This is part five of a multi-part series intended to help both newcomers and veterans learn profitable Daily Fantasy NASCAR strategy. In this series Nick Giffen (@RotoDoc) teaches you everything you need to know to jump right into the sport and succeed. Nick holds a Ph.D. in mathematics, has won multiple NASCAR GPP tournaments, and is a two-time qualifier for the DraftKings NASCAR Main Event Live Final – the King of the Speedway.
After mastering the art of finding dominators and high-floor drivers for your cash game lineups, it’s time to dip those toes into the guaranteed prize pool (GPP) waters. With the increasing pay scale for finishing higher in GPPs, the goal is to win and win big. As a result, we’re not concerned with downside or “what’s likely,” but rather, “what’s possible.”
Due to the increasing pay scale, we are focused on upside and ownership percentages. In this piece, we’ll focus on upside, with ownership in the next GPP strategy article. Starting with upside, there’s a preferential list that I use when trying to find drivers’ upside. It goes, in order:
- Front Runners
- Salary Relief Movers
- The Rest
So what do each of these categories mean, and why are they ordered this way?
Just like cash games, it is important to find the driver that dominates the race. The difference for a GPP is that finding the dominator is a prerequisite to winning, or even placing anywhere near the top.
The difference between cash and GPP, however, comes when the number of laps are factored in. As the number of laps increases, it becomes more and more important to find the top two dominators. The graph below shows the DraftKings (DK) points of the second highest scoring driver from each race, plotted against the number of laps.
From the fewest to the greatest number of laps, there is a difference of approximately 40 DK points. But how does that compare to the rest of the field?
To make that comparison, I simply looked at the average number of DK points remaining among the next 20 highest scoring drivers, because in any given race there are about 10-15 drivers that are unusable. We’ll use that as a proxy for a replacement level driver.
The change in average DK points from the fewest to the greatest number of laps was only seven points. More notable, at the races with the fewest number of laps, missing out on the second highest scoring driver only cost approximately 21 points by fielding a replacement rosterable driver (60 compared to 39), whereas at races with the greatest number of laps this cost 54 points (100 compared to 46).
The next most important kind of driver to roster is the back-to-front driver, or what I call a “mover.” What do I mean by this? Simply a driver that qualifies poorly, but has the talent and ability to finish well. These drivers rack up points for a good finishing position, as well as a lot of place differential points. A few of these drivers are even fast enough to pick up a significant amount of dominator points.
To illustrate the value of these drivers, I took a subset of the drivers that qualified 25th or worse, but finished 10th or better. Then, I removed the dominators. The absolute minimum scored by any of those drivers was 49 points, which is two points more than a replacement level driver in a 500-lap race.
Additionally, 86 percent of these drivers finished with a top six DraftKings score. The average production of the movers was 66.5 DK points, with an average of 64.5 and 69 in races with the fewest and most laps, respectively. This means, on average, these drivers are 23-25.5 points better than a replacement level driver, depending on race distance.
There have been 1.2 drivers per race that fall into this category since the start of the 2013 season, when NASCAR introduced the current style of car. These drivers account for 6.2 percent of all drivers that have started 25th or worse over that time frame.
So how to we find the movers? I ran a logistic regression on drivers starting 25th or worse to determine what factors led to a higher probability of one of these drivers finishing inside the top-10.
As it turns out, the most important factors were:
- Driver Rating (DRtg) over last eight races at similar tracks
- DRtg over the last 15 races overall
- Quality Pass percentage (QP%) over the last eight races at similar tracks
- QP% over the last 15 races overall
By using principal component regression, I was able to determine the probability of a driver finishing inside the top-10 based off of these four factors. To have a 50 percent chance of finishing top-10 a driver needed to have all of the following:
- DRtg of 80+ over the last 15 races
- DRtg of 85+ over the last eight at similar tracks
- QP% of 35+ over the last 15 races
- QP% of 45+ over the last eight at similar tracks
Now, there were only 58 drivers out of 108 races that fell into this category, so you won’t be able to find these guys every race. In that case, you may need to fill out your roster with a few more of the next type of driver.
The next category is the mid-to-front drivers, which I call the “contenders,” because they can contend for the occasional race win despite starting mid-pack. I classify these drivers as drivers who start anywhere from 13th to 24th, and finish inside the top-eight.
After removing the dominators that fall into this subset, these drivers average 58 DK points with a minimum of 41. Still well above a replacement level driver, regardless of lap count.
These drivers make up 19.75 percent of all drivers starting in positions 13-24 from 2013 to present. On average, 2.4 drivers per race fall into this category.
The same factors determine which of these drivers will rise to the front, DRtg and QP% overall and at similar tracks.
4. Front Runners
These are the drivers that didn’t dominate the race, who finish inside the top-1o, but who also qualified well enough that they don’t fall into any of the mover categories. They average 47.25 DK points, because they don’t get any points for place differential.
Front Runners make up 51.5 percent of all drivers that finish in the top-10. These tend to be your strong, high-salary drivers so it’s a trade off between starting these drivers and the next category of drivers, as their expected points are both approximately replacement level.
5. Salary Relief Movers
You’ll round out your roster with the back-to-mid drivers. Typically these drivers come at a discount, so I call them “salary relief movers.” I classify these drivers as starting 25th or worse who make up at least 10 spots of place differential and finish between positions 11 and 24.
On average, there are 3.1 drivers per race that fall into this category, with 16.5 percent of all drivers starting 25th or worse. These are your startable replacement level drivers. They score 44.4 DK points on average, with a minimum of 30 points.
With these drivers, the most important factors are average finish over the last 15 races, and having a better final practice rank than qualifying rank.
Depending on the lap count and salary requirements of each driver, you’ll typically want to fill a GPP lineup with these driver types. Which of these possible drivers you select may depend a bit on ownership percentages, which involves a bit of game theory. I’ll cover that in the next GPP strategy article.