/cdn.vox-cdn.com/uploads/chorus_image/image/69937995/1342143318.0.jpg)
So, in the dying embers of having to use xG data from last season, the prediction model had a bit of a down week. Since then, they’ve rebounded with two solid matchdays in a row, and seem well on their way to be extremely reliable, getting 6/10 in matchday six and 7/10 in matchday 7.
Matchday 7
A bit late, but our Matchday 7 edition of the prediction model had a good weekend- 7/10 correct results and 1 correct scoreline. Hopefully the model can keep up its run of good form since dropping last season's data! pic.twitter.com/gQpNp4N9kK
— Charlie Tuley (@analyticsLaLiga) September 28, 2021
The only three misses were us drawing Madrid (which was seen as the least likely result and thus is a flat miss by the algorithm), Atleti losing, and Celta beating Granada, the last of which was actually the second most likely result.
Matchday 8
If you’re eyes dart directly to the Villarreal match like mine do, you’ll see that we are favored to beat Real Betis by both our algorithm and the oddsmakers. In what is probably the headline event of the week, Atletico Madrid are massive favorites over a flailing FC Barcelona and their surly, soon to be jobless manager. The biggest underdog of the weekend is Alaves, fresh off a massive upset over Atleti, looking to continue that run against Athletic Club.
One thing that has me eyeing these predictions in a new light is the fact that salary caps were announced this week, with Barcelona coming in 7th and Valencia coming in dead last. Real Madrid’s salary spending advantage is incredibly massive over everyone else in the league, but fortunately they were kind enough to commit the better part of a million Euros a week to Gareth Bale and Eden Hazard, so they have largely self mitigated for the rest of us.
We are sitting at 35/68 matches correctly predicted overall, which as I always say is pretty darn good when there are three possible outcomes per match. I’m looking forward to seeing how we do this week.
Loading comments...