In the math example, I stated what would happen if Kovach got all of the Republican votes and Migliore got all of the Democratic votes, Kovach would need to win 3-1 among Others.
I do not expect either candidate to win all of the votes of their own party and should have used the term numbers "equivalent to all" because after each of them get whatever numbers they get from their other party I think it will be the numerical equivalent of each getting roughly all of his own party vote and none of the other party vote which still leaves Kovach with a 200-300 vote deficit before Others are counted in with the turnout estimates I used.
If turnout is mugh higher or much lower, the number would require adjustment, but the principle would remain the same.
The numbers I used put estimated total turnout at 20%. I assume it will actually be closer to the 18% that voted in the Nov 2007 & Dec 2007 special elections for vacated legislative seats in late 2007.