NASCAR DFS: Analyzing the Winning Lineup from a Wild Daytona 500

The Daytona 500 is now in our rear-view mirror. Kurt Busch came away with the win in NASCAR’s biggest race, adding to what is likely a Hall of Fame resume. I’m going to review the winning DraftKings lineup, as well as my 8th place lineup from the $300k contest, and give some brief thoughts on my picks and model results for this past weekend’s race.

As part of my commitment to you, I’ll be keeping track of my weekly results. Overall it was another solid week for me. I didn’t bank $26k profit like last week, but a $5k profit is nothing to sneeze at.

DaytonaScreen_LI (2)

(I hope you enjoyed the corrected watermark)

Let’s dive into the optimal lineup from last Sunday’s results, as well as how my model fared.

The Winning Lineup

There was a four-way tie for the optimal lineup in the $300k contest…congrats to the winners! Nobody in the $500k contest ended up on this lineup, so the winning score in that one was a bit lower at 412.75 points.


All six drivers started 30th or worse. And while I didn’t expect that strategy to win, I certainly said it was a viable strategy that could win in both my strategy article and my RotoViz Live stream before lock on Sunday. How did this happen? A bunch of wrecks took out front runners, including basically every favorite to win. That’s the exact kind of chaos we try to embrace at these restrictor plate tracks.

The chalk hit pretty hard, with Martin Truex, Jr., Ryan Blaney, A.J. Allmendinger, and Paul Menard all ending up in the winning lineup. I listed all but Allmendinger in my cash picks, and I pointed out on the RotoViz Live stream that Allmendinger was a complete oversight when writing, and I used a lot of him.

I wrote about both Gaughan and Waltrip in the GPP section, and also highlighted Gaughan in my strategy article. Safe to say, the RotoViz NASCAR subscribers — especially ones who used the new NASCAR Multi-Lineup Optimizer — had a strong day.

How did my model fare?

Model Results

Surprisingly well, for Daytona. The actual vs. predicted finishing position was craptastic, but that’s to be expected, especially in a race with as many wrecks as there were here. However, when all that randomness comes into play, the place differential becomes predictable. So in terms of actual vs. predicted DraftKings points, the model did great! And that’s what we care about, right?

2017-03-01 (1)

In fact, as fellow RotoViz editor Ben Gretch pointed out, the model nailed each of the top seven in DK points, and 10 of the top 13. Not too shabby.

I don’t want to sound too high about it though. A lot of it is luck, and certainly my model had some elements of luck to it, seeing as it’s the best it’s performed for any Daytona race so far.

However, I’m excited to roll out the improved version of my model for non-plate tracks this week at Atlanta. Stay-tuned for the Atlanta preview/strategy article out later this week, and then the picks and model projections article out after final practice.

Author Details
Co-Owner and Editor-in-Chief at RotoViz
Co-Owner and Editor-in-Chief at RotoViz. Mathematics Ph.D. 3x qualifier for the DraftKings NASCAR Main Event.
By RotoDoc | @RotoDoc | Archive

No Comment

Leave a reply