This is a really fun article for me to write because one of my first “big goals” when I started writing about IndyCar statistics was to apply the Elo rating system to IndyCar drivers. Elo ratings were developed by Arpad Elo to rank chess players, but they have since been used to rank sports from soccer to football to basketball and more. Elo ratings have a couple of great qualities that make them a good choice to rank and compare IndyCar drivers. The first part of this article will detail how the ratings are calculated, but feel free to skip past that for the results or come back to it later!
Today I just wanted to drop a very brief article here on a question I recently looked at. How useful is starting position in terms of predicting finishing position for different tracks? This is a slightly different question than the three I explored in this article on qualifying position, but if you’re interested in qualifying and its relationship to the race I urge you to check out that post as well.
Throughout the 2019 season I kept track of a stat called Expected Points (xPoints). xPoints is the number of points we would expect a driver to earn in a race based on their average track position during the race. The intuition behind xPoints is that crashes, mechanical failures, slow pit-stops, and more “bad luck” don’t reflect a driver’s true skill: these sources of bad luck are factored into traditional stats like average finishing position and the points table overall. A driver’s true skill can be measured by how they ran throughout the entirety of a race, not just by how they finished — or didn’t finish.
So, does measuring xPoints add to our IndyCar knowledge?
The IndyCar field is full of part-time drivers. Even outside of the plethora of drivers who just compete in (or attempt to compete in) the Indianapolis 500, there are a lot of drivers who are on partial contracts for the season and might only compete in a hand full of races. Most drivers don’t race part-time because they prefer part time, but rather because they haven’t gotten the opportunity to race in a full season yet or haven’t proven themselves. Ed Carpenter is the obvious exception to this rule since he chooses to only race on oval tracks.
Earning the most points throughout the season is the goal of IndyCar racing. Points are how a driver wins championships and every decision a driver and team make throughout the course of a season should be centered around how they can maximize the number of points they’ll earn at the next race. Using the stats I tracked this IndyCar season, we can look at how correlated different stats are with earning points over the course of the season. Correlation is a measure of how strong or weak a relationship is between two variables. It can range from [-1,1], with -1 being a perfect strong negative correlation, 1 being a perfect strong positive correlation, and 0 meaning there is no correlation between the variables. If the correlation between two variables is 1, that means when variable x goes up, so does variable y. If the correlation is -1, that means that when variable x goes up, variable y goes down.