I was pulling for Jimmie Johnson to make the playoffs last weekend. I cheered for him partially out of self-interest (go old people!) and partially because his last two seasons have been so tough and I don’t want to see him go out on a down note.
I was also pulling for Daniel Suárez to make the playoffs because he doesn’t have a contract for next year yet, and also for Clint Bowyer because he came so close in 2012 and (also) doesn’t have a contract for next year yet…
But it was not to be for Johnson or Suárez. Both drivers said all the requisite things afterward: There are still ten races left and we’re gonna focus on winning them.
But how likely is that for drivers who didn’t make the playoffs? And, for that matter, what about the six drivers entering the playoffs without any wins?
A Case Study: 2018
Let’s start by looking at last year because we were working with the same rules. Everyone who won a race was in the playoffs, plus six drivers who got in on points.
Which Drivers Won?
Here’s a pie chart that sums up when drivers won — or didn’t
This is a little confusing: I know because I confused myself a couple times. The sixteen drivers break down into four groups.
- No Wins: Four drivers (25% of the drivers in the playoffs) got in on points and didn’t win any of the 10 playoff races.
- Wins Everywhere: Five drivers (31%) drivers won regular season (RS) and playoff (PO) races.
- RS Wins, but No PO Wins: Another five drivers (31%) won in the regular season, but didn’t win during the playoffs.
- No RS Wins, but PO Wins: This is the one we’re really interested in. Only two drivers got into the playoffs on points, and then went on to win races during the playoffs. (That’s 13%)
What About the Races?
Since we’re interested in how likely it is that a driver wins a race, we need to look not just as what the drivers do, but how many races each wins — or doesn’t. This will end up getting a little confusing, so I hope my color code helps make it clearer.
- BLUE colors denote regular season (RS) races
- RED/ORANGE colors denote playoff (PO) races
So here’s what 2018 looked like, graphically.
- Of the 26 regular-season races
- Drivers who won both RS and PO races won 17 races
- Drivers who didn’t win in the POs won 9 races
- Of the 10 regular-season races
- Drivers who won both RS and PO races won 8 races
- Drivers who didn’t win in the RS won 2 races
This data implies that those drivers who haven’t won this year have about a 20% chance of winning a post-season race.
But That’s Just One Year of Data
Yes. It is, and good for you for noticing. But here’s the problem: NASCAR keeps changing the rules. I’m not complaining: they’re trying to make the sport more competitive. But that just wreaks havoc with data analysis.
Just as a reminder, so you have some context, there have been (to date) five different schemes for The Chase/the playoffs.
- The Chase/playoffs began in 2004 and involved all drivers in the top 10 in points, plus any drivers within 400 points of the leader.
- In 2007, the field expanded to the top 12 drivers and they got rid of the ‘within 400 points of the leader’.
- 2011 brought another switch: 10 drivers + two wild-cards: the drivers who had the most wins, but were outside the top 10.
- Starting in 2014, NASCAR introduced win-and-you’re-in, with 16 drivers in the Chase/playoff system, plus elimination stages.
- It wasn’t until 2017 that we got the current playoff scheme with stages.
What’s the Probability Using Only This Iteration of the Playoffs?
If we limit ourselves to 2017-2018 (where the playoff rules were the same as they are this year), we have two drivers in 2018 who only won in the playoffs and one driver in 2017, so the effective probability is 3/20 = 15%.
It Looks Pretty Good Then, Right? 15%?
Fifteen percent isn’t bad.
But is it a valid number?
Let’s look at the previous three years, where we had everything the same in the playoff scheme except for stages.
For three years in a row, no one won in the postseason who hadn’t won in the regular season. Putting the last five years of data together…
If we use the last five seasons on the argument that they are most similar to 2019, then drivers we were winless going into the playoffs won 6% of playoff races.
But this graph also raises another issue that will shortly become even more important. See that hashed section in 2017? That’s Joey Logano, who won a regular season race, but his win was encumbered. So I needed to make another distinction in my graphs.
- SOLIDS indicate races won by drivers who finished in the top 16 (i.e. made the playoffs).
- PATTERNS indicate races won by drivers finishing outside the top 16.
- A blue hash = a driver who finished outside the top 16, but won a regular season race.
- A red/orange hash = a driver who finished outside the top 16, but won a playoff race.
What About Going Further Back?
I went ten years back, with the caveat, again, that we’re getting into different playoff schemes, so the data may be less relevant.
Did Non-Winners Ever Win the Majority of the Playoff Races?
In 2011, Tony Stewart barely made The Chase, but had no regular season wins. He then went on to win five Chase races — and the championship.
Drivers who had won in the regular season only won three of the final 10 races that year. Drivers who hadn’t won in the regular season won seven races: Stewart’s five, plus one each from Clint Bowyer and Kasey Kahne.
But That’s an Exception, Right?
Well, let’s look at the number of postseason races won by new winners for all ten of the last years.
It’s not too surprising that the number dropped once NASCAR introduced win-and-you’re-in. This graph represents 20 drivers over ten seasons, which amounts to 20% of the playoff races won by drivers who hadn’t won in the regular season.
But given that those larger numbers come from a different set of rules, I’m more comfortable with the 6% figure.
So… 6% Then?
Well, not exactly.
The graph above fails to make an important distinction if we’re going to talk about the chances of, say, Jimmie Johnson or Daniel Suárez winning a race vs. Ryan Blaney or Clint Bowyer.
What about Post-Season Winners From Outside the Top 16?
The last three non-RS winners have all been from within the final 16. So do drivers from outside the top-16 ever win playoff races?
Here’s ten years of data. We’re looking for a winner not in the top 16 at the end of the year, which would be represented by a hatched area at the far right of the bar.
Do you see them? They’re small (1 race each), but they’re definitely there: one in 2013 and one in 2009. And the results aren’t very encouraging for those drivers sitting in positions 17 and up.
Over the ten most recent seasons, drivers outside the top sixteen won only two playoff races (2%).
2%??? Really? That’s All?
Actually, no. Let’s look at that 2013 point, which is Denny Hamlin.
Hamlin missed four races in 2013 due to medical issues. That was before the exceptions and exemptions we have today. Hamlin finished 23rd that year. So that data point sort of doesn’t count because if Hamlin hadn’t missed those races, he very well might have finished in the top 16.
That leaves us with only the one race in 2009. Before we get excited, let’s check and make sure it’s a legit data point.
It is! Jamie McMurray finished 23rd in the standings that year, but managed to win his first race in the fourth-to-the-last race of the year.
So the actual number is 1%, but that comes from data a long time ago in a playoff system far, far away.
So No One Outside the Top 16 Can Win?
Unfortunately for you 41, 48, 21, etc. fans out there, the reality is that the probability of your driver winning one of the next ten races is pretty doggone low.
But that doesn’t mean it’s impossible. Because who predicted this year’s list of winners would include Justin Haley? One percent means that the event would happen once in a hundred times. There’s nothing saying that Las Vegas this weekend isn’t that one time.
And if one of those drivers does win, this analysis shows that the win ought to be even more respected because it’s a pretty amazing feat.
Where Are the Best Chances for a Win for Those Drivers Outside the Top 16?
Road Courses and Plate Tracks: Blaney won the Charlotte Road Course and Almirola won Talladega last year. Our solitary data point (McMurray in 2009) won at Talladega. My analysis shows that these two types of tracks are wild cards and represent a much better chance of winning for less-likely-to-win drivers. So look to Talladega, or the Charlotte Roval.
Johnson almost won last year at the Roval, so that’s looking like #1 on my list of highest-probability tracks for him. But remember that Blaney (in the playoffs this year) won the Roval last year.
Johnsons’ best finishes in the regular season are at mile-and-a halfs, so you might look for him to do well this week, too.
THANK YOU to the great folks at racing-reference.info for all the info they provide and keep up to date. Their resources make doing this infinitely easier and faster.