Where have I been and What’s next
This blog has been way to silent and it’s been eating away at me for the last X years. So what better way to relaunch public interest in the site than writing about something only I care about? Seriously, not even my own grandmother will be able to slog through this monotonous topic nor will she be able to bear any more run-on sentences and cheaply attempted anecdotes.
There, have I lowered the bar enough? Good.
So let’s introduce the topic of interest, Analytics. More precisely the topic of racing analytics and why we deserve more. In my view there aren’t enough data science projects on that broach the topic. What you will find are articles telling us that there is a burgeoning market for data in motorsports. Articles that tells us how companies like SportMedia Technology(SMT) are helping to deliver metrics to fans in the hopes to deliver a more involved experience. And articles that tell us how F1 uses data for nearly everything that happens on and off the track, heck F1 teams are so ingrained in data that data management companies have become full fledge sponsors. But that’s it, just articles. And articles aren’t hands on. You can’t break stuff and learn how it works when all you have are blogs that tell us that these things exist, but to see it, well that’s proprietary.
El Plan
To satisfy my need for hands on projects I’m going to make my own and as a bonus I am going to take you on the journey. Be forewarned that although I may be a career data analyst I still consider myself a pretty poor analyst. So now that I lowered your expectations again here’s the plan, we’re going to pick apart into racing data from iRacing. For the unengaged, iRacing is a PC racing simulator featuring multiple disciplines of motorsport (road racing, oval racing, dirt ovals, and rallycross). iRacing also has a built in telemetry recorder that we’ll leverage for data gathering. Furthermore, there are a multitude of tools that can interpret telemetry data (We’ll go through the list of tools in a future article because boy are there a lot out there).
Things are coming together. We have our data source and a suite of tools to get us going. Now we need some direction for this project. The number of ideas are only limited by our imaginations but they should all serve to help us either make us faster drivers or better racers. These two items sound like the same thing but they’re not. Of course being the fastest driver on the track means lap by lap your times are faster but what if you are a really quick driver but poor at passing? Being the fastest driver doesn’t mean much when you’re stuck behind a slower car. How about being a fast driver but you burn through tires requiring you to pit early where you lose time to your opponents who don’t have to stop for fresh rubber?
For this reason we’re going to be splitting the any Key Performance Indicator(KPI) we create into two categories: Driver KPI’s and Race KPI’s. A driver KPI would be any measure that gives us insight to how to drive faster on a hot lap (a hot lap is when you’re the only car on the track and you are free to go as fast as possible without interruption). An example of a Driver KPI would be time spent at wide open throttle. A Race KPI would then be any measure that gives us insight on well you drive with other cars on the track. An good example of a Race KPI would be passing efficiency, which would measure how long does it take you to perform an overtake on a competing car.
So there you have it. We’re back in a very technical way. We have a loose plan mapped out that should turn out some interesting projects. Next article we’ll go through some of the tools we’ll be using and why I chose them.
In the mean time, let me know which Driver or Race KPI’s you’d like us to cover.