26 thoughts on “South Carolina (D) Thread and Predictions

  1. Clinton: 61%
    Sanders: 39%

    While unlike the Nevada Caucuses, we do have substantial polling this year and month, I could only find two polls that occurred since the Nevada Caucus, and one was a Clemson poll who’s results I find dubious to be representative (though I am no way insinuating fraud. I just think that they somehow culled an unrepresentative group).

    That being said, I have no doubt of a Clinton win, and a large one at that. I haven’t seen a poll this year having a Clinton lead of under 18, and recently her polling has been more like low to mid 20’s. So the question is: what will be her margin of victory?

    The good news for Clinton is that the two polls show her at 60% (Emerson) and 64% (Clemson). That’s a good sign that she’s likely to top 60%. The bad news is that her higher number is the Clemson poll, which has Sanders at 14% and 22% undecided. And… Those are numbers I can’t really believe. I don’t believe a) that there are 22% undecided voters and b) Sanders is at 14%. That would suggest that even if undecided voters broke 2 to 1 in Sanders favour, Clinton would just break 70% of the vote and Sanders would only end up at slightly under 30%. And while I expect a big Clinton win, nothing says she’s going to have that good a day. (I know I’m discounting others votes, but since I don’t expect that to top 1%, it still doesn’t change how ridiculous the situation looks).

    So while I cannot discount the possibility that Clinton does even better than the Emerson poll suggests, I decided to hedge my bets by giving it a bit more weight on my analysis to predict 61% of the vote.

  2. And @Aaron: this is the second time you’ve come in before me and basically used the numbers I was thinking about using. How do you do that?

  3. @Rophil, While I agree that Clinton is likely to be the nominee (and probably will rack up wins on Super Tuesday’s part 1, 2, and 3 which will cement that status), if she only wins South Carolina by 10 points, well that’s worrying for her. She’s projected to win by at least 15, probably around 20 points. So if she wins by only 10, well that’ll fuel fires of whether this is a repeat of 2008. Now, of course it’s better than losing, but a win by only 10 points weakens the claim that Clinton will win, not strengthens it.

  4. Yeah. I mean I just ran through the numbers with Sanders winning 2/1. But this suggests a more 1 to 1 split (which seems more reasonable for how undecided voters will vote in a two person race). This is a result I would have expected five months ago, not now.

  5. Yeah. Shame on me. (With the caveat that I can still safely discount polls with severe methodological errors, since their accuracy will be based on random chance).

  6. I am also very surprised by the margin of victory. Now Clinton has easily won her third state, doing better than expected by the polls, she looks unstoppable. Sanders should be worried.

Leave a Comment