How Big is the P.U.M.A. Factor?

An analysis of how many PUMA voters are required for McCain-Palin to win.

If you know the answer to this question, you know who our next President will be. While it is possible for McCain to win even if this factor is negligible, he would need to get virtually all of the undecided’s to break his way. That is theoretically possible but it is mostly wishful thinking on the part of McCain-Palin supporters. For every undecided voter either candidate picks up, it takes a vote out of the undecided column and adds one to that candidate's total. However, the spread between the two candidates only goes up by one vote. That is not the case with a P.U.M.A. voter, which is by definition, a registered Democrat voting for McCain-Palin. This takes a vote away from Obama-Biden and adds one to McCain-Palin, increasing the spread by two votes.



There is always a certain amount of voters who do not vote for their own party's candidate in any election. This desertion rate is pretty well known and is factored into the weighting algorithms used by pollsters. What is unusual about this election is that the way Hillary was treated by the Democratic National Committee and by the Obama campaign thoroughly annoyed a lot of Democrats. Add in the fact that Obama's positions make social conservative Democrats very nervous. Then there is the hypothetical Bradley Effect due to the presence of a black candidate. All these factors make up the P.U.M.A. coalition which is very hard to numerically pin down. That makes it difficult for the pollsters to properly allow for this factor. Given the spread between the various polls, it is obvious that some of the polls are making no allowance for the P.U.M.A. Factor at all.



As we have all learned by now, it does not matter what happens in the popular vote because only the electoral votes count. So it all comes down to the effect that the PUMA Factor will have in the battleground states. By knowing the size of the PUMA Factor in those eleven states, we can determine who will win and by how many electoral votes. By looking at the No Toss Up States electoral map on www.realclearpolitics.com, we see that at the moment, Obama-Biden has 353 electoral votes and McCain-Palin has 185 votes. Now look at the RCP average for the battleground states further down that page. Let's arbitrarily decide that there is no way the PUMA Factor can make up a deficit of 13 or more points in the RCP average for a state.



Now let's build a table of the eleven remaining states in ascending order of the amount that McCain-Palin is behind in the RCP average for each state. To this we add columns to show what happens as the PUMA Factor allows McCain-Palin to overcome the deficit in each state. As you can see, if the PUMA Factor is large enough to overcome a deficit of 6.2 points, we have a tie vote. Overcoming a 6.5 point deficit gives McCain a victory by 10 electoral votes. Anything higher is pure gravy. Using the latest state population estimates from the Census Bureau, we can apportion the estimated turnout of 130,000,000 voters nationwide into these eleven states to get an estimated turnout. The big question is how many PUMA voters would it take to give McCain-Palin the election? Since each PUMA voter has twice the effect, it takes only half of the RCP average of PUMA voters to swing the state into the McCain column. This allows us to calculate the number of PUMA's needed in a state.

STATE	VOTES	RCP AVERAGE	7/07 POPULATION	EST. TURNOUT	PUMA’S NEEDED	O-B VOTES	M-P VOTES
NC	15	1.3	9,061,032	3,905,343	25,385	338	200
FL	27	4.2	18,251,243	7,866,362	165,194	311	227
OH	20	5.6	11,466,917	4,942,289	138,384	291	247
VA	13	6.0	7,712,091	3,323,944	99,718	278	260
CO	9	6.2	4,861,515	2,095,333	64,955	269	269
NV	5	6.5	2,565,382	1,105,690	35,935	264	274
NM	5	7.3	1,969,915	849,042	30,990	259	279
PA	21	8.5	12,432,792	5,358,585	227,740	238	300
WI	10	11.0	5,601,640	2,414,330	132,788	228	310
NH	4	11.1	1,315,828	567,127	31,476	224	314
IA	7	11.6	2,988,046	1,287,860	74,696	217	321

Now you can see why the McCain-Palin campaign says that this election is not over. There is nothing like a numerical analysis to allow you to see what it would take to turn this election around. Now before you McCain-Palin supporters think this is a piece of cake, let's have a reality check. You cannot count a voter as a PUMA unless the Democrat voted for both Gore and Kerry. Democrats who did not do that are already accounted for in the polling models. This analysis assumes a uniform turnout across the country which is not going to happen. That means that it could take more PUMA voters if the turnout is higher than the average in any of these states. Likewise, if the polls have underestimated the Democratic turnout, then more PUMA voters will be required to offset that. Another assumption is that the number of Republicans voting for Obama-Biden does not exceed the usual percentage of Republicans doing that.



Will there be enough PUMA voters in the top six states in the table to give McCain-Palin the election? I do not know and neither does anyone else. I will say that if McCain-Palin only carries the top five states, we are going to have a complete mess on our hands because a tie throws the election into Congress to vote on a state by state basis. There is one possibility that the table does not cover and that is Congressional District 2 in Maine. If McCain-Palin could pick that one vote up, they can win without Nevada. Of course, that would take sending Sarah Palin to Bangor, Maine as I have repeatedly suggested. Instead, the McCain campaign is pinning its hopes on Pennsylvania. The problem is that the PUMA voters are not evenly distributed by state. There would seem to be a much smaller percentage of disgruntled Hillary supporters in the western states. That is why the McCain folks have leapfrogged over the three western states in favor of Pennsylvania.