Does Over Under impact NBA fantasy performance

On the surface this question seems relatively easy to answer.  Does the posted Vegas over under impact NBA fantasy production?  Thinking about it in general terms, you would assume that naturally, in higher scoring games, there will be more fantasy points scored.  I dug through 25,000+ game logs from the 2015-16 season to help answer this question.  We will be using the results of this research to help improve our already popular FTA+ daily NBA cheatsheet.

Does Over Under impact NBA fantasy production?

Section 1: Analyzing all game logs

The first thing I needed to do was pull all the data from last season’s NBA game logs (26,000+ entries), which was helpfully provided over on reddit.  I then whittled down the list to players who played in 20 minutes (helped to eliminate early injuries and bench players which could sway the stats).  The end product was this google spreadsheet which you can peruse at your leisure.

I then put together a simple formula which appears to the right on each spreadsheet, which gives us the average fantasy points produced for each posted total.

The first two tabs on the spreadsheet labeled “Overall” and “All” gives us the breakdown at each 0.5 pt from 179.5 all the way to 234.5.  Since it is only one season of data, not every total is represented, but you can see the trendline definitely moves up the higher the Over/ Under goes

Over Under impact NBA fantasy production


Conclusion for section 1: Does Over / Under impact NBA production? Seems to be a pretty obvious yes as the numerical values clearly climb based on the data provided.  However, more research is required…

Section 2: Breaking it down by position

The next question I wanted to answer is… does over under impact NBA players based on the position they play? Do PGs suffer when playing in a low scoring game more than a center? Perhaps there are more rebounds to go around that help to boost center production?

The main problem with the data at this point, is we have numerous players listed at G, F, G-F, F-C and so on.  To try to help clear up the data I listed the following positions for each tab:

  • G = G, PG or SG eligibility
  • F = G-F, PF, SF, F-C, C-F, G-F eligibility
  • C = C eligibility

So, does over under impact NBA players differently based on their positions? Here are the charts for G, F and C




So, here is where the data starts getting interesting.  While the sample sizes are shrinking, we clearly see that guards playing in lower over/under games are impacted the most.  This makes sense as they rely primarily on scoring and assists for their production.  Forwards still experience an impact although their scoring flattens out a bit more.

Centers are the most interesting, as there isn’t a noticeable trend.  While you would look at the spikes on both ends of that graph, keep in mind sample sizes are extremely small at the upper and lower bounds of over / under.  I’m far more interested in seeing the correlation that is shown between 190-220.  The average largely remains around 29 fpts, with no clear rise in production.  This tells us that true centers aren’t heavily impacted by the projected scoring / pace, which will certainly aid us when building those Fanduel NBA lineups.

Conclusion: Does Over Under impact NBA players at each position differently? Yes! Guards seem to be the most heavily affected, while the big men experience little to no difference in projected scoring.

As we move in to the final section of this article, things get really interesting. It occurred to me that the top point guards like Steph Curry and Russell Westbrook don’t play in many low scoring games.   In fact, the lowest posted total for a game in which Steph Curry played last season? 200! So perhaps the data itself is skewed as many of the stud point guards play for fast paced / high scoring teams?  I decided we needed to break this down by player and see how the over / under impacted them.

Section 3: Breaking it down by player

I actually ran over the limit of what google sheets is capable of as I tried to extrapolate as many players as possible.  The idea here was to check different guards from around the league, and see how their scoring fluctuates at each data point (over / under).  Keep in mind the sample sizes continue to get smaller, which leads to higher variance and less reliability in the data.

I settled on comparing 6 different guards that play for different paced teams: Goran Dragic, Isaiah Thomas, Stephen Curry, Rajon Rondo, Kemba Walker, Demar Derozan.  Lets see how their numbers look:

  1. Goran Dragic (Did not play in a game with an over / under higher than 219.5)image-5

  2. Isaiah Thomas (Did not play in games lower than 190 or above 225)image-6

  3. Stephen Curry (Did not play in a game with a total lower than 200)

  4. Rajon Rondo (Did not play in a game with a total under 195)image-8

  5. Kemba Walker (Did not play in a game with a total over 215)image-9

  6. Demar Derozan (Did not play in a game with a total over 215)image-10

Analysis: No players outside of Goran Dragic and perhaps Rajon Rondo showed any correlation between the over / under and their fantasy production.  While section 2 clearly showed that average fantasy production does increase between games, it seems this is simply due to the fact that higher producing players such as Westbrook and Curry play in those uptempo / high over under affairs, while players like Raul Neto and the Jazz will appear in the lower over under games.

Article Conclusion: Does Over Under impact NBA players production?

Based upon my analysis in section 3, I lean towards a surprising no!

As I mentioned in the intro, your initial thoughts would be that yes, production is clearly tied to the over / under and pace.  However, upon deeper analysis it appears that the individual players production doesn’t actually change as a result of the over / under.  If you expect DeMar DeRozan to score 35 fantasy points, and the over / under changes, this means you should not adjust your expectations. It appears that better players are in the higher over/under games, which gives the illusion that more fantasy points are being scored.  However, as you can see from the individual player charts, their production does not rise along with the over / under.

If you have any questions, comments, or think I’ve made an error in my analysis, feel free to post below.  Good luck in your NBA contests! If you are an FTA+ subscriber, don’t forget we will have live chat each and every day before tipoff to help you set those lineups and make sure you have the last second injury news the NBA is famous for.