THE BOOK cover
The Unwritten Book is Finally Written!
An in-depth analysis of: The sacrifice bunt, batter/pitcher matchups, the intentional base on balls, optimizing a batting lineup, hot and cold streaks, clutch performance, platooning strategies, and much more.
Read Excerpts & Customer Reviews

Buy The Book from Amazon


SABR101 required reading if you enter this site. Check out the Sabermetric Wiki. And interesting baseball books.
MOST RECENT ARTICLES
MAIL : You ask | We say

Advanced


THE BOOK--Playing The Percentages In Baseball

<< Back to main

Wednesday, January 26, 2011

Win Probability: pre-allocation method

By Tangotiger, 04:21 PM

Brian asks:

Don’t miss Andrew Foland’s 5-part tour de force on how to incorporate pre-game estimates of win probability with the in-game estimates. Awesome job, and I really appreciate the time Andrew has put in. This is something I’ve been asked to do for a while. I think it’s a great idea, and I developed a very similar method to do it.

However, I would be reluctant to make it the ‘official’ WP or WPA model. If the Patriots have a game against a weaker opponent pegged as a 70/30 match-up, and Tom Brady plays lights-out, should he only get 0.30 WPA instead of 0.50 WPA? Should he be penalized for being favored pre-game?

I’ve talked about this in the past, and I guess Brian post-dates those discussions.  No problem, as I presume lots of readers don’t go through my archives looking for stuff like that.

Ok, this is how it works.  Suppose that Halladay is pitching, and the Phillies are at home.  Let’s say that they have a .750 chance of winning.  In order to go that, you need to “preallocate” a certain number of wins to Halladay, to the Phillies being at home, to Rollins, and Utley et al.  Then, you let them accumulat their +.250 wins or -.750 wins depending on whether they win or lose.

And what happens after one million games?  Well, Halladay’s in-game WPA will be exactly ZERO.  The sum of his pre-allocated WPA will be exactly equal to his talent level.  In short, given enough games, adding up a player’s in-game WPA is useless.

But:
1. That is deathly boring
2. We really don’t know what his true talent level is for each game
3. There aren’t enough games to get a zero for in-game WPA, and now the whole thing will look strange

So, rather than doing the extremely complicated method of preallocating wins on a player by player basis, simply start everyone with ZERO pre-allocated wins, and let them earn their in-game WPA without knowledge of their talent level or their team’s talent level.  When you sum it up, you will obviously get zero pre-allocated wins, and the in-game WPA will be their entire impact.

If you are a professional bettor, well, you desperately need pre-allocated wins.

Check out the archives.  I have a few threads on the topoc.


#1          (see all posts) 2011/01/26 (Wed) @ 22:54

Am I correct in understanding that, by charting pre-allocated wins during play, you create a graph which basically reflects live, updated gambling odds? Sort of an up-to-the-play Vegas line?

That doesn’t seem too terribly boring, but I’m probably missing something.


#2    Xeifrank      (see all posts) 2011/01/27 (Thu) @ 01:42

From my understanding it means that you don’t start the Win Expectancy clock at 50.0/50.0, you first look at the closing Vegas line, which could be something like 75.0 / 25.0 (75% win prob for the favorite and 25% win probability for the underdog).  Then you dole out some of the WPA to the players that give the favorite the edge, before the game even starts.

After a while you can say that Roy Halladay is something like 0.28 wins “Against The Spread”, or -.18 wins ATS.


#3    Tangotiger      (see all posts) 2011/01/27 (Thu) @ 12:17

Right, the preallocation of wins will only make bettors happy.

It would be boring for the rest of us.


#4    Sky      (see all posts) 2011/01/27 (Thu) @ 12:28

When we pre-allocate wins, doesn’t that also imply that we’re customizing the in-game probabilities?  For example in football we give Tom Brady a better chance of scoring on a two-minute drill, or in baseball we realize that any closer, let alone Mariano, is going to allow runs much less frequently than the average pitcher (including starters)?  If so, that’s a solid, interesting improvement.


#5    Tangotiger      (see all posts) 2011/01/27 (Thu) @ 13:02

I guess I should be clearer: if you intend to have valid real time win expectancies (the chance of winning at any point in context with the game, the players, the bench, and so on), then you absolutely want to preallocate your wins.  You want to know if Pujols is batting or not, so you know what the change in win expectancy is going to be if you IBB him or not.  All that is great, if that is your objective.  It is a monumental effort to say the least.

Given enough games (like a million in one year), the in-game WPA of every single player will be exactly zero.  There will be no more point to track in-game changes and attributing them to players, because if you did your preallocation perfectly, then you will get no new information from in-game WPA.

If you did a half-a$$ed job in preallocation, the in-game WPA will capture the difference.  (Presuming 1MM games.)

If you decide to not do preallocation at all, the in-game WPA will capture all the difference (pretty much, but not totally exactly). (Presuming 1MM games.)

I’ll take back that it would be deathly boring.  It’s deathly boring depending on your objectives.


#6    Tangotiger      (see all posts) 2011/01/27 (Thu) @ 13:15

"It’s deathly boring depending on your objectives. “

I can actually say that about anything, so it’s useless for me to say it.  Silly that I brought it up actually…


Page 1 of 1 pages


Name (required)
E-Mail (optional; WILL be published)
Website (optional)

<< Back to main


Latest...

COMMENTS

May 25 15:37
What sabermetrics is NOT

May 25 15:28
Largest demonstration in Canadian history?

May 25 15:12
Do pitcher’s reach back for velocity when needed?

May 25 15:02
Pete Palmer’s new book: Basic Ball

May 25 13:04
“Why Kickstarter works”

May 25 12:51
Chad Curtis

May 25 11:32
Howard Stern

May 25 11:26
Lack of hustle during a game

May 25 10:58
Rooting for laundry

May 25 02:38
NFLPA lawsuit against collusion