Perot doesn't drop out

[Erc]

Using the results of the tracking poll in May that was most favorable to Perot...

(Perot 39-Bush 31-Clinton 25)



Perot 437
Bush 92
Clinton 9


(Bush might win one of the NE CD's, but since I don't have good data there, I can't really say).

Using the last one before he dropped out...
(Bush 33-Perot 32-Clinton 27)



Bush 234
Perot 225
Clinton 79

Note that the number of razor-thin margins in this scenario is ridiculous...

Illinois is a three-way race, Clinton 34 - Bush 33 - Perot 32...
Bush wins Hawaii by 2000 votes...
Bush wins Iowa by 15000 votes...
Perot wins Michigan by 17000 votes...
Perot wins Ohio by 6000 votes...
Bush wins Pennsylvania by 22000 votes...
Bush wins South Dakota by 4500 votes...
Bush wins Texas by 64000 votes...
Perot wins Utah by 5500 votes...
Clinton wins West Virginia by 6500 votes...

A .8 swing to Bush from Perot (within the rounding, let alone the statistical, error):



Bush wins 278-181-78

[Erc]

Thankfully, not possible under the current system (269/2 is not an integer).

Possible but worrying is:

What if two candidates tie for third and fourth?  Or if there's a three-way tie for second?