Hi, I’m relatively new to playing DFS and using the DG Optimizer. I was wondering if anybody had suggestions relating to the settings (specifically the amount of diversity to use and “Declump Lineups”). Not sure if it is more optimal to have a lot of exposure to one player and hope they do well or play it safer by having more variety across lineups.
So, this is a thought experiment that probably needs more than a quick explanation, but I’ll try my best 
Ideally, you think about it in 2 different terms - Whether you’re playing cash games or GPPs. (Although, cash games are much tougher than they used to be, so I barely play them anymore.)
For cash games, you just want to use the optimizer to create the best lineup possible, with no thought given to ownership %. For GPP, you want to start thinking about how to get leverage on the field.
So, if you do a 20 entry max GPP, if it’s a no-cut event like the Sentry this week, maybe set max ownership to like 50-60%. Bump up the guys you’ve identified as your main targets, and then choose Declump lineups. Usually for large field tournaments with a cut, I’ll set max exposure to like 25-30%, so one guy missing the cut doesn’t kill half my entries. Also, definitely use Some or Pushing diversity for GPPs.
After running the opto, I like to look at the output player exposure and identify any guys that I like that the opto left out, then manually bump them up and run it again.
Once you get into large field tournaments, you’ll want to start thinking about setting the max total ownership projection to something like 80 or 90%. You def want a sub 8% guy if you want any chance of taking down one of those large field tournaments. (Often even sub 5%.)
Start looking at the GPP winners and deconstruct how their lineup was set up. How many sub 10% guys did they have, what was their total ownership %, etc. You’ll start to see how you should set up the opto to attack fields and give yourself a chance.
I agree with @wallmr comment.
In general, you want unique lineups that have a lot of upside, especially if the contest is big. Whether you have a lot of exposure to a player in a given week or not is purely a matter of risk and reward. If you take a big shot on one player or a few players, you have higher risk (if they miss the cut, you’re screwed for the week) and higher reward (if they win the tournament, and they’re on all of your teams, you have a ton of live shots to win a DFS contest).
Rule of thumb: teams that can score a lot of points, that are hopefully unique. That is the recipe for winning DFS (also, get very lucky!).
1 Like
I’ve been using the optimizer for years for PGA DFS. Also built some custom models to help identify other things that will help DFS scoring… the biggest thing that most sites struggle with is identifying guys who won’t just finish on the leaderboard but “score” well.
You NEED the guys who are making birdies and eagles WHILE finishing high up on the leaderboard. Cam Young never wins real tournaments but he will win you DFS tournaments because he tends to score well (more birdies, eagles, birdie streaks, etc.).
In the recent events, DataGolf has metrics around “Great” and “Poor” shots. Cam Young is commonly very high in both compared to golfers of similar skill - always go for the upside.
2 Likes