Transcript


If you toggle manual mode on in the build settings, you'll see SaberSim sliders. And these sliders will ultimately control the different factors that lead to upside when you're building your lineups.

Let's talk about what each of these do. And I'm actually going to start with the sim precision slider at the bottom, because this slider is the most important slider to the SaberSim process. This slider ultimately controls the diversity, upside, and volatility of your lineups. And it does this by controlling the amount of simulations that are taken into account when building each lineup. The higher this slider goes, the fewer simulations we're using to determine the projections we're using to build each individual lineup. At sim variance 10, we're using a single game simulation of our thousands of sims of each game on the slate to determine the projections used to build each individual lineup. And at sim variant zero, on the other hand, we're using every simulation we have for that game. In other words, the average projection to build each individual lineup.

Another way of thinking about this is as the sim precision slider goes up, we are effectively widening the range of outcomes for each player on the slate. The sim precision slider is a much better approach to finding diversity and upside in a lineup portfolio than doing something like optimizing just for ceiling projections or using traditional randomness on another optimizer, because it's using the real distributions and simulation outcomes for each player when building every lineup.

Now, let's jump back up and talk a little bit about correlation. Correlation in DFS describes two different players who have a connection or a relationship in their point scoring. A very common example would be a quarterback and wide receiver in the NFL who are positively correlated, because when the quarterback throws a touchdown to the wide receiver, they both score points. Simulating games allows us to precisely quantify how correlated two players are rather than having to use historical averages or intuition. And when you build lineups with SaberSim, the correlation slider factors in how much to pull in the correlations from our sim data into your lineups. In a nutshell, the correlation slider, the higher this goes, the more we'll factor in those average correlations, building better, more correlated, more stacked lineups.

Ownership in DFS refers to what percentage of lineups in a contest roster that player. We project ownership for all slates and sports we have simulations for, and the ownership fade slider controls how much to factor in those ownership projections or your own custom ownership projections when building your lineups. This slider's pretty simple. The higher the ownership fade slider goes, the more likely we are to fade highly owned players or the chalk. The one important thing to note here is we also take into account the ranges of outcomes of players here. The more variant a player is, or the wider the distribution of his outcomes are, the bigger the adjustment will make when determining whether or not to put a player into the lineup. In other words, a safer, highly owned player is less likely to be faded than a very high variance, highly owned player.

Now, SaberSim has back-tested and optimized these sliders for different types of contests by building thousands of lineups at a bunch of different build settings and calculating which sliders perform the best for different sports, slate sizes, and contest types. So we believe that the best sliders for a given contest are the default sliders that match that particular contest. We've done a lot of the heavy lifting for you already in terms of optimizing these settings. So stick with the defaults unless you have some other reason to change them.

If you're interested in learning more about the rest of the lineup building process on SaberSim, check out the other videos in our support documentation. Thanks and good luck.

Did this answer your question?