One mostly for the Squiggle modellers and Monash Prob Comp participants
You think the probability of a home win is h, and of a draw is d. What value should you enter for Prob(Home win) in the competitions?
@uvmcda@maxlaughton Certainly an advantage, but nothing like 50/50 becomes 80/20. Maybe 65/35 for some teams, and that's assuming about a two-and-a-half goal HGA.
Who lives in a pineapple under the sea? SpongeBob SquarePants 🎶
Someone flying a Diamond DA42 Twin Star (N108LU) over Bellefontaine, Ohio, had a little fun — drawing a giant SpongeBob SquarePants on the map. Via @AirNavRadar
Check the flight data: https://t.co/LFbFhv1dhg
📢📢 I am excited to announce a new open source code repository on the application of AI and Data Science in the Cricket world. Over the course of next few days, I will be open sourcing the models that power some of the advanced analytics on Cricmetric.
First installment: a full look into the T10 Win Probability model that is currently used on the website. Detailed article: https://t.co/4vAi2EWFOs
The open source Git repository can be found here: https://t.co/OEACLpJN7z . Collaborators invited!
@Chudlubber@AflGlicko If you're genuinely interested and not just wanting to have a crack for its own sake, there are about a dozen different models whose team ratings and rankings are summarised here (and probably not a spreadsheet amongst them)
https://t.co/2UQysbbexy
@EmlynBreese Are there some really small counts here?
My (possibly flawed) understanding is that Disposals and Possessions should generally be very similar (recognising that tackled means poss no disp, and kicked off ground means disp and no poss)
@DanielCherny You're beautiful
You're beautiful
You're beautiful, it's true
I saw your mace in a crowded place
And I don't know what to do
'Cause I'll never be with you