week with totals under 44 points::
Are there any teams or units (i.e. – if you think a particular offense will be very explosive or one of these defenses is one we will be targeting regularly as the season goes on) from that list that you will be targeting this week as you think the market may be inefficiently pricing what they will look like in 2025?
