The Race Prediction Machine

Predictions Before Every Round

The same love for data that drives us to make fitness watch mounts started us down the path of seeing what we can do for forecasting race results. The Race Prediction Machine (RPM) uses machine learning based on data back to 2013 to predict what it thinks the most likely finish order will be. Before every round of AMA Motocross and Supercross, the RPM is fed things like the qualifying data and the weather conditions as shown at the bottom of the page. We then post its resulting predictions right here.

Of course motocross and supercross racing is difficult to predict... that's why we like it! The RPM can't be expected to get it perfect, but it gives its best shot at what it thinks the most-likely outcome will be. We do not change any of its predictions even if we don't necessarily agree or understand what it's seeing in a given prediction. Click here to check out the RPM's past results.

 

How to Use the RPM's Predictions (Short Video Autoplays)

   

What the RPM takes into Account

After every race, the RPM takes in the latest results and trains to better understand what matters most in making its predictions. While the exact percentages below will vary slightly from week to week, the graph below shows how much each of these factors weighs into its decisions. It should also be noted that some of the more incident-prone riders and riders without much historical data are more difficult to predict, so keep that in mind when using the RPM to help guide your fantasy picks!