Politech mailing list archives
FC: Bush set to win today? And election coverage links
From: Declan McCullagh <declan () well com>
Date: Tue, 07 Nov 2000 15:04:49 -0500
Some possibly-useful links... Vote-watching guide: http://www.evote.com/features/e2knightguide.asp Washington Post's guide: http://washingtonpost.com/wp-srv/onpolitics/elections/election2000coverage.htm Exit poll background: http://www.opinionjournal.com/extra/?id=65000557 What sites are doing: http://www.wired.com/news/business/0,1367,39964,00.htmlAlso Drudge has pledged to break the exit poll embargo on drudgereport.com, but his site is unreachable, so even if he did, it'll have little effect.
-Declan ********* http://www.cluebot.com/article.pl?sid=00/11/07/161240&mode=thread Poll: Bush Set to Win, and Election-Night Scorecard posted by cicero on Tuesday November 07, @10:47AM Who would've thunk it? George W. Bush's lead has increased sharply after his 1976 DUI arrest, and he's set to win in a possible landslide today, according to our final poll analysis. Our analyst writes below: "This may be because of the perceieved involvement of the Gore campaign, or because late-breaking Undecideds generally favor the challenger. In the final hours of the 2000 campaign, Bush has acheived his greatest lead since Gore's convention bounce began: 48.9% to 39.8%. It increased by 2.1% to 9.1% from 7.0%... It is possible that Bush will match Clinton's 1996 share of 49%, while Gore will fall short of Dole's 41%." We thank our anonymous author of these poll writeups, who must labor nameless because of unrelated, non-political consulting work he does for the Feds. [...] Here is my final analysis of the Rasmussen Research Electoral College numbers from Monday, November 6. I will follow this with a report of results of Monte Carlo simulations of the Electoral College results based on the probability distributions implied by the new data. I will also send you a scorecard to print out for use tonight. The new nightly tracking data from last night shows that Bush's lead has in fact increased sharply since the announcement of his 1976 DUI arrest. This may be because of the perceieved involvement of the Gore campaign, or because late-breaking Undecideds generally favor the challenger. In the final hours of the 2000 campaign, Bush has acheived his greatest lead since Gore's convention bounce began: 48.9% to 39.8%. It increased by 2.1% to 9.1% from 7.0%. The margin of error is about 1.6% by virtue of the three-night sample size of 3750. The trends of both Gore's and Bush's share were clear and month-long, but in the closing days Gore's increase from the undecided stopped or even reversed, and Bush's accelerated. Bush will likely get about 48.5% today, and Gore, about 41.0%. Given the trend in the last few days, the margin could be even larger. It is possible that Bush will match Clinton's 1996 share of 49%, whileGore will fall short of Dole's 41%. In Rasmussen's state-by-state data, what must be their last three surveys showed up this morning. Two -- New Jersey and Wisconsin -- were done Sunday night, while the third -- Iowa -- was done Thursday night. The margin of error in two of the three is unknown since they didn't report the sample size, though it was probably the same 4% or less that all of their others have been. Amazingly, there was no change in the margin in any of the three! In New Jersey, last surveyed just a week ago, there was no change in Gore's 6-point lead. Bush trails by 4 there, so New Jersey's 15 electoral votes continue to lean Gore. Wisconsin's 11 electoral votes continue to lean Bush. In it's first survey in two weeks, there was no change: Bush continues to lead by 4 points. In Iowa -- last surveyed between the final two debates -- Bush's lead was also unchanged at 4-points. There was thus no change in the 3.5%/7.5%, the 0%/0%, or the 4.5%/9.5% lean/solid scenarios. If Gore could capture all of the Nader vote -- some surveys indicate that he could only get half of it while Bush would get a quarter -- the data at hand would give him a narrow 273-265 victory. Interestingly, in all states but Washington and Wisconsin, Nader trails Undecided! Based on current national data Gore would trail Bush by 4.5% in the popular vote then. If Gore could recapture only half the Nader vote and Bush got no additional votes, Gore would need an additional 2.0% shift from Bush in key states to win in the Electoral College. If Nader voters stick to their avowed allegiance in the face of the loss for their traditional party -- and New Mexico's experience with Green party voters is that they will -- then Gore needs an enormous national shift of over 4.5% from Bush to capture the Presidency. Thus he needs to win not only the tossup states but 4 states that were leaning Bush when last surveyed: Maine, Iowa, Wisconsin, and Pennsylvania. Any change in the analysis below is solely due to correction of previous errors. The numbers still include 6 unsurveyed, but presumably solid electoral votes which break evenly Bush 3, Gore 3. These votes, omitted from Rasmussen's state-by-state surveys, include the presumed Bush state South Dakota(3) and the presumed Gore 'state' D.C.(3).} Under the reasonable lean/solid criteria of (>3.5%)/(>7.5%) Bush's lean + solid lead remains 276 to 114 with 148 in the tossup range of 3% or less. [These criteria mean that a state where one candidate has a 4% lead is leaning his way, and 3% is the margin of error in most of the state polls Rasmussen is doing now.] Of that lead, 44 votes (62-18) are in Senate-based electoral votes. Sol-G Lean-G Tossup Lean-B Sol-B ElecV 64 50 148 72 204 States 6 4 10 7 24 HouseEV 54 42 128 58 156 SenatEV 10 8 20 14 48 Total-G Tossup Total-B 114 148 276 Though a number of states have moved from leaning or solid Bush columns into the tossup column, and vice versa, Gore's states have been fairly stable. Today Gore manages a leaning or solid lead in only ten 'states': New York, Illinois, New Jersey, Massachusetts, Connecticut, Hawaii, Vermont, Rhode Island, Maryland, and the District of Columbia. There are only two tossup states with 57 electoral votes on his side of even -- California and Delaware. Bush now has a leaning or solid lead in 31 states. When the lean/solid criteria are set to 0%, Bush takes 81 of the tossup votes (7 states) and Gore takes 57 (2 states). Bush's lead then grows to 357 to 171 with only Minnesota's 10 in the tossup column. The Senate-based lead increases to 52 (76-24). Sol-G Lean-G Tossup Lean-B Sol-B ElecV 171 0 10 0 357 States 12 0 1 0 38 HouseEV 147 0 8 0 281 SenatEV 24 0 2 0 76 Total-G Tossup Total-B 171 10 357 With the lean/solid criteria set to the more conservative 4.5%/9.5%: Sol-G Lean-G Tossup Lean-B Sol-B ElecV 64 28 215 76 155 States 5 3 15 8 19 HouseEV 54 22 185 60 117 SenatEV 10 6 30 16 38 Total-G Tossup Total-B 92 215 231 Bush's lean+solid total is over twice Gore's. In the Closing Days. Change Since After Third Debate EVotes BushUp BushChng Wisconsin 11 4 0 New Jersey 15 -6 0 Washington 11 1 1 Pennsylvania 23 4 -1 Michigan 18 1 1 New York 33 14 3 Pennsylvania 23 5 2 West Virginia 5 3 1 Florida 25 2 1 Tennessee 11 1 -3 Missouri 11 5 1 Delaware 3 -1 10 Virginia 13 10 -2 Michigan 18 0 0 Pennsylvania 23 4 2 After Third Debate. Change Since Second EVotes BushUp BushChng Iowa 7 4 0 North Carolina 14 10 0 Maryland 10 -7 -1 Wisconsin 11 4 -2 Missouri 11 4 -3 Florida 25 1 -3 Washington 11 0 -11 Michigan 18 0 -4 Pennsylvania 23 1 0 After Third Debate. Change Since First EVotes BushUp BushChng Arkansas 6 2 4 California 54 -3 1 New Jersey 15 -6 2 Ohio 21 9 3 Missouri 11 4 3 Tennesee 11 4 1 Louisiana 9 7 3 After Third Debate. Change Since Before First. EVotes BushUp BushChng Indiana 12 19 12 Vermont 3 -6 0 Delaware 3 -11 -7 Utah 5 38 0 Virginia 13 12 7 Illinois 22 -4 8 Nevada 4 12 7 New York 33 -11 10 Georgia 10 10 2 Maine 4 4 3 New Hampshire 4 8 5 Kentucky 8 9 5 Idaho 4 38 ? Oregon 7 5 -4 Rhode Island 4 -29 1 Montana 3 28 6 West Virginia 13 2 -2 After Second Debate Since First EVotes BushUp BushChng Michigan 18 4 -1 Missouri 11 7 -2 Pennsylvania 23 1 -1 Wisconsin 11 6 4 North Carolina 14 10 8 Florida 25 4 3 After Second Debate Since Before First EVotes BushUp BushChng California 54 -4 2 Washington 11 11 2 Iowa 7 6 9 Minnesota 10 0 3 Maryland 10 -6 -5 Between the First and Second Debates EVotes BushUp BushChng Michigan 18 5 1 Missouri 11 9 9 Wisconsin 11 3 2 Pennsylvania 23 2 7 Ohio 21 6 -4 Tennessee 11 3 2 Arkansas 9 -2 5 Louisiana 6 4 6 Florida 22 1 1 North Carolina 14 2 -9 Texas 32 29 -3 South Carolina 8 16 7 New Jersey 15 -8 4 States not solidly in one camp or the other not surveyed since the first debate State EVotes Date of last poll New Mexico 5 9/26/00 If you want to see the full state-by-state electoral college breakdown go to http://www.portraitofamerica.com and click the 'Electoral Tracking' link. Notes on Rasmussen and 'margin of error' The margin of error in their state polls is now 3-4%, implying that if they did the same survey 20 times they'd get an answer within the margin of error of the one they got this time, 19 of those times. Softening the cutoffs to 3.5%/7.5% implies that my predictions based on their data would be right about 17 or 18 out of those twenty. On November 2 there were only 3 states where one candidate's margin is more than 3.5% but less than 4.5% -- they break 2 with 15 electoral votes for Bush and 1 with 22 for Gore. You might wonder about the significance of the Rasmussen Research data I'm reporting. They will not work for any party, candidate, or elected official. They use automated phone-polling technology to lower the cost of polling and increase the number of interviews they obtain. Their nightly tracking poll produces 1000 interviews, and they roll three nights data together to get a single poll with a margin of error of 1.8%, about half of that for other tracking polls. They report results from only 'likely' voters, as determined by responses they receive. A study they sponsored showed that their pre-election polls during the primary season were the most accurate of any polling organization. There are a couple of possible reasons for that. Talking to a computer by telepho ne more closely reproduces the voting booth experience. There is no possibility of the interviewer's opinion influencing the interviewee. And, there is no possibility of the interviewer fabricating the data to suit themself. I know of no other national polling organization that is reporting state-by-state polls from which electoral data can be determined. Many of the news organizations are giving state-by-state electoral totals, but they aren't saying how the numbers are being determined. The Associated Press runs an electoral college story every week, but admits to determining it from interviews with political observers in each state. Others may be relying on local polls, each with its own methodology, standards, and ax to grind. **************************************************************** (cut here) **************************************************************** ELECTION NIGHT SCORECARD Gore Adds State %B - %G GoreL+S Tossup BushL+S ABC CBS CNN Fox +NBC PBS Massachusetts -21 12 - - D.C. -20 3 - - Hawaii -20 4 - - Rhode Island -19 4 - - New York -14 33 - - Connecticut -12 8 - - Maryland -7 10 - - Vermont -6 3 - - New Jersey -6 15 - - Illinois -4 22 - - California -3 - 54 - Delaware -1 - 3 - Minnesota 0 - 10 - Washington 1 - 11 - Tennessee 1 - 11 - Michigan 1 - 18 - Arkansas 2 - 6 - Florida 2 - 25 - New Mexico 3 - 5 - West Virginia 3 - 5 - Maine 4 - - 4 Iowa 4 - - 7 Wisconsin 4 - - 11 Pennsylvania 4 - - 23 Oregon 5 - - 7 Missouri 5 - - 11 Louisiana 7 - - 9 New Hampshire 8 - - 4 Alabama 8 - - 9 Mississippi 9 - - 7 Kentucky 9 - - 8 Ohio 9 - - 21 Virginia 10 - - 13 Georgia 10 - - 13 North Carolina 10 - - 14 Nevada 12 - - 4 Arizona 12 - - 8 North Dakota 15 - - 3 Colorado 16 - - 8 South Carolina 16 - - 8 Oklahoma 18 - - 8 Kansas 19 - - 6 Indiana 19 - - 12 ZSouth Dakota 20 - - 3 Nebraska 20 - - 5 Alaska 24 - - 3 Montana 28 - - 3 Texas 29 - - 32 Idaho 38 - - 4 Utah 38 - - 5 Wyoming 42 - - 3 Total 114 148 276 ------------------------------------------------------------------------- POLITECH -- the moderated mailing list of politics and technology You may redistribute this message freely if it remains intact. To subscribe, visit http://www.politechbot.com/info/subscribe.html This message is archived at http://www.politechbot.com/ -------------------------------------------------------------------------
Current thread:
- FC: Bush set to win today? And election coverage links Declan McCullagh (Nov 08)