Just as our Adam Levitan does each week during the NFL season, every Monday we’ll be using the week-long nature of DFS golf to take advantage of the extended feedback loop by analyzing what went right and what went wrong in the previous week’s slate, regardless of results.
Each week, we’ll review a variety of cash-game & GPP lineups in an effort to take you inside our thought process as we approached Thursday’s lock and how our projection- and ownership-based expectations stacked up against the field.
I am back again for another edition of the GPP Lineup Review! As Discord senior lsahr noted, anyone that has a big score over the weekend is apparently signed up to do EXTRA work around here at ETR, so here I am writing another article. I’ll be reviewing my $200 Driver single-entry lineup here, but I’ll also be reviewing my overall MME approach on the slate, as it really defined the way I build both optimized and hand-built lineups. I will not be reviewing my R4 lineup approach and will cover that in my Large-Field GPP Breakdown later this week. Here is a screenshot of my winning lineup:
I’ll of course start with the winning lineup and talk through my process here a little bit. If you read my article last week, you’ll know that I was taking a contrarian stance on the week by being way underweight Rory McIlroy and fully fading Keith Mitchell. Generally, I wanted to attack my contests with unique starts and overall builds, which was exactly how I approached this lineup. In smaller-field contests (The Driver this week was 2,200 entries), I fully expected ownership to condense on Rory, Bryson, Mitchell, and Mackenzie Hughes, which Sam noted in his small-field ownership notes. For those of you playing in these contests, this is a very key aspect of building for these contests — there are instances where I will deem a player unplayable in small-field contests but more of an underweight position in large-field tournaments.
The good news for me was that Sam was correct on the small-field ownership steam. Rory (39%), Bryson (34%), Mitchell (31%), and Hughes (15%) all came in above where they landed in large-field contests, especially Mitchell, who was shockingly in the 30s. Knowing this likely would happen, I wanted to leverage the field by playing two players that would likely squeeze on the opposite end of steam. Those two players were Scottie Scheffler and Jon Rahm, who had strong projections from the ETR team and were coming in at lower ownership than in large field. These are some of the best ways to attack both types of GPPs. Scottie and Rahm at a combined 30% ownership was an absolute gift, and I felt great about it when the cards flipped Thursday morning.
One of the other key points in my article on Thursday was how the field seemed to be overconfident in the $5K and $6K ranges in an effort to justify Rory-Bryson lineups. I thought that if these player ranges were so strong, why were only two players taking on tremendous amounts of ownership? Well, that’s probably because the perceived strength of the ranges wasn’t actually true. I wanted to take advantage of this by trying to limit the amount of low $6K and high $5K players I had, leading me to this build. I ended up on the above, which was a 12/10/7/7/6/5 that came in at 50% cumulative ownership. I typically end up with very low-owned lineups in single-entry contests, as I frequently take aggressive positions against chalk even if the player projects well. As the ownership gap widens between chalk and contrarian, the contrarian plays often stand out as better options, and I am not shy about playing them.
Talking Through MME Sets
While I simply broke even on the week in MME classic, I still think it’s important to talk through how I built my overall set. I’m not going to post my full exposures as I’ve done in the past, as the focus for this article was on my single-entry build. But, everything that I laid out in my GPP Breakdown on Thursday was implemented in my overall MME set. If you go on Labs and take a look at my exposures, you’ll find the following:
- Extreme underweight positions on Rory and Bryson, shifting ownership to take overweight positions on many of the other $10K+ players. I was also overweight the combinations of these players (Scottie + Rahm, Xander + Morikawa, etc.). I achieved this through the use of boosts in The SOLVER, where I use the basic IF statements to boost salary ranges with a player.
- Overweight positions on many of the $8K and $9K players, as I believed these ranges were way underplayed given the skill level. The field was overconfident in Bryson and Rory lineups; I wanted to attack them by getting more balanced or normal types of builds. This meant I had to be overweight many of these players given that I was going to have two or even three in these ranges in my lineups.
- Very few players taking chalk in the low prices, as these are typically the most volatile players and the chalk that I am most comfortable fading. As you know, each reaction has an opposite reaction, thus all my Mitchell and Hughes ownership was dispersed to others in the range or lost to different build structures.
With this information, you can start to understand how I built my full MME set. There were a portion of lineups that started out with contrarian stars builds that utilized low-owned cheap plays (Taylor Moore being one of them). Another portion would feature a high-priced player and two or three $8K players with him. Depending on the price of that stud, maybe it was a 10/9/8/8 start, for example. With the large player pool that we had, I was able to create 150 lineups in this manner, all within or near my cumulative ownership targets, that I felt were going to be unique from the field. It’s important to remember that while 150 lineups may seem like a lot, it really isn’t when you’re playing large-field GPPs. You have to make the most out of these lineups, and I feel like the best way to do this is to simply build differently when I know the field is not doing so.
Using Optimizers to Your Advantage
I talk constantly about how DFS players simply do not know how to use optimizers to their advantage. The field views optimizers as an easy way to reach 150 lineups that you can submit and move on. What an optimizer can be, however, is a way for you to manipulate a 150 set to your advantage by adding some rules to your sets. Too often, I hear people say things like “The SOLVER gave me 40% of this guy” as a reason for why they played someone heavily. There is a lack of understanding of what the optimizer is actually doing here, and that leads to a lot of losses for the field. Below, I’ve screenshotted an example of one of my rules that I used for the PGA Championship:
This rule is specifically set for Scottie Scheffler, but I have multiple rules with other players and also combinations with Scottie as well. What I’m doing here is decreasing my frequency of Scottie + Rory-Bryson builds and directly funneling that ownership to Scottie + another stud. I also wanted to boost $8K and $9K players with Scottie as a way to avoid a typical 13/7/7/6/6/5 type of build. While I have full decimal places in this screenshot, some of my other rules feature more exact numbers that I landed on by constantly re-running the optimizer to get my desired combination ownership for my set.
I give this as an example because, at the end of the day, DFS is a peer vs. peer game. You have to think about what everyone else is doing and how you can beat it. I know for a fact that the PGA DFS streets are riddled with poor players takes, lazy optimizer builds, and also just bad projections. My thinking when playing PGA DFS is this: If I can effectively give myself a way to be unique while also using the best projections out there as a base (ETR), I am at an advantage for every single slate. My performance in PGA DFS has proven this notion.
Finalizing an MME Set
This is a copy + paste from my last GPP Lineup Review, but I think it’s a good refresher on how I build my MME sets. I never just take what The SOLVER spits out and immediately upload it. I am probably pressing that optimize button over 300 times throughout my process, constantly massaging player ownerships and combination ownerships to my liking. I will also go through and sort and X lineups based on a variety of different roster-construction things that I am liking on the week. I typically like to attack my MME set with these steps:
- Build my player pool and get a good idea of the player ownership ranges I am looking for. For example, I mentioned in my Large-Field GPP Breakdown that I’d likely come in around 25-30% Corey Conners ownership. I am constantly editing projections, rules, etc., and optimizing to get every single player to the range I want them to be at.
- Set up my rules, complete with min/max, docks and boosts, and projection edits. Continuously optimize and tweak until I get to a point where my player ownership looks appropriate.
- Once I have a set that I like, I start to filter through my lineups. How many Scheffler-Rahm combinations do I have? Who are my most-played players in those types of lineups? Is that comfortable for me? I am cycling through a variety of these scenarios to ensure that I am confident that all 150 of my lineups are representing a lineup that I would play in a hand build if I did that.
Now, everyone is going to have a different process in building their MME set, different rules they like to lean into, etc. I am simply trying to outline my process to help you become a better DFS player. You’ll notice that in my GPP Lineup Reviews, I almost never mention anything about the players themselves — I simply do not care all that much whether or not Scheffler is a better course fit than Rory (or whatever). Here is where I trust that the projections will generally lead me in the right direction, and I can use that as a base to what I’m doing on the week.
There you have it. Another GPP Lineup Review from me. Hopefully, I’ll be writing another one in the near future! But if not, just know I will still be continuing to find out edges against the field in the meantime. If you have questions for me, please feel free to ask me in Discord. I hope you found this article helpful in your DFS journey. If you didn’t, well, kudos to you for getting to the end. I’ll see you all on Wednesday for my weekly Large-Field GPP Breakdown article.
