|
|
Post by uconncat on Mar 17, 2015 15:15:05 GMT -5
Gives predictions for every team in every game, including chances of winning it all. Uconn given a 74% chance of winning the national championship, SC with a 10% chance, ND 9% and Md 2%. 538 (Nate Silver) started out as a political prognosticator and has since taken its methodology to sports, etc. fivethirtyeight.com/interactives/march-madness-predictions-2015/#womens |
|
|
|
Post by Icebear on Mar 17, 2015 15:19:25 GMT -5
Actually, he originally did sports, baseball, before 538 went to politics and now back doing sports.
|
|
|
|
Post by linkster on Mar 17, 2015 17:17:45 GMT -5
Gives predictions for every team in every game, including chances of winning it all. Uconn given a 74% chance of winning the national championship, SC with a 10% chance, ND 9% and Md 2%. 538 (Nate Silver) started out as a political prognosticator and has since taken its methodology to sports, etc. fivethirtyeight.com/interactives/march-madness-predictions-2015/#womensUconn given a 74% chance of winning the national championship, SC with a 10% chance, ND 9% and Md 2%. And Baylor given 2%. Every other team has less than a 1% chance. Kind of says a lot about the steepness of the dropoff from the elite teams to the second level. And the silliness of suggesting that there are any " power conferences" in wcbb. |
|
|
|
Post by radylady on Mar 17, 2015 18:29:31 GMT -5
that's pretty cool
|
|
|
|
Post by semper on Mar 17, 2015 19:25:17 GMT -5
I'll consult it when I set up my bracket tomorrow. Thanks!
|
|
zorro
Husky Puppy
Posts: 225 
|
Post by zorro on Mar 17, 2015 23:23:05 GMT -5
UT 2% also.
|
|
|
|
Post by meyers7 on Mar 18, 2015 7:51:50 GMT -5
Gives predictions for every team in every game, including chances of winning it all. Uconn given a 74% chance of winning the national championship, SC with a 10% chance, ND 9% and Md 2%. 538 (Nate Silver) started out as a political prognosticator and has since taken its methodology to sports, etc. fivethirtyeight.com/interactives/march-madness-predictions-2015/#womensReally? Only 74%? Tony will have something to say about this slight. |
|
|
|
Post by chicagogg on Mar 18, 2015 9:29:23 GMT -5
I had never seen that site before. It was very interesting. I would love to know more about how they arrive at their conclusions, but I am probably not mathematically sophisticated enough to really get it.
|
|
|
|
Post by Pinsky on Mar 18, 2015 10:59:37 GMT -5
Like Ice said: Nate Silver did sports. Then he became super famous predicting elections. With mind blowing accuracy. Now, his company re entered the sports world big time. I think that's right. No?
|
|
|
|
Post by msf22b on Mar 18, 2015 11:44:59 GMT -5
Gives predictions for every team in every game, including chances of winning it all. Uconn given a 74% chance of winning the national championship, SC with a 10% chance, ND 9% and Md 2%. 538 (Nate Silver) started out as a political prognosticator and has since taken its methodology to sports, etc. fivethirtyeight.com/interactives/march-madness-predictions-2015/#womensHis political record seems exemplary |
|
|
|
Post by phil on Mar 18, 2015 13:31:03 GMT -5
I wouldn't have picked Texas that high. |
|
|
|
Post by phil on Mar 18, 2015 13:41:26 GMT -5
I know that Silver started in sports, but I didn't know his work then. I became acquainted with him when he started following politics (hence the 538, the number of national elections). I wanted to dislike him, because I didn't like his prediction, but it turned out to be quite accurate. Some of his prediction have been exemplary - IIRC he got all house races right one year, although even he conceded there is some luck involved. After seeing his results, I paid more attention, and bought his book The Signal and the Noise: Why So Many Predictions Fail-but Some Don't which is quite good (although some shortcomings when he ventures into new areas). I don't know the exact model he uses for wbb but I know he favors looking at existing models and including those results as weights, in addition to possibly a proprietary model. |
|
|
|
Post by Icebear on Mar 18, 2015 14:10:24 GMT -5
Readily acknowledges that there are unknown uncertainties and variables when moving into a new field that take some time to sort out. One of the first major breakthroughs he had in politics was the recognition of old phone polling techniques were skewed and outdated because of the emergence of cell phones, especially, among the youngest voters who frequently had no landlines. He quickly sought adjustments to 538 polling process and was much more accurate. There is another excellent young political pollster out of Princeton whose name escapes me at the moment.
Silver cut his teeth on baseball metrics.
|
|
|
|
Post by Pinsky on Mar 18, 2015 16:23:26 GMT -5
Sam Wang is out of Princeton. I think he's a neuroscientist as well. In the 2014 elections They had a little public disagreement and Nate won. They are both as good as you get.
|
|
