Using Our New NFL Salary Comparison Tool
You’ve probably seen the salary comparison tool normally up on our site. It looks at the differences in salaries for various players on FanDuel compared to DraftKings.
Comparing salaries on multiple sites can help you identify values. If one player is much less on DK than FD, he may become play worthy because his price relative to his value is good.
Comparing salaries can also help you figure out how to efficiently expose yourself to different players. You may really like a play this week like Darren McFadden (I do at least), but given a much lower price on DK, you should be focusing on playing a ton of DMC on DK while limiting your exposure on FD where he is still good but less optimal. Being able to accurately compare salaries is extremely important for playing on multiple sites, which is why we’ve made our salary comparison tool more advanced.
The problem with just looking at salary differences is they don’t take into consideration the scoring differences and differences in salary structure for each site. DK and FD have different salary algorithms and therefore to accurately compare salaries at each site you have to be conscious of how the algorithms differ overall and how they differ for each position.
There are some obvious differences between the salary structures at FD and DK. First off, the minimum salary is different for each position aside from QB, which is $5000 on both sites. For RB and WR, FD has a min salary of $4500, while DK goes as low as $3000. DK goes even lower to $2500 for TE while FD keeps the same $4500 floor.
There are also some textural differences in the salary distributions on each site. Top tier QBs tend to have salaries in the $8000 to $10000 range, while DK has a tighter distribution, where almost all QBs have a salary of $8000 or less on DK. TE salaries on FD are basically all in the 4500-5500 range besides a few guys, while DK has a much wider salary range. The salary textures for both RB and WR are pretty close to the same.
With these factors in mind, I’ve made a tool that compares salary values at each site using Z-Scores. A Z-Score is the number of standard deviations away from the mean an observation is in a given group. It’s used to measure the relationship between an observation and the mean of a group. An extremely simple example: If you have a data set with a mean of 3 and a standard deviation of 1, 6 would have a Z-Score of 3.0.
When calculating Z-Scores for DFS NFL players, I’ve had each position in reference it’s own group, besides RB and WR, where salary structure is so similar that it makes sense to look at the data together. However, before the Z-Scores are calculated, I make a salary adjustment to take into account the half point more per reception you get on DK compared to FD. I make a very simple receptions projection for each player, and subtract $200 from their DK salary per expected reception. This number is somewhat arbitrary, I’m basing it on a general 2.5 DK points per $1000 of salary target for each player. With a receptions adjustment, wide receivers who catch a lot of balls will have a salary that is appropriately more appealing on DK.
How To Use the Z-Score Chart
Using that chart is really easy. The far left column gives the differences between Z-Scores for players on each site. Positive numbers indicate a more favorable salary on DraftKings, while negative numbers represent a salary that is more favorable on FanDuel.
Let’s examine some salary comparisons for this week. Arguably the most favorable FD salary is Greg Olsen. Now Olsen’s salary is actually better in absolute terms on DK when we adjust for receptions (Although his $6500 DK salary is actually higher than his $6400 FD without an adjustment). But relative to the salary structures for each site, Olsen is very cheap on FD. You only have to spend a couple hundred dollars more in salary to upgrade to Olsen from a mediocre TE. His 1.69 Z-Score on FanDuel is less than his 2.29 Z-Score on DraftKings.
The best DraftKings salaries are Heath Miller and Stefon Diggs. Both of them already have small DK salaries, but with a high receptions per game their effective salaries become minuscule. Both seem like great plays, with Miller having a great matchup against Oakland and Diggs looking like one of the top receivers in the league the past several weeks.
It’s interesting to note that in general, you have much better salary values on DraftKings than FanDuel. This makes sense because the min salary is lower on DK. But FanDuel’s salary algorithm clearly a lot more aggressive at raising prices and keeping them high. They’ve moved Todd Gurley’s salary this week all the way to 9000 this week, while DraftKings had nowhere near as drastic as a salary increase.
For questions and comments, head over to the discussion thread on the forum. Special thanks to schoenmh who suggested using Z-Scores and helped me code it up.View all posts by Daniel Steinberg