November 2004, Part 1

Jim Miller on Politics

Pseudo-Random Thoughts

Bird Watcher Saves Millions:  And maybe some lives.
The Israeli ornithologist Yossi Leshem is no ordinary twitcher. An expert in bird migration patterns across the Middle East, he has turned his passion into a small industry responsible for saving the Israeli air force more than half a billion dollars in hardware - and, no doubt, the lives of several pilots.
Israel is a crossroads for birds from three continents used, Yeshem estimates, by a billion birds each year, so the hazards to aircraft there are serious.

Leshem's studies have helped in other ways, too.  He has been able to help farmers encourage birds that destroy pests.  And he has reached out to his Arab and Muslim neighbors to try encourage cooperation in many other ways.  Sadly, one of his most interesting projects, which brought Arab and Israeli children together to observe birds, ended after Yasser Arafat started the intifada.

(For more, you can go to his site, which you can switch to English, if you don't happen to read Hebrew.

Americans may need to know that "twitcher" is British slang for bird watcher.  I'm not sure of the derivation, perhaps it's a shortened form.)
- 2:42 PM, 8 November 2004   [link]

Which Poll Company Had The Best Results?  DJ Drummond of Polipundit has a ranking of 11 polling companies.   The worst was the Los Angeles Times.
[T]he LA Times made predictions in 6 states, and in 5 Battleground States.  Nationally, the LAT had 3 calls right, 3 wrong, and was off by an average of 7.67 points.  None of their picks was the closest in a state, and one of their final polls was off by more than 10 points.   In the Battleground States, the LAT made 2 calls right, 3 wrong, and was off by an average of 7.80 points.
You could do better by simple extrapolation from past voting, I suspect.

The best was Survey USA.
SUSA made predictions in 30 states, and in 9 Battleground States. SUSA got 29 right and 1 wrong, and was off by an average of 3.70 points.  So, why does SUSA win with 29/30, and beat RR take second with 33/33?  It comes down to hitting the bullseye.  EIGHTEEN of Survey USA's final polls were the closest for that state, almost twice as many as every other major poll PUT TOGETHER!  Also, none of their polls were invalidated for being more than 10 points off.  In the Battleground States, SUSA got 8 right and 1 wrong, and was off by an average of 3.44 points.  Three of SUSA's final polls in Battleground States were the closest for that state, again the best of any poll.
Rasmussen was a close second, according to the Polipundit.  One interesting detail:   According to a commenter, both Rasmussen and SUSA use automated polls.  I know Rasmussen does.  An automated poll might reduce the problems of interviewer bias.

Real Clear Politics has its own rankings, for both the national polls and the state polls.  The three polls with the best results nationally were Battleground, Pew, and CBS/New York Times.  The best state polls, according to RCP, were Mason-Dixon, Rasmussen, and SUSA.  RCP found that some of the polls were biased toward one candidate, though they don't say by how much.

Both sets of results support an argument that I have made here.  The state polls are not as accurate as the national polls.  Often, you can do better at predicting a state's vote by using a national poll and past election results, than by using a state poll.

This year was a disaster for my favorite pollster, Gallup, which was wrong nationally, predicting a tie, and wrong in Florida, Ohio, Pennsylvania, and Wisconsin.   Three pollsters, Battleground, Fox, and Marist, predicted the wrong winner nationally, and were wrong by enough so that they should check their methodology.

Finally, one point that neither the Polipundit or RCP mention.  Only one of the national polls predicted too high a vote for Bush, and that by just 0.2 percent.  Two were almost exactly right, and thirteen underestimated Bush's success.  Why this systematic error?   Two explanations occur to me.  There was a late movement to Bush, and the Republican ground game of getting their voters to the polls was better this year.  Both could be true.
- 11:17 AM, 8 November 2004   [link]

Sometimes Parties Find Principles, even central principles, inconvenient.  In Canada, members of parliament provide Canadian flags to veterans organizations, just as congressmen and senators do in the United States.  For members of the separatist Bloc Quebecois this poses a problem, since they want to break up Canada, not celebrate it.
MONTREAL (CP) - The Bloc Quebecois reversed itself Sunday and promised to send Canadian flags to insulted Quebec war veterans after all.

Bloc Leader Gilles Duceppe said rookie MP Andre Bellavance's decision not to provide flags to the Royal Canadian Legion branch in Richmond, Que., was misunderstood and politically exploited.   "It's not a question of not providing flags," he told reporters outside a Parti Quebecois meeting in Montreal, where Duceppe explained his vision of a sovereign Quebec.

"There is a difference, I think, between providing a service and promoting (Canada)."
Of course there is, especially when you get on the wrong side of guys with gray hair and chests full of medals.

(The demand for flags that have flown over the US Capitol is so large that, if you visit at the right time, you will see a whole series being raised and lowered, just before they are sent out to grateful constituents.  I guess being over the Capitol for a few seconds still counts.)
- 6:24 AM, 8 November 2004   [link]

George W. Bush Has A Harvard MBA:  There are times when I wonder whether a single journalist knows that, or at least whether any have thought about how it affects his thinking.  Consider, for example, this article on how Bush won Florida.   The Bush organization identified where potential Bush voters lived, built an organization to contact them, and followed through efficiently before the election, and on election day.  
In Pasco County and other places, Republicans relentlessly called and visited members of their base and the so-called lazy voters in the 72 hours before Election Day, extracting promises that they would vote.  Mr. Bunting rented and borrowed six vans to drive voters to the polls on Election Day.  When surveys of voters leaving the polls showed Mr. Kerry leading that afternoon, volunteers crammed into tiny offices to make hundreds of last-minute calls.

Late on Tuesday afternoon, a polling place near Pensacola Christian College ran out of ballots, forcing more than 1,000 voters to wait in the rain for hours.  The Republican Party handed out bottled water to the voters, who sang "God Bless America."  When the polls in most of the state closed at 7 p.m., volunteers statewide placed more than 100,000 get-out-the-vote calls to Escambia County, home of Pensacola, which is on Central time and had another hour of voting.

"It's the hard work and it's the organization that does it," Mr. Bunting said. "If you don't keep contacting them and talking about issues, they're going to be complacent and sit back."
This is just the kind of campaign one would expect from a Harvard MBA, but the reporters don't seem to understand that.  Bush's faith looms so large for them that they don't see all those people with spreadsheets behind him.

(The emphasis on organization is something of a family tradition.  Bush's father began in politics with a party position, not an elected office.)
- 5:38 AM, 8 November 2004   [link]

Worth Study:  The New York Times chart that accompanies this article, showing exit poll results from 1976 through 2004.  The Times is even nice enough to supply it in several versions; there's a picture of the chart in a .gif file, there's an Excel spreadsheet file with the data, and there's an interactive feature which lets you look at parts of the data for the elections from 1976 through 2000.  (I'm not sure why the different versions of the data have different sets of elections.  Only the spreadsheet has the 1972 data, for some reason.)

What can you find in the chart?  How about this: You would have to avoid all news sources in the last few days not to hear the theory that Christian conservatives swamped the polls and determined this election.  It's true that George W. Bush received 63 percent of the white Protestant vote this year, but it is also true that his share was lower than his father's in 1988 (66 percent), Reagan's in 1984 (72 percent), and Nixon's in 1972 (76 percent).  And it was the same as Reagan's in 1980, when Reagan received about the same proportion of the total vote.

And if his opposition to gay marriage alienated gays and lesbians en masse, it is hard to prove it from the table.  Bush received 25 percent of that vote in 2000 and 23 percent this year.   (The number of self identified gays and lesbians in the sample, 4 percent, is small enough so that it is entirely possible that there was no difference in their vote at all.)  It is worth remembering that some gays oppose gay marriage, sometimes because they oppose marriage in general, sometimes because they oppose it for themselves.

There's much more there.  Take a look, for instance, at the curious relationship between voting Republican and education.  And expect to see many news stories that are contradicted by the data in this widely available table.
- 5:27 PM, 7 November 2004   [link]

Willie Horton And Gay Marriage:  When political parties lose, they often latch on to an excuse that makes them feel better about losing.  For example, as I have mentioned before, party activists believe, almost universally, that they sometimes lose because they are not as nasty as the other party.  (Sports fans can probably think of parallels.)   After their defeat in 1988, Democrats fastened on to the Willie Horton ads to explain why George H. W. Bush defeated Michael Dukakis.  This year they have already decided that gay marriage explains why George W. Bush defeated John F. Kerry.

I've seen many examples of the gay marriage excuse already.  Here's one from a Seattle Times columnist and here's another from Bill Clinton, himself, though as always with Clinton one can't be entirely sure that he believes his own story.   (And, as you may know, Clinton is having it both ways, saying in some places that the Republicans used the issue to win, and in other places that he had advised Kerry to use it, too.)

These two excuses have two things in common; neither is plausible as a reason for their candidate's defeat, and both allow the Democrats to feel noble.  They lost, they can tell themselves, because the Republicans used an illegitimate appeal, to race in 1988 and to "homophobia" in 2004.  (I put homophobia in quotes because it is a scare word that almost never fits where it is used.  Literally, it means the fear of homosexuals, but few who oppose gay marriage fear homosexuals.)

In fact, there is a simpler explanation of both victories.  The economy had been doing very well in 1988 and is doing well this year.  George H. W. Bush is more likeable than Michael Dukakis, and George W. Bush is more likeable than John Kerry.  Both Bushes hold values that are closer to those of most Americans than their opponents — and not just on gay marriage.   Excuses like Willie Horton or gay marriage are not required to explain the Democratic defeats in 1988 and 2004.  If the party and its allies in the media are smart, they will not cling to these excuses and set themselves up for more defeats.

(Need a refresher on Willie Horton?  He is a brutal criminal who murdered a man in Massachusetts and received a life sentence.  At that time, Massachusetts had a policy of allowing even convicts with life sentences to have weekend leaves.  Horton escaped on one such leave, traveled to Maryland and brutalized a couple there.  He was re-captured and returned to Massachusetts.  Dukakis refused to meet with the couple.

A small Massachusetts newspaper raised the leave policy as an issue and did a series of articles attacking Dukakis on the subject, eventually winning a Pulitzer prize for their work.  The Readers Digest picked it up and gave it further publicity.  It became an important issue in Massachusetts politics and a symbol of how out of touch Dukakis was.  In the 1988 primaries, Al Gore raised it, at the suggestion of the New York governor, Mario Cuomo.

In the general election, the Bush campaign raised the issue again, showing a revolving door with criminals getting out, but with no picture of Willie Horton.  An independent group thought the ad too mild and did another ad on the subject with a picture of Horton, who is, as it happens, black.  The Bush campaign asked them to stop running the ad and they did.  Almost immediately, the Willie Horton ad became proof for Democrats, and their media allies, that Bush had used racial appeals to win — unlike, say, Jesse Jackson, Al Sharpton, Al Gore, et cetera, et cetera.  The Bush campaign presented the facts that won a Pulitzer for a a small newspaper — and was forever thought to be racist by many Democrats and journalists.   The facts did not matter, just as they do not on gay marriage.

Finally, some may be wondering why, as I did at the time, prisons would give weekend passes to convicts with life sentences.  Some prison bureaucrats liked them because the passes gave them a carrot to help control the prisoners.)
- 10:31 AM, 7 November 2004   [link]

Did Gay Marriage Make The Difference?  Not judging by the results in Washington and Oregon.  You have probably heard the argument that John Kerry lost because he didn't oppose gay marriage.  For a certain kind of person, this is an agreeable idea, because it allows them to see his loss as caused by his noble refusal to appeal to — as they see it — bigotry.

There are two problems with this argument.  First, John Kerry did oppose gay marriage and explicitly endorsed a Missouri initiative banning it — at least before he opposed the initiative.  In spite of this flip-flop, most of the time during the campaign Kerry opposed gay marriage.  (Some of his supporters didn't believe him, though I can't imagine why.)

Second, it doesn't fit the data, at least in Washington and Oregon, two states that are broadly similar politically.  Oregon had an constitutional amendment on the ballot banning gay marriage on the ballot, which passed easily; Washington did not.  In both Washington and Oregon, Kerry ran better than Gore had in 2000.  (That's going by the current returns.  There are many ballots yet to be counted in Washington, and, I believe, in Oregon as well.)  Kerry's margin was 4 points better than Gore's in Oregon and just 2 points better in Washington.  The gay marriage amendment does not seem to have helped Bush in Oregon.
- 11:01 AM, 6 November 2004
More:  I was lazy and just looked for an example; Paul Freedman was not and looked at the results in all the states with gay marriage on the ballot.
It's true that states with bans on the ballot voted for Bush at higher rates than other states.   His vote share averaged 7 points higher in gay-marriage-banning states than in other states (57.9 vs. 50.9).  But four years ago, when same-sex marriage was but a twinkle in the eye of the Massachusetts Supreme Judicial Court, Bush's vote share was 7.3 points higher in these same states than in other states.  In other words, by a statistically insignificant margin, putting gay marriage on the ballot actually reduced the degree to which Bush's vote share in the affected states exceeded his vote share elsewhere.
Those facts may not settle the argument, but they put the burden of proof on the other side.   Without strong contradictory evidence, I would conclude that the gay marriage issue did not help Bush this year.

There's a general point worth remembering.  The average voter can be remarkably uniformed.   I am absolutely certain that many voters, probably a majority, did not know that Bush opposed gay marriage and that Kerry only sort of opposed it.
- 5:49 PM, 7 November 2004   [link]

Two Election Indicators Fail, and one gets modified.  You have probably already heard that the Redskins indicator failed to predict this election.  So did another indicator, the team colors indicator.  According to Standard and Poors, the Democrats win 71 percent of the time "when a team with red uniforms wins the World Series".   And Bush's win lowered the margin for the October Dow Jones indicator.  In the past, every time the Dow fell more than .85 percent in October, the incumbent lost; if it fell less than .25 percent, he won.  It fell .52 percent, and Bush won.

(Not sure how the Redskins indicator works, or I should say worked?  Here's the rule:
With no exception, a Redskins win in the final home game before a presidential election has meant that the incumbent party wins; a loss has meant the incumbent party is sacked.  Until now.
It's fun to try to think of an explanation for these indicators. The third is pretty obvious; the Dow is a measure of the economy, and even more a measure of expectations about the economy.   The Redskins, located in Washington, D. C., are the team of the incumbents, and red is the traditional color for leftist parties, even though almost all our election maps now use red for Republicans.

Taken from an article in Thursday's Wall Street Journal.)
- 10:07 AM, 6 November 2004   [link]

One More Bit Of Evidence That Votes Were Stolen In Palm Beach In The 2000 Election:  I have been writing about this question since 2000, and included it in a comprehensive Q&A on the 2000 election.  Briefly, there is strong evidence in the returns that about 15,000 votes were stolen from George W. Bush in Palm Beach county, most likely by invalidating punch cards, since there was a very high rate of spoiled ballots in the county.  (For more evidence, see this post.)

Yesterday, it occurred to me that a comparison between this year's election results and the 2000 election results could give us one more bit of evidence that some Democratic operative(s) in Palm Beach almost stole the 2000 presidential election.  One thing that struck me in 2000 is that Bush's percentage of the vote in the county was lower than the percentage of Republicans registered there.  That's extraordinary in a close election.  Was the same true this year?  No.  According to the data I unloaded from the office of the Florida Secretary of State, 32 percent of the voters in Palm Beach registered as Republicans, and 45 percent registered as Democrats.  Bush's 39 percent of the vote this year is just what one would expect from those registration numbers — and much higher than it was in 2000.

I am even more certain that votes were stolen from Bush in Palm Beach in the 2000 election, and I think that John Fund's estimate of 15,000 votes is about right.

There's a point from my September post that bears repeating.
Finally, one bitter point about the media.  Fund and others have tried to interest journalists at "mainstream" news organizations in the Palm Beach vote story.  None have thought it worthwhile to study the near theft of a presidential election.
I am not John Fund, but I have also tried to interest journalists at "mainstream" news organizations in stories of vote fraud.  I haven't had any success either.

(One thing I wondered about in my September post is why the operative(s) did not quite steal enough votes to tip the election.  It is just a guess, but I think that they may not have had enough time to do the calculations correctly and spoil the Bush ballots, so they guessed, and luckily for all of us, they guessed wrong.)
- 10:53 AM, 5 November 2004   [link]

Republicans Will Regret that this news couldn't have come out on Monday.
The U.S. economy created jobs last month faster than in the previous six, the Labor Department reported today.

The number of new jobs in October, 337,000, far exceeded economists' forecasts of about 169,000.

Three days after the election -- in which job creation was an issue -- the Labor Department also announced that its previously reported job-growth figures for August and September had been too low, considerably so.
. . .
Revisions -- upward and downward -- are common as statistics become more complete.  Today's revisions were larger than usual, however.  The hiring increase in August shot up from the previous estimate of 128,000 to 198,000.  The original number for September, 96,000, jumped to 139,000.
There were 71,000 new construction jobs in October, some created by the hurricanes in the Southeast, though the Post doesn't say how many.  The same hurricanes may have depressed job growth in September.
- 9:19 AM, 5 November 2004   [link]

Republican Like Me:  Seattle PI columnist Robert Jamieson passes as a Republican and finds Seattle not completely tolerant of that minority.
During the presidential campaign, when I would bump into folks on the street, I would say: "Hi. I'm with a group called Blacks for Bush.  We are trying to help our president win four more ..."
. . .
But my unscientific experiment, involving encounters with dozens of people, made me realize a truth in what conservatives often kvetch about: The Emerald City ain't so pretty if you are down with the GOP.

To embrace a Bush state of mind here means keeping one's head down and not saying a word that could generate a mean stare, raised eyebrow or verbal dart.
Jamieson ends by arguing people in Seattle should not dismiss Republicans as "rednecks or religious numbskulls".  Many in Seattle will find his argument novel and even distasteful, since their sense of superiority rests on their contempt for the majority in the country.   Kudos to Jamieson for making it, anyway.  (But don't expect the PI to hire a Republican to write on politics any time soon.  That's almost as unlikely as a deep South newspaper in the 1950s hiring a black to write on race relations.)

(Younger readers may miss the reference in the title of this post.  I am putting a twist on the title of one of the great civil rights books, Black Like Me.
Concerned by the lack of communication between the races and wondering what "adjustments and discriminations" he would face as a Negro in the Deep South, the late author, a journalist and self-described "specialist in race issues," left behind his privileged life as a Southern white man to step into the body of a stranger.  In 1959, Griffin headed to New Orleans, darkened his skin and immersed himself in black society, then traveled to several states until he could no longer stand the racism, segregation and degrading living conditions.
The book affected my thinking profoundly, as it did many other whites in both the North and the South.  And the sympathy I gained for American blacks is one of the reasons that I later left the Democratic party for the Republican party.  I became convinced that Democratic measures, especially the vast expansion of welfare in the 1960s and 1970s, had damaged the people they were intended to help, especially the black poor.  Democrats seemed incapable of realizing that, though it was obvious to many recipients.  Even now, many Democrats reject welfare reform, which has been, on the whole, an enormous success.)
- 4:24 AM, 5 November 2004   [link]

Rural Areas And Charter Schools:  Washington state voters again rejected charter schools, voting against Referendum 55 by 58.5 percent to 41.5 percent.  The reason is not hard to see.  Many Republican leaders backed it, but rural Republican voters did not.  Take a look at the current results, by county, and you will see what I mean.  In little Garfield county, just 1,077 voters had voted on the issue, when I checked this afternoon, and just 35 percent of them had voted yes.  The same county was giving 735 votes to the Republican candidate for governor, Dino Rossi, who supports charter schools, and just 387 votes to his Democratic opponent, Christine Gregoire.  If you look at other rural counties, you will see the same pattern, strong opposition to charter schools and strong support for Republican candidates, many of them favoring charter schools.  If rural voters had given the same support to charter schools that they did to Republicans, Referendum 55 might have passed.

I hope that I do not disappoint some of my friends by saying this, but I think this combination of attitudes by rural voters is rational.  They are right to support Republicans, who are far more likely to listen to their concerns.  And they are not wrong to oppose charter schools, which offer little positive to most rural areas.

The great advantage of charter schools, in my view, is that they introduce competition into education.  But to have competing schools requires a density of population not found in most rural areas.  My little high school had a mere 130 students when I graduated &mdash and several very long bus routes.  (For some from big cities, I should add that 130 was the number for the entire school, not the senior class.)  And it was that large only after two towns consolidated while I was in junior high school.  It was difficult for the two towns to support a single school system; it would have been terribly difficult for them to support two systems.   And if a charter school tried to serve a large rural area, it would find that many of its students had to spend half the school day traveling.

And that isn't all.  A charter school in a rural area would threaten the existence of an existing school system far more than it would in an urban area.  A large suburban school can lose hundreds of students to a charter school and adapt; a rural school would be badly damaged by the same loss.  This possibility would dismay most rural communities, where the schools are, more often than not, their psychological centers.

Given the opposition of rural voters to charter schools, what should their proponents do?  I believe that they should try for a more comprehensive reform, one that offers something to rural areas.  The best approach, I believe, would be to combine charter schools with a measure returning power to local school districts, which would draw strong support in rural areas.  (Another possibility would be to limit charter schools to very large school systems.)  I say returning power because the local school districts here in Washington state have lost much power to the state, and to the Washington Education Association.  I think some decentralization would be good for most schools, as well as being good politics.

That's as far as I want to take this argument for now.  Before I say more, I want to do some research, and I would like to hear reactions from charter school proponents (and opponents).

(The charter school proponents in Washington state could probably have done a better job of selling the proposal to rural areas.  I have some thoughts on that, too, for those who are interested.)
- 1:55 PM, 4 November 2004
More:  Joanne Jacobs, who runs a great edublog, linked to this post here and got some interesting comments.  Most agreed with my argument that charter schools are impractical in rural areas, though one noted Oregon cases in which rural voters are considering replacing their current school with a charter schools, in order to save them.  That, I think, supports my argument that the way to win rural support is to give rural areas more control over their own schools.

And, I received a very thoughtful email pointing out that the defeated referendum "included strict controls on establishing charters, such that a 'takeover' of an existing small district was a virtual impossibility".  That may be so, but it is not an easy argument to convey to voters, especially against the opposition of the WEA.

Let me give some rough numbers to illustrate the political possibilities.  Eighteen percent of Washington's residents, and, I would guess, about the same proportion of Washington's voters, live in rural areas.  They tend to be Republicans who distrust unions and state bureaucracies, so they should be responsive to proposals that trim the power of those groups.  If charter school proponents moved one third of those voters to support their cause, majority support for the reform would be in sight, and they would only need a few more percent from the urban population.
- 5:50 AM, 5 November 2004   [link]

Catching Up:  With the election over, I have been catching up on my sleep and hope to catch up on my email and site maintenance this weekend.  This weekend, because today is sunny, the forecast says that tomorrow will be sunny, and the same forecast says that the weekend will be rainy.  I am still enough of a farm boy to let the weather control my schedule, at least some of the time.
- 8:46 AM, 4 November 2004   [link]

Remember Bobby Jindal, the son of immigrants from India, who narrowly missed becoming governor of Louisiana?  He was just elected to Congress from Louisiana's 1st Congressional district.

The New York Times may not have paid much attention to this story, despite Jindal's breakthrough victory and his remarkable personal story, but the Times of India did, putting Jindal's picture ahead of Bush's picture in this illustrated story of election celebrations.

And the Times of India has much more to say about the US election.  In this story, they say that Indian techies are happy.
The Indian IT industry Thursday expressed delight at US President George W Bush's re-election, hoping he would continue to be a proponent of free trade and outsourcing.

"The US is a powerhouse of the global economy and the Indian IT industry is delighted in being a partner to US corporations, helping to make them more competitive," India's National Association for Software and Service Companies (NASSCOM) president Kiran Karnik said in a statement.
In this story, they list the candidates of Indian descent, who are roughly equally divided between Republicans and Democrats.  I was fascinated to learn that the Republican candidate for secretary of state in North Carolina (!) was Jay Rao, an Indian-American woman with a lovely smile.   (She lost to a two term incumbent, Elaine Marshall.)

I have long thought that the Republicans should make more effort to gain support from East Asians, mostly those of Japanese, Chinese, and Vietnamese descent.  I now think that they should do the same with South Asians, almost all of Indian descent.

(Jindal is not a favorite of the New York Times, despite his brilliance and his personal story; Michelle Malkin doesn't hold back as she describes their condescension.
The condescension of The New York Times toward minority conservatives is so thick, you need a Cuisinart electric carving knife to slice it.
Jindal is the second Congressman of Indian descent; the first was Dalip Singh Saund who won the 23rd district of California in 1954 and held it for many terms.  Like Jindal, Saund had an impressive personal story.

Changing the subject, I noticed that the Times of India has a livelier front page than the New York Times.  There is a picture story on hot November babes, a story on a beauty queen "held for sex trade", and a story on an unsatisfied wife.  The New York Times might cover most of these subjects, but not in quite that tabloid style.)
- 7:19 AM, 4 November 2004   [link]

Just Wondering:  If Dan Rather weren't so close to retirement, would CBS fire him or demote him?  Not because he campaigned to drive George Bush from office, but because he failed in that campaign.  I don't have any inside information, but when Rathergate came along, CBS treated him like an old quarterback who had thrown too many interceptions.  They didn't disagree with his goal, but they weren't sure he could still deliver.
- 6:03 AM, 4 November 2004   [link]

Bush Wins, even though the "mainstream" media and their candidate, John Kerry, have not conceded.  Bush wins since he won Ohio by a large enough margin — 136,221 votes as I write — to overcome any plausible Kerry gain from provisional ballots.   Bush wins because his lead in New Mexico is large enough — 11,620 votes as I write — to survive a recount.  And Bush may win Iowa and Nevada as well: he had a narrow leads in both last night, when they stopped the counts.

Bush is winning the national popular vote by, as I write, 3.5 million votes.  (That margin may decline a little as they count the absentee ballots in California.)  His vote total — 58.3 million votes as I write — gives him both an absolute popular majority and the largest number of votes for any presidential candidate, ever.

Finally, we now know why John Kerry chose a one term senator, with no political accomplishments, for his running mate.  Kerry planned, all along, to try to win the election in the courts, if he could not win it at the ballot box.  And that is why Kerry sent trial lawyer Edwards out last night to make this odd statement.
John Kerry and I have made a promise to the American people that with this election every vote would count and every vote would be counted.  Tonight we are keeping our word and we will fight for every vote.
Note that Edwards does not say "every legal vote".

(The election provided a glaring example of "mainstream" media double standards.  Kerry won the state of Pennsylvania by, as I write, 121,818 votes, which is smaller than Bush's margin in Ohio.  There were many reports of irregularities in Philadelphia and other parts of the state.  The networks declared Pennsylvania for Kerry early, but some have still not declared Ohio for Bush.   None of the networks seem much interested in the Pennsylvania irregularities.  All are fascinated by Ohio.)
- 7:35 AM, 3 November 2004
Update:  Kerry has phoned Bush and is going to formally concede later today, for which we may all be grateful.
- 8:42 AM, 3 November 2004   [link]

The Results Look Good Enough For Bush so that I am heading off to the Republican party in Bellevue.  Given the fact that Bush appears to be doing better than the polls projected, there may be good news in Washington state races, too.
- 7:17 PM, 2 November 2004   [link]

Another Good Sign:  Peter Jennings was interviewing former Nebraska senator Bob Kerrey and asked him how the Democrats would treat John Kerry if Kerry was defeated.   Jennings has looked resigned in the bits that I have seen this evening.  He's my favorite anchor on Republican election nights, because he catches on early and shows his unhappiness at the results.  Dan Rather can be funny, but I can't help but suspect that his "Ratherisms" are written for him by some English major from Yale.
- 6:35 PM, 2 November 2004
More:  Just saw Katie Couric.  I didn't expect the election results to make her happy, but I was still taken aback by the angry, bitter woman I saw on TV.   She was wearing a black dress that would not be out of place at a funeral, and she looked as if she had lost a child to a brutal murderer.  Peter Jennings was fun to watch last night, but Couric was more than a little disturbing this morning.
- 8:50 AM, 3 November 2004   [link]

Early Results:  With 34 percent of the vote in, Kentucky is giving Bush a bigger margin than in 2000, 16.6 percent versus 15.1 percent.  And his margin has been growing while I have been watching it.  With far fewer votes reported (707 of 5024 precincts), Indiana is giving Bush a much bigger margin than in 2000, 23.4 percent versus 15.7 percent.
- 4:57 PM, 2 November 2004
More:  With 92 percent of the precincts reporting, Kentucky is giving Bush nearly a 20 percent lead.  It looks almost certain that Bush will do better in this border state than he did in 2000.
- 6:21 PM, 2 November 20004   [link]

Need An Election Night Guide?  Then print out a copy of this John Fund column, which tells you what to watch for, and when.  For what it is worth, Fund is predicting a Bush win by a very small margin, 1.5 percent.
- 3:36 PM, 2 November 2004   [link]

Final Election Prediction:  (Some of you will be saying, "about time", and you would be right.)  I have hesitated because I am genuinely uncertain about the outcome, as I was not in 2002.  The polls almost all give Bush a small lead, but if you look at this chart giving the poll averages, it is not hard to make a case for Kerry trend through the last half of October.   The pollster I trust most, Gallup, rated the race a tie — after allocating 90 percent of the undecideds to Kerry, which seems dubious to me, .

So, let's go back to the basics.  Currently, I believe that the two parties are very close in identifiers.  The Democrats might have a small lead, but that is more than balanced by the Republican edge in participation, which I expect will continue this year.  Republicans, according to the polls, are more loyal to Bush than Democrats are to Kerry, in spite of all the talk you hear about Republicans switching to Kerry.  Issues are about even, with Bush advantages on some counterbalanced by Kerry advantages on others.  Bush is the incumbent.   Voters like Bush as a person more than they like Kerry.

All that still adds up to a victory for Bush in my estimate, probably by 4 percentage points.   But I do worry that I am letting my partisanship affect my analysis, although I have tried very hard to avoid that common trap.  On the other hand, I take some comfort in this remarkable concession column, from Mark Mellman, who has been polling for Kerry this year.
First, we simply do not defeat an incumbent president in wartime.  After wars surely, but never in their midst.  Republicans have been spinning this fact for months, and they are correct.

Democrats have spoken often and powerfully about the nation's economic problems.  But by historical standards, they are not that bad.  The "misery index" is 7.8 today but was 20.5 when Jimmy Carter was defeated.  Economic models of elections show Bush winning 52-58 percent of the vote.
And near the end, he comes up with this summary figure.
Taking all that and more into account, an expert forecasting model suggests that Bush will get 51.6 percent of the two-party vote.
Which differs from my estimate by just 0.4 percent.

Naturally, Mellman ends by saying he is optimistic, but no Democrat should find reason for optimism in his analysis.  I take his analysis seriously because he has been immersed in poll data for months, including much private data that is unavailable to you and me.

Finally, I should give you a warning in case Mellman and I are wrong.  The betting markets, when I looked a few minutes ago, had swung sharply against Bush for the first time in months.  (They may be over-reacting to early, and incorrect, reports on exit polls.)
- 1:38 PM, 2 November 2004
Update:  How close was my prediction?  As I write, Bush has 51.55 percent of the two party vote; I had predicted 52 percent.  That's close, but not as close as the model Mellman cites.  I'll do a comparison of predictions later, when the returns are final.  By way of comparison, the average error in the 2000 national polls was 1.8 percent, if I recall correctly, but I am not going to claim I can replace Gallup and company.
- 9:15 AM, 3 November 2004   [link]

There Were No Lines when I went to vote at 11:40 this morning.   Five precincts vote at the Lakeview elementary school.  The machine that read my optical ballot had a counter, which showed that mine was the 339th vote cast in those five precincts.   So, judging only from that single location, in what is not considered to be a swing state, I would not expect the more extravagant predictions for turnout to come true.  At a guess, it will be similar to the 1992 turnout.  (I should add that in every election, a higher proportion of Washington voters use mail ballots.)

As it always is in King county, the ballot was poorly designed.  It began with the propositions and didn't get to the presidential candidates until the bottom right corner of the front of the ballot.  As always, I was annoyed by the flimsy voting stands, which offer little privacy and wiggle as I fill in the ovals.  And I was saddened when I walked back and noticed that, on the corner just west of the school, both Democratic and Republican signs had been stolen.
- 12:56 PM, 2 November 2004   [link]

Matt Drudge is reporting vote fraud in Philadelphia.
Before voting even began in Philadelphia -- poll watchers found nearly 2000 votes already planted on machines scattered throughout the city... One incident occurred at the SALVATION ARMY, 2601 N. 11th St., Philadelphia, Pa: Ward 37, division 8... pollwatchers uncovered 4 machines with planted votes; one with over 200 and one with nearly 500... A second location, 1901 W. Girard Ave., Berean Institute, Philadelphia, Pa, had 300+ votes already on 2 machines at start of day... INCIDENT: 292 votes on machine at start of day; WARD/DIVISION: 7/7: ADDRESS: 122 W. Erie Ave., Roberto Clemente School, Philadelphia, Pa.; INCIDENT: 456 votes on machine at start of day; WARD/DIVISION: 12/3; ADDRESS: 5657 Chew Ave., storefront, Philadelphia, Pa... MORE... A gun was purposely made visible to scare poll watchers at Ward 30, division 11, at 905 S. 20th St., Grand Court.  Police were called and surrounded the location... Developing...
This looks like the raw reports from Republican poll watchers.  Some may turn out to be false alarms, but given Philadelphia's long history of fraud and violence at its elections, I would not be surprised if most turned out to be true.

Let me repeat something, because it is important.  The increase in vote fraud, mostly by Democrats, means that every narrow win by a Democrat will be suspect.

(The apparent tampering with the voting machines in Philadelphia reminds me of a great irony.   Many oppose the use of touch screen voting machines because some of the machines do not print records of the vote.  But the same is true of mechanical voting machines, which were introduced to prevent fraud — and often succeeded in that.  It is possible, as the Drudge report shows, to tamper with mechanical voting machines, but it requires skills that not everyone has.  Anyone who can see paper ballots can tamper with them.)
- 8:07 AM, 2 November 2004   [link]

Looking Back Over The 2004 Campaign what I find most striking is this.
Sen. John Kerry has gotten the white-glove treatment from the press, garnering more praise from journalists than any other presidential candidate in the last quarter-century, according to a new analysis of almost 500 news stories released today by the Center for Media and Public Affairs.

"It's not just that John Kerry has gotten better press than President Bush before this election, he's gotten better press than anyone else since 1980.  That's significant," said Bob Lichter, director of the D.C.-based nonpartisan research group.
Bob Lichter (or S. Robert Lichter, the name he uses on his research work) is a careful scholar with enormous experience in studying the media.  I trust his results.

That better press has extended to a refusal to pressure Kerry to release records that most presidential candidates (and vice presidential candidates) release.  In 1964, the press did not rest until the husband of the vice presidential candidate, Geraldine Ferraro, had released his tax returns.  There has been no pressure on Teresa to release her full tax returns, even after the two pages she did release suggested many possibilities for investigation.  There has been no criticism of Kerry's choice of a running mate with no executive experience and no accomplishments in the Senate, a man who makes Dan Quayle look overqualified.  And there are many other areas that might have drawn the attention of the press; for examples, see this critique of Kerry from a supporter, Mickey Kaus.

The media decided, early in the election season, to hold their noses and to try to drag Kerry over the finish line.  Evan Thomas of Newsweek said that the press was worth 15 points to Kerry, then later lowered that to 5 points.  He was more frank, but not partisan, than most in the media.

There is an example from the 1992 campaign that shows just how openly partisan the media have become, and how estranged they are from much of their audience.  In that year, ABC, along with the other networks, received thousands of (often justified) complaints about their coverage of the campaign.   ABC took the complaints seriously enough to investigate them and to do a special in the spring of 1993, which conceded, I thought, some points to the critics.  (And produced a few delicious moments when journalists were faced with the same "gotcha" questions they like to direct to others.)  Does anyone think that CBS, which must have received far more complaints this year than ABC did in 1992, will do a similar self examination?

And that is in spite of the fact that this bias is costing the newspapers and the networks millions of dollars.  If you look at this latest circulation report, I am sure you will be struck, as I was, by the losses at the Los Angeles Times, which tried to save Gray Davis, and the gains at the New York Post, which has supported President Bush.  The gains for Fox and the losses for ABC, CBS, CNN, and NBC are even more striking.

I am not sure many in the media understand how much they risk by this blatant bias,  The worst result for them would be a Kerry victory, which would further embitter all those who feel the coverage has been unfair.  If I were a stockholder in a major media company, I would be hoping for a Bush victory.
- 7:25 AM, 2 November 2004   [link]

What Do The State Polls Say About The National Race?  If pollsters are doing the work correctly, the state polls should, collectively, give us about the same answer as the national polls. To see whether they do, I tabulated the margins from all the state polls at Real Clear Politics yesterday and the margins from the 2000 election.  If the RCP site had a poll average for a state, I used that.  If not, I used an average of the last three polls, and, in a few cases, the last two polls or the only poll. For those who want to take the analysis further, I have added the electoral votes for each state.

Bush Margins in 2000 and 2004, by State

Delaware-13.1 - - 3
D. C.-76.0 - - 3
New Hampshire1.3-1.2-2.54
New Jersey-15.6-7.08.615
New Mexico-
New York24.317.07.331
North Carolina12.97.7-5.215
North Dakota27.624.5-2.13
Rhode Island-29.1-17.012.14
South Carolina15.9 - - 8
South Dakota22.713.7-9.03
West Virginia6.38.52.25

What the table suggests is that Bush has gained in most states since the 2000 election, when he lost the popular vote by 0.5 percent. If the gains in larger states such as California and New York are real, then it is hard not to conclude that Bush has gained at least two or three points from his 2000 vote.  If that is the case, then Bush would win the popular vote by 1.5 to 2.5 percentage points.

(Several caveats: Some of these polls, especially in the states that are not close, were taken some time ago.  As I have mentioned before, I do not think that state polls, in general, are as well done as the best national polls.  Some of the polls were of registered voters, not likely voters.  And by using yesterday's data, I almost certainly missed some late polls.

There are some interesting patterms, subject of course to those caveats, in the data.  Bush is running behind his 2000 margin in all three upper New England states, Vermont, New Hampshire, and Maine, and ahead of his 2000 margin in the three lower New England states, Connecticut, Rhode Island, and Massachusetts (!).  Bush gains in Tennessee and losses in North Carolina show the effects of favorite sons in those two states.)
- 6:41 AM, 2 November 2004   [link]

John Kerry's Missing Records:  There are three missing sets of Kerry records, Teresa's full tax returns, Kerry's full medical records, and his complete military records.  Unlike President Bush, he has not authorized the full release of his military records.  As with the tax returns and the medical records, this has led to speculation about what he might be hiding.

Now, Thomas Lipscomb suggests that Kerry may be hiding a less than honorable discharge.
A former officer in the Navy's Judge Advocate General Corps Reserve has built a case that Senator Kerry was other than honorably discharged from the Navy by 1975, The New York Sun has learned.

The "honorable discharge" on the Kerry Web site appears to be a Carter administration substitute for an original action expunged from Mr. Kerry's record, according to Mark Sullivan, who retired as a captain in the Navy's Judge Advocate General Corps Reserve in 2003 after 33 years of service as a judge advocate. Mr. Sullivan served in the office of the Secretary of the Navy between 1975 and 1977.
I am unqualified to evaluate the evidence presented in this piece.  But I can say that these suspicions, and similar suspicions about Teresa's tax returns and his medical records, will never end while he refuses to release all the records.
- 3:13 PM, 1 November 2004   [link]

You Find My Vote Choices at Sound Politics, along with those of other contributors to that group blog.
- 2:55 PM, 1 November 2004   [link]

The Red Sox Championship was a win for the numbers guys.
While credit goes to Boston's Terry Francona for making the managerial moves, he is no more a genius than St. Louis's Tony La Russa is a moron for becoming the first manager in Series history to be swept twice (with the A's in 1990 against Cincinnati and manager Lou Piniella).
. . .
That's why the real hero in Boston is [general manager Theo] Epstein, who along with stats-guru consultant Bill James, built a club that any competent manager could manipulate successfully.
Many baseball fans knew this, despite all the talk about the "curse of the Bambino" and other such nonsense.  And so, apparently, did the bookies, who made the Red Sox favorites in all three series.  This reliance on statistics isn't new; Bill James has been writing his Baseball Handbook for many years and has, along with others, changed our understanding of baseball strategy.   But there's a general lesson here that many have missed, and there's a political lesson that almost all "mainstream" journalists have missed.

First, the general lesson: When you understand a problem well enough to put numbers on it, you have, nearly always, a better understanding than when you can use only words.  Measurement is, nearly always, an essential first step in solving problems.  That is especially true of the policy problems faced by elected officials.  The late Senator Patrick Moynihan made this argument often, usually quoting decades old authorities when he did.

Second, the political lesson:  An office holder who understands numbers has great advantages in problem solving over one who does not.  Now then, which of othe two candidates, Bush or Kerry, is better at numbers?  Maybe it is the one who has a Harvard MBA (Bush), rather than the one with a Boston College law degree (Kerry).  (Lawyers, like journalists, are nearly always good with words, and often terrible with numbers.)  Maybe it is the one who is backed by the numbers guys.   (If you missed this column, Brooks showed that those who work with numbers, for example, accountants, were much more likely to contribute to Bush than those who worked with words, for example, journalists.)

The idea that Bush, rather than Kerry, is better at analytical thinking would surprise many journalists.  This analysis by John Harris is typical of many, and the quote he uses from David Gergen (a words guy) gives the conventional wisdom in newsrooms.
David Gergen, a writer and political commentator who worked for Clinton as well as a succession of Republican presidents, said American politics has rarely offered such a stark contrast in leadership models.  The choice, he said, is "between fact-based versus intuition-based policies."   Confronted with a policy decision, Gergen believes, Kerry's instinct is to study and seek to master the complexities of his choice; Bush's instinct is to act quickly, in the belief that a leader is better off to drive events and circumstances rather than be driven by them.
Before you swallow that, let's recall a bit of Bush history.  He was, according to a number of accounts, a fine poker player while he was at Harvard.  That tells me that he is good at estimating probabilities, and at fooling other people into thinking he is using "intuition".   In thinking otherwise, Gergen is not the first to be fooled, and won't be the last.

Why doesn't Bush show this analytical side in public?  Because he knows many voters don't like cold analysis from politicians.  But you do get glimpses from those who meet with Bush in private.  When he worked for education reform in Texas, he prepared himself by meeting with many education experts and studying the main findings.

Some think that Kerry's public dithering shows a man who analyzes problems carefully.  In fact, one of the things taught by most business schools is that managers should make decisions when they have enough information.  (For some problems, there are even formal solutions that specify exactly how much information is "enough", something well known to MBAs, but not to lawyers.)   Dithering when you know enough to decide is always wrong, and sometimes disastrous.

That Bush is better with numbers than Kerry is not the only reason to vote for Bush, but it is a good one, if you want an effective problem solver.  (And if Bush could talk Bill James into coming to the White House and putting his talent to work on policy problems, that would be a plus.)
- 7:58 AM, 1 November 2004   [link]

Washington Post Columnist Backs Increased Voter Fraud:  E. J. Dionne, whom I used to respect, argues that more voter fraud is an acceptable price for greater participation.  As journalists often do, he lets an "expert" do the heavy lifting.
Michael McGerr, a historian at Indiana University, notes that in the post-Civil War years, before these restrictions came into force, the American system was remarkably open -- as long as you were a man. "There wasn't a cumbersome registration system," said McGerr, a proponent of the theory that we're now going through another political revolution.

Often, you didn't have to register at all.  You just showed up on Election Day.  Some jurisdictions, McGerr notes, even allowed immigrants then coming to America in droves to vote, as long as they declared their intention to become citizens.  "The system was willing to risk a certain amount of corruption in order to make sure that anyone who had the right to vote did vote," McGerr said.  "We've made a different tradeoff in the last 100 years.  In order to protect against potential corruption, we've made it too hard for the less educated and less well-off to vote."
North Dakota even now does not register voters, because it is mostly rural.  When most of the nation's population was rural, we didn't need registration, at least outside the cities.

Whatever may have been true in the past, currently vote fraud mostly occurs in areas with low participation.  States with relatively clean elections, such as Washington and Minnesota, have higher rates of voting than states that do not.  There is a strong argument, in fact, that cleaning up elections encourages people to vote, that the prevalence of fraud in many urban areas is one reason that many there don't vote.

If Dionne were right that allowing more fraud would increase participation, then we would have had higher participation after the passage of the 1993 "Motor Voter" Act.  But we haven't had an increase in participation, as he must know, but we have had an increase in fraud, as he may not know.  Nor have other measures that increase fraud, such as the widespread use of mail ballots, increased participation in those states that have tried them.

Those who want to see election fraud decrease favor requiring photo IDs for voting.   Those who want to see election fraud increase oppose requiring photo IDs for voting.  Here's what Dionne says:
Another matter of national concern is the requirement that first-time voters who registered by mail show identification at the polls.  This, too, is a product of the Help America Vote Act, and it would seem a simple barrier against fraud affecting relatively few voters.  But as Alec Applebaum reported recently in the New Republic, some election officials are interpreting the provision to mean that all voters must present ID.  An estimated 5 percent of Americans have no photo ID, and they tend to be "poorer, less educated and more urban," Applebaum writes.  In the absence of explicit national rules, a well-intended but loosely written law might become, in some places, the Help Some Americans Not to Vote Act.
So Dionne opposes photo ID even for voters who have registered by mail and have never shown identification to any election official!  I think it fair to conclude that he is pro fraud.   I suppose that we can give him some credit for frankness.

(Dionne makes several factual errors in the piece.  The Florida Secretary of State in 2000, Katherine Harris, was not "chosen by a GOP governor whose brother was one of the candidates".   She was elected by Florida voters in a general election — after defeating a candidate backed by Jeb Bush in the Republican primary.  Nor did she make "all the critical decisions in her party's favor".  For example, she proposed that all the separate recount disputes in Florida be combined in an argument to the Florida Supreme Court, which would probably have led to a statewide recount.  The Gore legal team rejected the idea, not knowing that the Bush team opposed it, as well.

And he uses a truly funny example to make his argument for the direct popular election of the president, the experience of France.  In the last presidential election in France, the voters had to choose, in the runoff, between a crook (Chirac), and a right wing extremist (Le Pen).  Did Dionne not notice that?  I know the Washington Post covered the French election.)
- 6:06 AM, 1 November 2004   [link]

Osama Borrows From Michael Moore:  Osama bin Laden's latest tape apparently borrowed themes from Michael Moore's Fahrenheit 9/11, as even the Guardian concedes.
Perhaps, some speculate, Bin Laden has managed to see - and be inspired by - Michael Moore's film, Fahrenheit 9/11, with its footage of Mr Bush's slow performance on the day.
(More fair minded people might think Bush's performance was about right, continuing what he was doing while others gathered the facts that he needed and made the preparations to get him in the air.)

When even the Guardian admits that the world's most famous terrorist is borrowing themes from its favorite movie maker, I think we can agree that it is true.  And though the Guardian doesn't go into it, the similarities between Fahrenheit 9/11 and bin Laden's latest tape extend to details such as the fascination both have with the story that the president was reading at the time of the attack.

Thanks to a better translation from MEMRI, we can see another idea that the two men share and that bin Laden may have borrowed from Moore.   Immediately after the 9/11 attack, Michael Moore wrote that he could not understand the attacks on places that had not voted for Bush.  (When the reactions to this outrageous statement came in, Moore removed it from his web site, though as far as I know, he never really apologized for it.)  Here's MEMRI's correction.
The tape of Osama bin Laden that was aired on Al-Jazeera(1) on Friday, October 29th included a specific threat to "each U.S. state," designed to influence the outcome of the upcoming election against George W. Bush.  The U.S. media in general mistranslated the words "ay wilaya" (which means "each U.S. state")(2) to mean a "country" or "nation" other than the U.S., while in fact the threat was directed specifically at each individual U.S. state.  This suggests some knowledge by bin Laden of the U.S. electoral college system.  In a section of his speech in which he harshly criticized George W. Bush, bin Laden stated: "Any U.S. state that does not toy with our security automatically guarantees its own security."

The Islamist website Al-Qal'a explained what this sentence meant: "This message was a warning to every U.S. state separately.  When he [Osama Bin Laden] said, 'Every state will be determining its own security, and will be responsible for its choice,' it means that any U.S. state that will choose to vote for the white thug Bush as president has chosen to fight us, and we will consider it our enemy, and any state that will vote against Bush has chosen to make peace with us, and we will not characterize it as an enemy.
Let's summarize:  Michael Moore implied in 2001 that terrorist attacks on states that supported Bush were understandable.  Osama bin Laden says in 2004 that he will make terrorist attacks on any state that supports Bush.  Now where did bin Laden get that particular evil idea?

One last thought:  Wouldn't it be delightful if a reporter for the New York Times, the Washington Post, or one of the major networks asked Moore about the similarity between his ideas and bin Laden's?

(Moore's original statement, which has been forgotten by most journalists, deserves more comment.  As I am sure you noticed, Moore rests his argument on the discredited idea of collective guilt.  If you live in an area that voted for Bush, you may deserve punishment, even if you personally did not vote for Bush.  This is a common idea among primitives, but rejected by almost all in the West.  That it is held by Moore shows the level of his moral arguments.)
- 4:28 AM, 1 November 2004   [link]