Who heard that McCain was ‘suspending his campaign’ and then that Bush had asked Obama to come to DC today for some kind of summit on the financial markets and assumed that both were stunts to try and keep the VP debate next week from happening.
Hilzoy lays it out
I’m guessing that once the VP debate was ‘postponed’ from next Thursday, and the campaigns began negotiating a new date, the McCain campaign would be happy to re-schedule for any time starting, oh, in May of 2017
Thursday, September 25, 2008
Wednesday, September 24, 2008
Checking in on the polls
Back during the fever swamp of the Republican National convention and the Palin announcement I said we should wait several weeks before looking at any polls and trying to asses what was happening. Classically, each party gets a short-term bounce from their convention, these events happen close enough together to jumble things for a while, previously ‘undecided’ voters change their minds a lot during a relatively short period of time, and a result is tons of variance from one poll to the next.
The conventional wisdom is that it takes 2-3 weeks after the last of the party conventions to see how they dynamics of the race have changed, if at all. We’re three weeks post-MN this week. Go to this poll-aggregator... ...and click on the chart to get a larger version in a pop-up. No extra points for spotting the dates of the Democratic convention, the dates of the Republican convention, and where things have gone since then.
A few thoughts:
1) For months we’ve been saying McCain is in a tough spot because his poll numbers were stuck between 40 and 45, while Obama’s numbers drifted between 45 and 50. Palin’s galvanizing affect on the Republican base seems (as of this week) to have gotten McCain up to around 45%. He needs much more then this.
2) Conventional wisdom holds that after the conventions are over the debates are the last predictable event that has the potential to move these numbers significantly. That said, it’s the rare presidential race where the debates move these numbers more then a few percentage points. Most often all the in-debate stumbles and zingers of each candidate net out to very little change in the polls
3) A few years ago Nissam Taleb came out with a very good, interesting/provocative book called The Black Swan. The core argument was that you cannot accurately predict the incidence or impact of highly improbable, high-impact events on any endeavor. While not primarily about politics, it’s easy to apply this to US elections. When highly-improbably events happen, we grope for historical references (Palin = Spiro Agnew) because we have nothing else to go on. Part of what this groping should tell us is that we really don’t know how things are going to play out. Or what or when the next highly-unlikely, high-impact event is going to be. Just looking at recent political history:
- Elian Gonazlez lives, his mother dies; tens of thousands of Cuban-Americans shift their votes before the 2000 election
- The White House has interns and Bill Clinton notices one
- Obama gives a great convention speech
- McCain chooses an unknown evangelical woman to be his VP
- Paulsen worries about moral hazard, makes the decision not to bail out Lehman and let it fall into bankruptcy, and financial markets panic
- What’s next? Will it happen before November?
4) Much has been written about the Bradley/Wilder effect, and the speculation that significantly fewer voters will actually vote for a black men then are saying in polls that they’ll support Obama. All the research I’ve seen on the primaries indicates that no Bradley/Wilder effect showed up in the contests between Clinton and Obama; which is to say that all the pre-caucus/primary errors in polling were more convincingly explained after the fact by other variables. This week the Pew center released research on the fact that people who own cells phones but have no land-line are consistently under-represented in telephone polls, which overwhelmingly don’t call cells phones. These people are largely under 30, urban, and pro-Obama. Pew estimates that Obama is actually running 2-3% better nationally then polls indicate because of this effect. Ironically, this 2-3% range is exactly the same as the estimates I’ve seen for the size of the Bradley/Wilder effect, should one exist. So perhaps all this nets out to zero, and Obama is running exactly where he appears to be. Perhaps not.
5) The fundamentals are still very strong for the Democrats. In addition, it’s easy to construct an explanation for how national attention on the economy and Wall Street helps Obama and hurts McCain. The Clintons, RCC and the McCain campaign have now spent ~7 months and (last time I saw numbers) over $100M on negative attacks designed to tear down Obama’s favorability ratings. The net results of all this has been to move Obama from ~60% favorability in February to ~58% favorability today. If Kerry, Gore or Dukakis had been so resilient to negative attacks they would have won the presidency in a landslide. With six weeks left one could argue that McCain is running out of time to move these numbers, a task made harder if people are obsessing about Wall Street and their 401(k)s and paying less attention to character attacks. The media and the McCain campaign have spent the last month falling into an all-out war, sparked by a number of disputes, which is going to make it much harder for McCain to get positive press coverage or control the news cycle during the home stretch.
Peggy Noonan, Regan’s former speech-writer and conservative matriarch of the Wall Street Journal’s op-ed page, apparently doesn’t need to see any more. Two weeks ago she was captured on CNN explaining to a few other talking heads—none of whom knew their mikes were on—that “it’s over”.
The conventional wisdom is that it takes 2-3 weeks after the last of the party conventions to see how they dynamics of the race have changed, if at all. We’re three weeks post-MN this week. Go to this poll-aggregator... ...and click on the chart to get a larger version in a pop-up. No extra points for spotting the dates of the Democratic convention, the dates of the Republican convention, and where things have gone since then.
A few thoughts:
1) For months we’ve been saying McCain is in a tough spot because his poll numbers were stuck between 40 and 45, while Obama’s numbers drifted between 45 and 50. Palin’s galvanizing affect on the Republican base seems (as of this week) to have gotten McCain up to around 45%. He needs much more then this.
2) Conventional wisdom holds that after the conventions are over the debates are the last predictable event that has the potential to move these numbers significantly. That said, it’s the rare presidential race where the debates move these numbers more then a few percentage points. Most often all the in-debate stumbles and zingers of each candidate net out to very little change in the polls
3) A few years ago Nissam Taleb came out with a very good, interesting/provocative book called The Black Swan. The core argument was that you cannot accurately predict the incidence or impact of highly improbable, high-impact events on any endeavor. While not primarily about politics, it’s easy to apply this to US elections. When highly-improbably events happen, we grope for historical references (Palin = Spiro Agnew) because we have nothing else to go on. Part of what this groping should tell us is that we really don’t know how things are going to play out. Or what or when the next highly-unlikely, high-impact event is going to be. Just looking at recent political history:
- Elian Gonazlez lives, his mother dies; tens of thousands of Cuban-Americans shift their votes before the 2000 election
- The White House has interns and Bill Clinton notices one
- Obama gives a great convention speech
- McCain chooses an unknown evangelical woman to be his VP
- Paulsen worries about moral hazard, makes the decision not to bail out Lehman and let it fall into bankruptcy, and financial markets panic
- What’s next? Will it happen before November?
4) Much has been written about the Bradley/Wilder effect, and the speculation that significantly fewer voters will actually vote for a black men then are saying in polls that they’ll support Obama. All the research I’ve seen on the primaries indicates that no Bradley/Wilder effect showed up in the contests between Clinton and Obama; which is to say that all the pre-caucus/primary errors in polling were more convincingly explained after the fact by other variables. This week the Pew center released research on the fact that people who own cells phones but have no land-line are consistently under-represented in telephone polls, which overwhelmingly don’t call cells phones. These people are largely under 30, urban, and pro-Obama. Pew estimates that Obama is actually running 2-3% better nationally then polls indicate because of this effect. Ironically, this 2-3% range is exactly the same as the estimates I’ve seen for the size of the Bradley/Wilder effect, should one exist. So perhaps all this nets out to zero, and Obama is running exactly where he appears to be. Perhaps not.
5) The fundamentals are still very strong for the Democrats. In addition, it’s easy to construct an explanation for how national attention on the economy and Wall Street helps Obama and hurts McCain. The Clintons, RCC and the McCain campaign have now spent ~7 months and (last time I saw numbers) over $100M on negative attacks designed to tear down Obama’s favorability ratings. The net results of all this has been to move Obama from ~60% favorability in February to ~58% favorability today. If Kerry, Gore or Dukakis had been so resilient to negative attacks they would have won the presidency in a landslide. With six weeks left one could argue that McCain is running out of time to move these numbers, a task made harder if people are obsessing about Wall Street and their 401(k)s and paying less attention to character attacks. The media and the McCain campaign have spent the last month falling into an all-out war, sparked by a number of disputes, which is going to make it much harder for McCain to get positive press coverage or control the news cycle during the home stretch.
Peggy Noonan, Regan’s former speech-writer and conservative matriarch of the Wall Street Journal’s op-ed page, apparently doesn’t need to see any more. Two weeks ago she was captured on CNN explaining to a few other talking heads—none of whom knew their mikes were on—that “it’s over”.
Tuesday, September 16, 2008
Subscribe to:
Posts (Atom)