When it comes to predicting percentages based on recent polls, South Carolina is one of the toughest challenges yet. I base my numbers on five polls taken in the past three days, but only two of them took Rick Perry out of the equation. His numbers were on the rise in Gallup tracking polls nationally, but he was well under 5% in South Carolina polls and faced the unpleasant possibility of being compared to the "Herman Cain" (actually Stephen Colbert) comedy campaign.
So now we are down to four. Almost all the recent polls agree on the order: Gingrich, Romney, Paul, Santorum. (Only PPP switches Santorum over Paul.) Based on a weighting of the median and the average of those five surveys, here's my prediction for South Carolina, rounded to the nearest half a percent. Nate Silver's numbers are given in parentheses.
This means Silver is bullish on Gingrich and Other, while I am hoping for strong showings Romney, Paul and Santorum.
Colbert is a wild card, being both a South Carolina favorite son and a TV star, but I am betting the vast majority of conservatives realize his conservative persona is making fun of them. Also, there is the weird public appearance mojo of Friday. Gingrich had to cancel an event for low attendance, while the joint appearance of Colbert and Cain drew about 3,000. Let's assume most of the those at the Colbert rally are not likely primary voters.
Also helping the Other votes is Perry dropping out so late. There could be a fair number of mail-in ballots with his name on them. Still, I'm not going to repeat my late fascination with Buddy Roemer in New Hampshire based on very little hard data.
As dissatisfied as the polls show the likely voters are with this field - the median count in the five polls for NoneOfTheAbove was 8% - I still think when South Carolina voters step inside those booths today, they will cast their votes for one of the four guys who showed up at the last debate. Polls have shown voters are paying close attention to the debates. As a person who believes in democracy, I find that encouraging and discouraging at the same time.
These are my final numbers. I updated Silver's numbers from his noon Eastern time prediction on Saturday. I will report back when 100% of precincts report in, which will likely be Sunday morning.