Voter to Donor ratio thus far - about 10:1 and predictions
Please check out my other post for a technical explanation of how I came to these numbers: http://www.dailypaul.com/node/20531
Iowa, with an average rate of donation had a voter to donor ratio of about 8:1. New Hampshire, with a high rate of donations had a voter to donor ratio of about 11:1. Michigan, with an average rate of donation had a voter to donor ratio of 11:1.
I think having three examples, it's useful to make predictions for some of the upcoming primaries, provided that nothing drastic happens.
First off, South Carolina. If we apply the 10:1 voter to donor ratio to the estimated 2100 SC donors, we get a vote total of 21,000, which, if you look at the 2000 GOP primaries (see the above link), would be good enough for ~4%.
Nevada, on the other hand, is VERY encouraging. Using the 10:1 ratio and current number of donors,, there should be 25,000 votes for Ron Paul. Unfortunately, I don't have the numbers for past Nevada results (if you have any please link me!) but I think 25k votes is very competitive for NV.
Finally, Florida: With 10:1 ratio, we can expect about 107,000 votes, which in 2000 would have been good enough for 15%. I find this to be very encouraging.
Your Donors Metric correlates with my meetup metric
I've been doing the same thing but with meetup members as my metric
http://www.dailypaul.com/node/25914
I came up with about the same number of projected votes. The uncertain thing is what the turnout will be. In a Nevada article there was a projection of 40,000-60,000. At first this seemed very low. But then let's take a look a Iowa. The population of Iowa is around 3 million. The population of Nevada is only 2.5 million. Iowa had about 120,000 for their caucus turnout. I really doubt Nevada will have as big of a turnout as Iowa. I wouldn't think it would be a stretch to say that Nevada would have only half of the turnout. That would be only 60,000 which would make your calculations turn into a huge victory as far as percentages go.
Right now my meetup metric is projecting about 75,000 votes in Florida which would be about 11%.
neat
I remember reading that the 2000 Nevada turnout was exceedingly low (although I don't have numbers). I don't know if this necessarily explains why the other candidates have been so dismissive of the state. In fact, I would expect just the opposite.
Will caucus format affect prediction?
Do you think the caucus format will affect the prediction? I tend to think the caucus hurts Ron b/c, if its like IA, he won't get many second choice picks (except maybe Thompson first-choicers). Otherwise, I think that's an interesting model you've applied.
maybe
It's hard to tell because the only test of the caucus format was IA (with the exception of Wyoming, but that doesn't count). However, Iowa's ratio was lower than the primary states of NH and MI. But I don't think Iowa's format directly resulted in a lower ratio of 8:1 rather than the 11:1 of MI and NH. If we would have applied 11:1 to IA, Ron Paul would have received 15% or over 18,000 votes, 5,000 more than McCain or Thompson and good enough for a strong 3rd.
I hope you're right.
This is very encouraging!
Interesting...
... thank you for this post. Interesting take on things.
Just to think that we needed 6 times of the final vote count to come first in Michigan.