By Brad Friedman on 11/26/2004, 5:38pm PT  

Warren Mitofsky, head of the consortium which gathers and publishes the exit poll data for all of the major networks and newspapers, has been claiming of late that those who relied on and/or published his exit polling data during the course of Election Day were mistaken to have done so. The reason being, Mitofsky recently emailed Keith Olbermann at MSNBC, is because those raw numbers, leaking out when they did, are more aptly compared to "the score at half time at a football game" rather than information to be used to judge against the final election results.

I believe Joseph Cannon gets it exactly right, however, when he responded yesterday with this thought:

No, Mr. Mitofsky. They should be compared to the score at half time at fifty football games. Better analogy: Fifty tosses of a coin. Error should skew in both directions; if the coin keeps coming up heads, something is wrong with that coin.

Of course, without the raw data from the exit pollsters, everyone is largely guessing. So I will associate myself with pollster John Zogby's sage demand on MSNBC last Wednesday wherein he said, "I think that the gentlemen who are responsible for the exit polls should be fully transparent, release their data, discuss their methodology. Let us see what exactly it is that happened, and why it happened...I'll take this opportunity right now to say I think that it's in the interest of healing this country and restoring some unity to this country for us to have a thorough investigation of what happened both to the election and with the exit polls."

Hear, hear, Mr. Zogby!...

Tis a shame though that rest of the media in this country were dilligently covering the "unacceptable" disparity between exit polling and final results over in the Ukraine, as the Bush Administration has spoon-fed them to do, instead of looking into those same disparities here. So unfortunately, Zogby's call likely falls on deaf ears. Now if only he were a Republican on the other hand...

Folks just now wrapping their brain around all of this would be well advised to peruse the published work --- neatly summarized in this Detroit Metro Times article --- of Steven Freeman, a professor of statistical analysis at the University of Pennsylvania with a doctorate from MIT. Freeman's analysis of the exit poll data versus the final results in just Ohio, Florida and Pennsylvania peg the probability that the exit polls could all swing so wildly for Bush so late in the day as they did in all three states as a 250 million to 1 longshot.

Freeman's study was double-checked for accuracy, and concurred with a few days later, by colleague John Allen Paulos, a professor of mathematics at Temple University in Philadelphia. Paulos further goes on to make a crystal-clear point to skeptics who may feel that all of the "irregularities" combined so far would still not be enough to have any meaningful effect on the election itself. He smartly notes that if "only about a half dozen voters per precinct in [Ohio] - switched their votes, Kerry would be president-elect. Considerably fewer switches would be required if, as is likely, most provisional and spoiled ballots were good and went for Kerry."

But never mind such questions about our own democracy...Back to that "outrage" in the Ukraine!