In the announcement and follow-on coverage of the return to the Chase Format, NASCAR discussed its analysis of the format and ran simulations to better understand it. While they’ve provided some high-level results from this work, I wanted to dive deeper, so I’ve run the same race simulations that powered last season’s Playoff simulations through the new Chase format, which allows us to explore the results further. As NASCAR described in their process, these are all anonymized to avoid ascribing biases to drivers.
First, I looked at when the championship will be decided. I looked at this in two different ways, the Mathematical decision: when it is mathematically impossible for another driver to score enough points to beat the leader, and second the Probabilistic decision: when, according to the model, the champion has a 95% chance of winning the championship given their current points position and the likely points scored by other drivers going forward.
In both cases, the most likely outcome is that the championship is decided in the final race. The mathematical decision has a 7.5% chance of occurring in the penultimate race, while the leader has a 20% chance of having probabilistically clinched after the penultimate race. There remains a slim chance that the leader probabilistically clinches all the way back to the fourth-to-last race. We should expect to see a driver clinch before the penultimate race at most once every 10 years. Those rare outcomes are an opportunity to celebrate a truly elite season and to continue building the narrative of a superstar driver.
Let’s drill down into the competitiveness of the final race and look at the distribution of the gap between first and second in points both entering the final race and in the final championship standings.
The median gap between the top two championship contenders is likely to be 22 points, while the modal gap is 11 points. To put that in context, 22 points can very plausibly be made up by winning the race and placing highly in both stages if the leader has a less-than-stellar day; we should expect to continue to see championships decided in the final race, and for championship battles to be close.
For some additional context, how should we expect the Chase to play out? Let’s look at the points gap for each position between 2nd and 8th (we’ll get to why 8th is important in a minute), over the course of the Chase. Each position’s median is the primary line; the shaded region is the Interquartile Range, the range of outcomes between the 25th and 75th percentiles. Position’s lines drop off the chart when it gets beyond the maximum points available in a race.
In the 50% of Chase simulations, it will be mathematically possible for drivers all of the way back to 5th to win the championship going into the final race, and while it’s unlikely that a 5th placed driver at that point in the season would win the championship; drivers in 2nd (22 pts back) and 3rd (42 pts back) both could very well upset the driver in 1st going into the championship if they had a particularly good day and if the driver in 1st had a particularly bad day.
With that in mind, let’s look at the biggest question: what is the likelihood that an “undeserving” champion emerges due to the format, and, relatedly, is the 16-driver field too big? To attempt to answer these questions, I looked at the probability of winning the championship based on your seeding entering the Chase.
The leader entering the Chase has a 21% chance of winning the championship, with that dropping to 13% for the driver in 2nd. The additional 25 bonus points provide a 7% increase likelyhood of winning the championship. Beyond 8th place, the odds of winning the championship drop below 5%, and 14th through 16th each have a 1.8% chance. So while I understand some fans frustrations with the apparent dilutive effects the expansion of the Chase has, the chances of one of those drivers winning the championship are incredibly low, and aren’t worth re-re-opening the charter agreement, every drivers contract, sponsor agreements etc. and the field size aligns with that of other sports ([tktk NFL, MLB, NBA]).
Finally, let’s see how the different championship positions in the seasons likely play out. The table below presents the average summary statistics for each final championship position throughout the season.
| Pos | Wins | Top 5s | Top 10s | DNFs | Avg Finish | Avg Stage | Finish Pts | Stage Pts | Avg Pts |
|---|---|---|---|---|---|---|---|---|---|
| 1 | 3 | 12 | 19 | 3 | 12.39 | 12.85 | 26.46 | 6.14 | 32.63 |
| 2 | 3 | 11 | 19 | 4 | 12.84 | 13.12 | 25.73 | 5.93 | 31.69 |
| 3 | 3 | 11 | 18 | 4 | 13.20 | 13.38 | 25.20 | 5.74 | 30.96 |
| 4 | 2 | 10 | 17 | 4 | 13.54 | 13.62 | 24.72 | 5.58 | 30.33 |
| 5 | 2 | 10 | 17 | 4 | 13.84 | 13.87 | 24.34 | 5.38 | 29.75 |
| 6 | 2 | 9 | 16 | 4 | 14.12 | 14.15 | 23.98 | 5.21 | 29.22 |
| 7 | 2 | 9 | 16 | 4 | 14.40 | 14.42 | 23.61 | 5.02 | 28.66 |
| 8 | 2 | 8 | 15 | 4 | 14.75 | 14.72 | 23.17 | 4.83 | 28.03 |
The champion typically has 3 wins and finishes one-third of races in the top 5. With an average finish of 12.39 and average stage points of 6.12, they’re consistently running in the top 10 and often in the top 5 throughout the season. They have some room to breathe, with an expected 3 DNFs throughout the season, but a fourth DNF is the biggest deciding factor between 1st and 2nd place in the championship.
The new Chase format, combined with a larger spread in points between first and second, does a good job of balancing winning with full-season consistency, while the points reset for the Chase allows drivers to DNF or miss a race without ending their season. My expectation is that we will have a competitive championship battle that better rewards a deserving driver, and I’m excited for a fresh start.
Leave a Reply