Interesting...when I enter the dual into Wrestlestat it gives me 30-9. Minnesota victories predicted at 141, 197 and 285. That said, I personally still don't use Wrestlestats for team comparisons or wrestler comparisons. Not enough matches for too many guys to make the mathematical Elo model effective.
Slight difference between when Steen ran the dual in Wrestlestat and and just now when you did. I see that someone changed the starters for Minnesota, Zain now wrestling Brolsma instead of Stroker; Wrestlestat predicted a MD over Stroker but a fall over Brolsma, accounting for the two points. No idea whether Brolsma is actually going over Stroker.
Stroker needs a lot more weight room work to even think of dealing with Zain. He's slick, but he'll get thrown all over the mat
An algorithm picked a MD for Zain over Stroker. Algorithms get stronger as they feed on more data and tend to be conservative to start. Zain is at the far end of the curve so I think most predictive models would tend to pull him back to the mean.
My algorithm forecasts a lot of pain for Stroker. My predictive model shows mean Zain pulling Stroker to his back.
True...but the model, unless I'm mistaken, includes a wrestlers entire body of work...equally. In other words, Zain's results from 2013-14 are weighted the same as last year and this year's results. With 21 Decisions, and only 4 Falls his first year, the model will not effectively "calculate" his current dominance. He's a dramatically different wrestler today...not necessarily in wins or losses, but in how much he dominates.
I agree, Zain will either Tech or Pin Stroker if they wrestle.
Yeah, I think @andegre already made tweaks to account for the fact that an earlier version this year made predicting pins almost impossible, and Zain has been making them look routine.
Remember that, Jason and Bo too!! My question to andegre (if you're out there) is whether recent results weigh heavier than older results?
I'll be there Duster with my Gopher wife. Will cheer for you.
