Reply
 
Thread Tools Display Modes
  #1  
Old 11-02-2016, 12:20 PM
Saint Cad Saint Cad is offline
Guest
 
Join Date: Jul 2005
Location: N of Denver & S of Sanity
Posts: 12,252
Question on probability involving handicapping.

Let's say I'm playing a game and there is a certain probability of me winning p(win). But to complicate things there are multiple factors affecting my chance to win which (luckily) can be quantified. For example
Standing on one foot => -20% chance to win
Being 5 feet closer => +15% chance to win

How do I use those to calculate my chance to win? The method I use is to treat them as multipliers. new p(win) = (1.15)(0.80) x p(win). My logic is that suppose there is a factor making it impossible to win then the multiplier would be zero and it doesn't matter what the other numbers are since new p(win) = 0.

Is this the right approach & logic?
  #2  
Old 11-02-2016, 12:40 PM
Thudlow Boink Thudlow Boink is offline
Charter Member
 
Join Date: May 2000
Location: Lincoln, IL
Posts: 24,097
Quote:
Originally Posted by Saint Cad View Post
How do I use those to calculate my chance to win?
Essentially, you're asking "What do I mean by '+15% chance to win'?" (Or, what does someone who says such a thing mean by it?) This is something that the person saying such things ought to be clear about, but IMHO yours is the most natural interpretation.
  #3  
Old 11-02-2016, 04:20 PM
muttrox muttrox is offline
Member
 
Join Date: Dec 2000
Posts: 2,103
Assuming these are all independent factors, your approach is reasonable. If going 5 feet closer is somehow correlated with standing on one foot, it won't work.
  #4  
Old 11-03-2016, 12:50 PM
Buck Godot Buck Godot is online now
Guest
 
Join Date: Mar 2010
Location: MD outside DC
Posts: 4,354
Quote:
Originally Posted by Thudlow Boink View Post
Essentially, you're asking "What do I mean by '+15% chance to win'?" (Or, what does someone who says such a thing mean by it?) This is something that the person saying such things ought to be clear about, but IMHO yours is the most natural interpretation.
The first part of your statement in correct that this begs the question as to what you mean by a +15% chance to win. But while the OP's method is pretty good, it suffers from some draw backs.

First suppose I initially had a 50% chance to win, then had one factor that gave a +20% chance and a second that gave a -20% chance. Ideally I should get back to a 50% chance but instead I have a 0.5*0.8*1.2=48% chance.

Also suppose I have a 90% chance to win and then step 5 feet closer to the basket. Using the4 OP's method I end up with a (0.9*1.15)=103.5% chance to win. This can't happen even if I'm willing to give %110.

So what statisticans do instead is use the logistic function.

P(win)=e^X/(1+e^X)

Where X is the sum of your positive and negative factors. So stepping closer to the basket my add 0.3 to X while standing on one foot may subtract 0.4 from X. When X=0 you have a 50% chance of winning. As X gets very large, the probability of winning goes to 1, and as X gets very negative the probability goes to zero. Also factors are additive so a postive factor if I have a postive factor of +0.3 and a negative factor of -0.3 they will cancel out.

Also notice that P(lose)=1-P(win)=e^-X/(1+e^-X) , so positive factors that increase your chance to win, decrease your opponents chance to win by the exact same amount.


Unfortunately it no longer becomes quite so simple to describe what an increase or decrease of a particular amount means. Starting from a 50% chance to win, adding 0.3 to X, will decrease my chances of winning to 42.6% (so according to the OPs method about 15.8%) , but if my chances of winning was at 42.6%, adding another -0.3 will change my chances to 0.3543 or (according to the OPs method by 18.7%).

Last edited by Buck Godot; 11-03-2016 at 12:54 PM.
Reply

Bookmarks

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off

Forum Jump


All times are GMT -5. The time now is 11:23 PM.

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2017, vBulletin Solutions, Inc.

Send questions for Cecil Adams to: cecil@chicagoreader.com

Send comments about this website to: webmaster@straightdope.com

Terms of Use / Privacy Policy

Advertise on the Straight Dope!
(Your direct line to thousands of the smartest, hippest people on the planet, plus a few total dipsticks.)

Publishers - interested in subscribing to the Straight Dope?
Write to: sdsubscriptions@chicagoreader.com.

Copyright 2017 Sun-Times Media, LLC.

 
Copyright © 2017