Santorum Surge, Part 144
Let’s ignore that three of his four wins (including two last night) have come in non-binding caucuses and take a quick look at The Size of the Santorum Surge.
Over at Skippy, Our Leader posted a link to a discussion of whether “Romney’s strengths” could beat Obama. I wisecracked, without looking at the data, that the total Republican votes in all three states were lower than the daily NYC subway ridership.
My bad. That’s not even a ballpark comparison. On its lowest day, the NYC subway averages more than 2,350,000 riders. (That’s a lot of elitists, Newt.) It averages more than three million (3,000,000) on Saturdays, more than four million (4,000,000) riders a day, and more than five million (5,000,000) every weekday.
By comparison, here are the number of votes cast yesterday in Republican primaries/caucuses:
Clearly, we need a smaller comparative to make the turnout appear more impressive. So let’s look at one of my favorite demographic measures, Metropolitan Statistical Areas (MSAs).
Keeping in mind that the MSA includes the areas around the cities named, the total voters in the three Republican elections yesterday would occupy…the 144th largest MSA,* just behind Eugene-Springfield, Oregon (351,715).
The Santorum Surge: Smaller than Peoria (#135), bigger than Kalamazoo (#148).
*I resist saying “Gross!” at this point.
Point being…?
Point being the Santorum surge is a ripple, not a tsunami.
Consider also, the number of delegates at stake in these events.
Zero.
Cheers!
JzB
Santorum is polling ahead of Romney nationwide, as of yesterday. If what you care about is outcomes, not a horserace within a horserace, then Santorum showing up big in state votes and then showing up ahead of Romney has significance. Santorum is also looking like a win in Tennessee, and maybe in Ohio.
Now, if we are going to talk about numbers, we need to know what the numbers imply. Anybody know what sample size is needed in a poll of this sort – not at all a random draw from the population – in order for there to be any predictive ability? I don’t, but claims that a win in three states is irrelevant – whatever delegate count is involved – is a claim that the sample is too small for the results to be meaningful. I’m willing to be convinced that’s true, but see no reason to believe thats true.
Santorum is polling ahead of Romney nationwide, as of yesterday. If what you care about is outcomes, not a horserace within a horserace, then Santorum showing up big in state votes and then showing up ahead of Romney has significance. Santorum is also looking like a win in Tennessee, and maybe in Ohio.
Now, if we are going to talk about numbers, we need to know what the numbers imply. Anybody know what sample size is needed in a poll of this sort – not at all a random draw from the population – in order for there to be any predictive ability? I don’t, but claims that a win in three states is irrelevant – whatever delegate count is involved – is a claim that the sample is too small for the results to be meaningful. I’m willing to be convinced that’s true, but see no reason to believe thats true.
Santorum is polling ahead of Romney nationwide among Republicans, as of yesterday. If what you care about is outcomes, not a horserace within a horserace, then Santorum showing up big in state votes and then showing up ahead of Romney nationwide has significance. Santorum is also looking like a win in Tennessee, and maybe in Ohio. Dismissing Santorum when he is polling well is a) what Romney wants and b) more or less parroting what “serious people” say on talking heads TV. The race is still underway.
Now, if we are going to talk about numbers, we need to know what the numbers imply. Anybody know what sample size is needed in a poll of this sort – not at all a random draw from the population – in order for there to be any predictive ability? I don’t, but claims that a win in three states is irrelevant – whatever delegate count is involved – is a claim that the sample is too small for the results to be meaningful. I’m willing to be convinced that’s true, but see no reason to believe thats true.