[EM] Approval _is_ like sincere Borda.

[MIKE OSSIPOFF]

Sure, on the average, Approval is like sincere Borda. Surely often they'd 
both give the same result. Sorry if I misunderstood and replied as if  you'd 
meant  they always do. One is more likely to vote for one's 2nd choice than 
one's 3rd choice, and so, overall, the effect could be like Borda.

Approval is different from IRV and Plurality, and is like Borda, in that 
Approval and Borda both look at a candidate's overall likedness. They both 
reflect and express ratings. In Approval whether one votes for a candidate 
depends on ratings, while in CR the ratings are explicit, and in Borda 
they're assumed to vary linearly down the ranking.

So sure, the approximation that you (Bart) used would, overall, on the 
average, be like Borda. A 1/3 probability someone will vote for his 3rd 
choice, a 2/3 probability someone will vote for his 2nd choice, a 3/3 
probability he'll vote for his 1st choice, and a 0/3 probability he'll vote 
for his 4th choice.  So the expectation for what he'll give a candidate in 
that Approval approximation is the same as what he'd give him in Borda. That 
dramatizes the ratings-influenced similarity between Approval & Borda.

Mike Ossipoff

_________________________________________________________________
Take advantage of our limited-time introductory offer for dial-up Internet 
access. http://join.msn.com/?page=dept/dialup