Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : PRESIDENT GEORGE W. BUSH -- Ignore unavailable to you. Want to Upgrade?

To: Neil H who wrote (99149)	12/3/2000 1:23:44 AM
From: fuzzymath	Read Replies (3) \| Respond to of 769667

It would appear that Yale Professor Nicholas Hengartner, statistician witness for Al Gore in the case before Judge Sauls in Leon County, has proven that a full recount of undervotes in all Florida counties using punch card ballots would likely substantially increase George W. Bush’s margin of victory. While this was not explicitly stated by Prof. Hengartner (who was tasked by the Gore team with predicting “the probability that the results of the election might be overturned if a full recount would proceed in Miami, Broward, and Palm Beach” counties), the numbers he presented suggest that this indeed would be the case if a statewide undervote recount occurred. Prof. Hengartner uses Broward County as an example for comparing the voting ratios for recorded votes and votes added during the manual recount. His description is as follows: HENGARTNER: “So let me walk you through this graph. In Broward, after the mandated machine count, there was 31 percent (inaudible) major share. So we seek to look at the Bush-Gore vote. Thirty-one percent of the votes went for Bush, where 69 shares went for Gore. Now, I do the recount, and there was a determination made that some of the ballots could be assigned Intend to vote, and these ballots which could be identified intend to vote, once I looked at for whom the vote fell, 34 percent of these ballots that were determined in the manual recount were 34 percent, whereas there was 66 percent for Gore, which is pretty much in line with those numbers. If anything, if it would be significantly different, it would be a difference in favor of Mr. Bush.” Statistically, these numbers suggest that there is a stronger correlation between the intent to vote for Bush and an undervote than between the intent to vote for Gore and an undervote. In other words, there was a greater tendency for voters who intended to vote for Bush to undervote than there was for voters who intended to vote for Gore. In Broward County, about 25% of undervotes were awarded to a candidate during the hand recount (Boies). This means the identifiable undervote was about 0.3% of the total vote. Out of 100,000 votes counted by the machines, 69000 were for Gore, 31000 for Bush. For each 100,000 votes counted by the machine, about 300 undervotes were identified for a candidate. Of these, 198 were identified for Gore, 102 for Bush. Thus, for every 100,000 Bush machine counted votes, there were 329 more votes identified by the hand count of the undervotes; for every 100,000 Gore machine counted votes, there were 287 more votes identified by the hand count of the undervotes. This means Bush voters were 15% more likely to undervote than Gore voters. Or, stated another way, given a 50%-50% mix of Bush and Gore machine counted votes, we would expect Bush to hold a 53.4% to 46.6% margin in votes that would be added by hand recounts of undervotes. The conclusion, from the numbers presented by Prof. Hengartner, is that a statewide count of all undervotes for counties that used Broward-type punch card ballots would result in a net gain for George W. Bush, assuming the undervotes were distributed evenly statewide. While I do not have the statewide data on this, a look at the locations of the counties that use punch card ballots suggests a fairly even distribution. How many votes might Bush gain from a statewide count of undervotes? A rough estimate can be made. 25 out of 67 counties use punch card ballots. There were 185,000 undervotes, with a majority of those registered in counties using punch card ballots. If 25% of an assumed 120,000 punch card ballot undervotes were identified for a candidate, and Bush voters were indeed 15% more likely to undervote statewide (as they did in Broward County), then Bush would gain 2040 votes in a statewide count.