21st December Bookings Model Update
It’s been two weeks since the last update, but I feel there are a few changes to address, so I’ve decided to post an impromptu update.
December so far has been a positive month, building on from November, which has been great. I’m not going to focus too much on results in this article though – I’ll do another update in a couple of weeks hopefully, more detailed on how December played out.
First thing to address here, is the introduction of a new competition…
AFCON
I’ve been trying to get more involved in the international competitions this year – in the summer it was the Club World Cup, then the World Cup qualifiers during the international breaks, and now it’s AFCON.
I’ve set up the code to be able to pump out predictions, although I’ll be taking them with caution as the card data for the lesser-known teams is a bit rogue, as is the referee data. With all games televised in the UK, I really want to try and pay special attention to AFCON this year. Even if it is only going to be a tournament which takes place every four years, I feel like there could be a lot of lessons to learn, especially with the 2026 World Cup coming up.
So, predictions will be released, but take them with a pinch of salt for the entire tournament. Next is the big thing to address.
An Updated Script
If you read my previous article, you would know that I’ve been working to update my coding script to essentially use the data I have collected over the last year or so as the new source, rather than scraping an external website. I mentioned how this would take some time, and set myself the target of getting it sorted by the end of the year.
Well, I’ve managed to get everything working slightly earlier, which is ideal. I still want to add a few useful features which I have thought of along the way, but the main foundation code is functional and works well.
This allows me to be a lot more efficient with predictions, while also hopefully improving accuracy and removing some of the inefficiencies with the project. I’m going to take some time to track how things go, whether we can maintain a similar accuracy, and hopefully keep pumping out winners.
A positive update, and I’ll hopefully be ready to start sharing predictions very soon. I’m hesitant to put an exact date on it, but the 25th January 2025 was the first day of tracking this model, and I did say I wanted at least a year of data to go off. So lets pencil that date in for now, and when we get there, I’ll re-evaluate the progress!
As always, I appreciate your support on this project if you’ve read this article, read more than one of my articles, or even just been in the group. It’s come a long way, and will keep progressing.
