Your browser does not seem to support JavaScript. As a result, your viewing experience will be diminished, and you have been placed in read-only mode.
Please download a browser that supports JavaScript, or enable it if it's disabled (i.e. NoScript).
Hello everybody, I've been running a predictive analytics model since '20 now to identify rating changes of players as well as an accompanying model that predicts increases/decreases in overall based on those rating changes. Last year I was sitting at a .7 r^2 for accuracy of updates and I'm constantly looking to refine my model.
It pulls in players' real-life stats over the past 3 seasons and compares them with samples of players who have been upgraded/downgraded per rating category, so theoretically the longer it runs and the more data I feed to it, the more accurate it will be.
Here's a few guys that it has getting upgraded, along with a % probability that they're heading upwards:
Bonus deep cut upgrade potential:
Christian Javier, based on my model, has a 94% chance of going gold and a 63% chance of going diamond.
Is there a roster update today?
@jmcdonald1817 said in Predictive Analytics for Roster Update Gems!:
Next week
Well, I just bought a shitload of Bryants yesterday at around 2.2-2.5K, so I hope you're right.
Knowing my luck, though, he'll pull a hammy and go on IR today.
He got Xander right