I like this approach. But I think you've multiplied by a factor which should be dividing:<br><br><div class="gmail_quote">2011/6/23 <span dir="ltr"><<a href="mailto:fsimmons@pcc.edu">fsimmons@pcc.edu</a>></span></div>
<div class="gmail_quote"><span dir="ltr"></span>...<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
Here's a way to resolve it.<br>
<br>
Let N=W+L. Let p=W/N. Let q=L/N. Let sigma = sqrt(N*p*q).<br>
<br>
Measure defeat strength by S=(p-q)*sigma.<br></blockquote><div><br></div><div>Let's say one team happens to consist of penguins. They score 0. The chances of them winning a rematch are 0. Their defeat strength should be... infinite. But with your formula, it comes out to 0.</div>
<div><br></div><div> ....</div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
What is the heuristic behind S=(p-q)*sigma?<br>
<br>
The Binomial distribution has standard deviation sigma = sqrt(N*p*q).<br>
<br>
>From the point of view of the winning team the fraction p is the proportion of favorable outcomes, while<br>
the fraction q is the proportion of unfavorable outcomes. So (p-q) is the difference in estimates of the<br>
underlying fractions of favorable and unfavorable outcomes. We multiply this estimate by the standard<br>
deviation to take into account "sample size."<br></blockquote><div><br></div><div>You should be dividing by the standard deviation. Higher standard deviation, more uncertainty about the repeated result.</div><div>
<br></div><div>And even that would only give you number of standard deviations. To convert this to a probability, you'd need to use a bell curve distribution.</div><div><br></div><div>Of course, if the game were soccer(UK football) instead of basketball, where a score of 0 is common and nothing to be particularly ashamed of, this wouldn't work. I think you'd have to add some epsilon to both scores to avoid the infinities.</div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<br>
If a real statistician, like Jobst were reading this, he (or she) could refine this estimate, or at least give a<br>
better explanation.<br></blockquote><div><br></div><div>Disclaimer: I am not a real statistician by academic credentials, though actually I've been gainfully employed as one in Guatemala.</div><div><br></div><div>JQ</div>
</div>