Dr. Diandra: Predicting Kansas contenders
Last week, I developed an algorithm to predict the first four drivers to be knocked out of the playoffs. Since we have to wait two races before evaluating how well that algorithm worked, let’s try something else.
Let’s try predicting the contenders in today’s Cup playoff race a Kansas Speedway (3 p.m. ET on USA Network).
Predicting Kansas with only one Kansas race
Picking drivers based on only May’s Kansas race won’t give us very robust results. So we’ll add a couple of comparable tracks that have already hosted a race this year.
The track most similar to Kansas is Las Vegas Motor Speedway. Both are 1.5-mile, D-shaped tracks with low tire wear. Las Vegas has progressive banking between 18 and 20 degrees while Kansas’s banking ranges from 17 to 20 degrees.
The other track is Michigan International Speedway. Kansas and Michigan share the same shape and level of tire wear. They’re different lengths, but only by a half-mile. The corner banking at Michigan is 18 degrees, similar to Kansas’s turns.
I weighted averages to make the Kansas data equal to the other two tracks combined. Basically, I added Kansas twice and Michigan and Las Vegas once and divided by four. When I refer to average finishing position or average running position for the dataset, I mean the weighted averages.
I developed the algorithm using the same techniques I described before, but valued winning a little more here. I used the driver’s average finish over the last five races to capture momentum. For reference, all drivers but one have average finishes of 10th or worse in their last five races.
The top three contenders
No one will be surprised to see Kyle Larson at the top of my list. He’s at the top of a lot of lists. The graph of weighted finishing positions below shows his 3.25 average finishing position. That’s half the average finishing position of the second-place driver, Martin Truex Jr.
Larson enters Kansas confident, but realistic about the level of competition he expects from Toyota.
“I’m sure they’ll get the pole,” Larson said. “I’m sure they’ll be in the hunt all race long. It’ll be tough, but they can be beat, for sure.”
Larson is one of only three drivers to have top-10 finishes at all three tracks. The other two are Truex and Alex Bowman. Although Larson didn’t win any of the three races, he put up a fierce battle with eventual winner Kurt Busch at Kansas in May. He also led 7.6% of all laps run at these three tracks this year.
Drivers two and three score well behind Larson, but very close to each other.
Although Christopher Bell had a DNF at Michigan, he still had the fifth best average running position for that race. Bell led 13.5% of laps at the three tracks in 2022, the second highest total of any driver. And he’s one of only two drivers to lead laps at all three tracks this year.
Denny Hamlin has the second-best average finishing position in his last five races at 10.8. That boosted him to third place in the Kansas prediction list. Despite not finishing the Las Vegas race due to a drivetrain problem, Hamlin led 9.3% of all possible laps at the three target races.
The last of the top six Kansas predictions
The final three drivers are tightly grouped in score, but below Bell and Hamlin.
Ross Chastain takes fourth place. He’s led the most laps of any driver (15.7%) at our three-track sample. He also led each one of the three races we’re considering. Bell is the only other driver who’s done that.
“It’s pretty wild that our wins have not come on 1.5-mile tracks,” Chastain said, “because I feel like those have been our strongest races.”
Strongest or not, Chastain’s position on the list isn’t higher because his average finishing position for the last five races is 23.2. That’s the worst of any of the top-nine drivers the algorithm identified.
Las Vegas winner Alex Bowman comes in fifth, with a 7.0 average finish position. Although he had a series of P30+ finishes in the middle of the season, his average finishing in the last five races is 13.4. He’s also one of the three drivers to have top-10 finishes at all three races in the sample.
No. 6 on my list is Kevin Harvick. Aside from winning Michigan, he doesn’t have a single standout stat like the other drivers. Just solid numbers in every metric. If he can channel last week’s frustration into a strong run, he just might solidify his spot in the next round of the playoffs this Sunday (3 p.m. ET on USA Network.)
With the second-highest average finish at the sample tracks, how did Truex rank only seventh? He finished in the top 10 in all three races, but never cracked the top five. He only led one lap between all three tracks. But if you’re looking for a driver to run top 10, he’s a solid choice.
Joey Logano is on a hot streak with the best last-five-race average (5.8). But his average finishing position at these tracks is 13.0, with zero laps led.
Chase Elliott, who had a disappointing 36th-place finish at Darlington, has a 19.5 average finish at the three comparable tracks this year.
Chase Briscoe and Daniel Suárez have average finishes of 25.8 and 32.0 respectively. While Suárez has shown improvement since his Sonoma win, his average finishing position at the last five races is still only 18.2.
Kansas predicts tires for the playoffs
Goodyear continues to develop the 18-inch Next Gen tire. Just as teams have learned a lot about the car during the season, so have the tire engineers.
They’ve changed the construction and the tread compounds for Kansas relative to the May race. The right-side tire code was run at Pocono and Michigan, but the left-side is new.
Texas and Las Vegas are scheduled to use the same tire codes. So in addition to points, teams will scramble to gather tire data in the hopes they’ll be able to use it to their advantage in the fourth and seventh playoff races.