[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
THE BOOK cover
The Unwritten Book
is Finally Written!

Read Excerpts & Reviews
E-Book available
as Amazon Kindle or
at iTunes for $9.99.

Hardcopy available at Amazon
SABR101 required reading if you enter this site. Check out the Sabermetric Wiki. And interesting baseball books.
Shop Amazon & Support This Blog
RECENT FORUM TOPICS
Jul 12 15:22 Marcels
Apr 16 14:31 Pitch Count Estimators
Mar 12 16:30 Appendix to THE BOOK - THE GORY DETAILS
Jan 29 09:41 NFL Overtime Idea
Jan 22 14:48 Weighting Years for NFL Player Projections
Jan 21 09:18 positional runs in pythagenpat
Oct 20 15:57 DRS: FG vs. BB-Ref

Advanced

Tangotiger Blog

A blog about baseball, hockey, life, and whatever else there is.

Saturday, December 21, 2013

Predicting huge comebacks

?More interesting data as to the idea that models can't properly estimate comebacks in extreme conditions.  In the 109 games where the team was favored to win 51% to 90% of the games (average expected win rate of .723), they actually won .716.  That difference is less than 0.2 SD, which, well, that's called nailing it.

In the other 111 games where the odds of winning was expected to be greater than 90% (average expectation of .982), they actually only won .919 times.  That's either 2.5 or greater than 4 SD depending on how you calculate it.  In any case, it's big enough to make us reconsider the extreme game, that maybe our win probability charts don't work.  And they don't work because the whole strategy has changed.

The easiest way to see it is by looking at the NHL, where teams can pull their goalies.  While their goals allowed rate shoots way way up, their goal scoring rate also shoots up (but not anyway near as much).  So, why do the teams do it?  Because whether they lose by 1 or lose by 2, it doesn't matter.  So, it's a high-risk, high-reward play, and so, when you try to figure out the chances of winning, you have to modify your charts to include this change in scoring environment.

It's possible a similar kind of idea applies in NFL as well.  If you are losing bigtime, well, the trailing team will be taking far more risks.  And those plays will eventually pay off, just a bit more, than if they simply played their normal way.

We see it also in MLB in the bottom of the 9th, where my win probability charts do not work.  I don't know anything about NBA, but I presume a same scenario applies there as well.

So, I'd like to see the work in the article applied to the other sports.  But rather than 4th quarter, I'd like to see it much more closer to the end of game.  In hockey, it's easy: look at the last 2 minutes, or the last five minutes.  In baseball, the bottom of the 9th, or if you want, the top of the 8th.  In football, I'd say the last 3 minutes, maybe the last 5 minutes.  NBA?  I'm guessing the last 3 minutes.

Latest...

COMMENTS

Nov 23 14:15
Layered wOBAcon

Nov 22 22:15
Cy Young Predictor 2024

Oct 28 17:25
Layered Hit Probability breakdown

Oct 15 13:42
Binomial fun: Best-of-3-all-home is equivalent to traditional Best-of-X where X is

Oct 14 14:31
NaiveWAR and VictoryShares

Oct 02 21:23
Component Run Values: TTO and BIP

Oct 02 11:06
FRV v DRS

Sep 28 22:34
Runs Above Average

Sep 16 16:46
Skenes v Webb: Illustrating Replacement Level in WAR

Sep 16 16:43
Sacrifice Steal Attempt

Sep 09 14:47
Can Wheeler win the Cy Young in 2024?

Sep 08 13:39
Small choices, big implications, in WAR

Sep 07 09:00
Why does Baseball Reference love Erick Fedde?

Sep 03 19:42
Re-Leveraging Aaron Judge

Aug 24 14:10
Science of baseball in 1957

THREADS

December 21, 2013
Predicting huge comebacks