“We can tell which party will control things for the next few years by the fact that Democrats are losing by a lot less than they usually do!”
Should actually say: “We can tell which party will control things for the next few years by the fact that Democrats are losing, in Republican districts, by a lot less than they usually do!”
Which means that when you get to districts that are more competitive, “losing by a lot less that usual” could actually mean winning.
For some crazy reason, I’m not going to get my hopes too high based on 538! :rolleyes:
And if wishes were horses, beggars would ride.
538 did a pretty great job with the 2016 election – they predicted Trump had a significant chance to win (around 35%, IIRC). Things that are 35% likely happen quite frequently (7 times in 20, in fact). That’s what the polls showed, and the polls were very close to dead on nationally, and close to dead on in most (but not all) states.
The open question is whether or not trump voter will maintain both their enthusiasm and allegiance to the degree necessary to keep most of the Republican seats in that column. Especially in the face of what is likely to be a highly motivated opposition (that’s putting it mildly).
The history of “movement politicians” (arguably including Obama) is not promising from their perspective.
For those hoping for some actual information in this thread, the election will be on June 20. This is a runoff from a “jungle primary” where Ossoff, the Democratic candidate, took 48.1% of the vote, while the leading Republican, Karen Handel, took 19.8%. 48.1% is of course very close to 50%, but that’s not as significant as it looks, since it mostly just means that almost all of the Democrats voted for Ossoff, while the Republicans had a more crowded field. It does at least suggest that this election is highly unlikely to be a blowout for the Republicans, as the PVI of R+8 might suggest.
Cite? 
So Republicans are looking for Handel to be their Messiah?
Given the actual results, how is the prediction of 35% a “pretty great job”? :dubious:
A lot of the other aggregators had HRC’s percentage chances at over 90% - I know Sam Wang over at Princeton had her at 99%+. 538 still had him as the underdog, but also had him with much higher chances in some of the states that he did end up winning, like Pennsylvania, and also had his chances of winning consistently higher than other models. After all, 35% isn’t great, but you shouldn’t be that shocked if something with a 1-in-3 chance happens. They were also pretty close in the prediction of the popular vote total, even if they didn’t fully account for the vagaries of the electoral college.
Egotist te absolvo. Go, and sin somewhat less.
It wasn’t a prediction of 35%, it was saying Trump had about a 35% chance to win. Better than 1 in 3. And their prediction of the national popuolar vote margin was pretty much on the nose, and possibly the best in the country.
Perhaps you have extraordinary abilities and predicted this election perfectly – if so, you probably have a wonderful career that you’re missing.
If you are referring to the Ayn Rand on Ice roadshow, you may be pleased to know that investment opportunities are still open.
And he won, anyway. I still don’t understand why you think that 538 did a “pretty great job”.
Because they were very, very close in the national popular vote and all but a very few states.
As we all learned by the time we got to junior high, the popular vote is irrelevant in Presidential elections.
It’s not, however, irrelevant in assessing predictions about the overall results of a presidential election (which include individual state results, national popular vote results, and the electoral college), which is what I thought we were discussing.
Ok, great, they accurately predicted the national popular vote. What practical effect does that have?
It demonstrates that they have a good (great, I’d say, taking '08 and '12 into account) track record of predictions, and therefore I think it’s reasonable to seriously consider their analyses of other things related to elections.
Oy, get a room you two. 