As stated in one of my previous posts, I will show the ratings that my model predicted for the 2021 Russian GP. Just as last time, right now I’m just posting the predictions created by the model that I created a few races ago. Once the media outlets release their actual ratings, I will update this post to show the performance of our model.
I decided to now give the top 2 predictions that the model spits out in order to give it a fighting chance. Let’s see if this makes the model’s results a bit more exciting.
It wasn’t our greatest day. It’s pretty clear that the model lacks information to make proper predictions under changing conditions (losing positions in the last few laps due to a massive downpour for example).
On this occasion, we got the rating right only for 18.5% of the ratings that were given out. With our second prediction, this number increased only to 36.1%, which is not satisfactory if you ask me. Having said that, the model is making predictions without any human input. Yes, I provide the model with the data, but I don’t influence the model in any shape or form.
You can see that for drivers like Norris and Leclerc, the model made very wrong predictions. This makes perfect sense when you think about it. How is the model supposed to know that Norris was leading and lost the race after making a gamble? A more robust model could’ve perhaps made a better prediction, but then again, it would’ve been more time consuming to create a model like that.
First of all, thank you for taking the time to read this article. I will update this post once the ratings from the media outlets are released. No matter how or model did, good or bad, I want to show you that the world of analytics is not perfect, and that sometimes there are obstacles on the road.
If you want to support me, there are multiple ways you can donate some money to help me keep this project alive. You can find all the options on either the about tab or the my supporters tab in the main menu.