The 2020 F1 season will begin at the Red Bull Ring in Austria in a few weeks, and it can’t come soon enough. The virtual racing has been fun, but it can’t really compare to the real deal - even if there won't be any spectators for the first eight races.
And while this season is different for many reasons, you’ll be seeing extra changes in the form of more on-screen graphics. New ‘car performance scores’ will take live data from on-board sensors - each F1 car has 300 of them - which is then crunched on Amazon servers and then presented in a way that anyone can understand.
While purists want to see the raw data synced perfectly with the on-screen video, the new rankings will offer a data-based insight into how each car (and driver) is performing both overall and on specific parts of the track. These, naturally, will complement the commentary from each broadcaster.
The data will use four key metrics: low-speed cornering, high-speed cornering, straight line, and car handling (oversteer and understeer). After being processed, the stats are presented as a score out of 10, with 5 being average and 10 meaning that the car should be ‘unbeatable’ on those parts of the track.
You’ll be able to see cars’ relative performance and also where a particular car’s strengths and weaknesses are. The variety of graphics will compare multiple cars as well as two direct rivals and indiviuals.
F1 has been working with Amazon for this type of data for a while now, having shown graphics last year for metric including predicted pit stop strategy, battle forecast and tyre performance.
The six new ones are:
- Car Performance Scores
- Ultimate Driver Speed Comparison
- High-Speed/Low-Speed Corner Performance
- Driver Skills Rating
- Car/Team Development & Overall Season Performance
- Qualifying and Race Pace Predictions
Some of these will factor in historical data (70 years’ worth is available), and some, such as the Driver Skills Rating, won’t appear until the second half of the season. This will basically show you who is the best overall on the track at that moment, being based on data such as tyre management, race pace, plus overtaking and defending styles.
Predictions, likewise, will be drawn from data from practice and qualifying laps, and it’ll be interesting to see whether this machine learning will be able to accurately predict which driver will qualify on pole and which will win the race.
Whether the raw data is being obfuscated for the viewers’ benefit or to prevent rival teams from seeing it we’ll probably never know. But for the more casual viewer, the graphics will hopefully be more than just a gimmick.
For more on the upcoming season, here's how to watch F1 races.