Everybody has seen that wonderful Pollkatz graph:

[Jackpine Radical]

I just conducted an eyeball analysis of the graph for another thread, but decided it deserved its own thread. Here are my observations:

Those downward segments of the graph between the up-bumps seem to show a decay rate in Shrub's popularity of about 2 points per month.

That should result in about an 8-point loss between now & Nov 2 unless there's another bump (e.g. an October Surprise).

But note that the magnitude of the bumps themselves is declining at a predictable rate. The first (9-11) bump was about 30 points. The second, at the time of the invasion, was about 15. The last, in December 2003, was about 7 points. Each bump is half the magnitude of the last.

If that pattern were to continue, we might predict that the October Surprise would result in only a 3 or 4-point bump, about half of the magnitude of the additional decay we might expect by election day. That seems to mean that Bush will be 8 points further down than today without a bump, and 3 or 4 points bwelow today even with the bump.

[unblock]

1. any october surprise has to be an actual SURPRISE

i think people have already discounted capturing osama. if it happens, people will cheer, but they won't necessarily change their mind about shrub.

a surprise must be something you really don't expect.

frankly, i don't think another terrorist attack here at home would even do it. if kerry has half a brain, he would zoom straight to ground zero number two and make an empassioned speech before shrub shows up. in fact, were i his strategist, i would have already advised him to prepare such a speech for just such an occassion....

[Jackpine Radical]

3. Well, I think you've provided some cogent reasons

why any "Oct. Surprise" will result only in a minuscule bump. The graph kinda suggests that his bump-potential is wearing out.

[Don_G]

2. I'm Not Psychic

And don't pretend to predict the future...

But the Pollkatz graphic charts 15 polls as it happens and the trend is a continual decline over the reliable polls quoted so far.

It can't predict a lot of variables nor "things" like people facing another Vietnam facing a "good" economy with a record defifit...

Why don't we compare notes and save it to another thread?

[BillZBubb]

4. It looks to me like he's got another bump going now.

is that from the Reagasm?

I think he'll get a big bump from the Repug convention. That always seems to happen. Hopefully it will be way too little way too late.

[Jackpine Radical]

6. There may be no bump at all due to the convention,

especially if things get ugly in NY. Anyway, I'm just looking at the trends in the graph, sort of a "technical analysis," and noting that any anticipated bump from any cause is likely to be only 3 or 4 points if it is to remain within the observed pattern.

[tritsofme]

5. I believe his hardcore base is probably 40-45%

and these people will vote for him no matter what he does. Thus I don't think we will see the absolute deterioration of his approval rating that we saw in his father and Jimmy Carter, who both at this point found their approval ratings in the 30% range. I think that due to the polarization of this country along party lines bush will never sink that low. That is why I discourage anyone from telling themselves that this election is over and that bush is toast. He is still in the range where this election could go either way. Work everyday like we're down 10 points in the polls.