This is good advice and the media shouldn't proclaim Bloomberg's campaign dead (or anything close) without seeing the evidence.
But also it wasn't clear how alive Bloomberg's campaign was in the first place and the media may have jumped the gun in proclaiming him a frontrunner. twitter.com/EsotericCD/sta…
Sanders didn't get hit too hard, but Warren having a good night isn't a great development for him. twitter.com/NickRiccardi/s…
Prediction markets are sometimes dumb but don't remember someone crashing this much over the course of a debate (well, 45 minutes of a debate). electionbettingodds.com/4hr.html
Don't get TOO excited but I've got an analysis of Bloomberg's current position in the campaign up on the liveblog. fivethirtyeight.com/live-blog/demo…
Our model has grown much, much less skeptical of Bloomberg, but mostly because he's in *striking distance* of having a really big Super Tuesday. Get up to the mid-to-high-20s and he could do some real good for himself. High teens or even low 20s, not exactly.
Other thing to keep in mind re: Bloomberg is that if he just gets what he's getting in the polls *right now* that's…not actually great for him. It tends to lead either to a Sanders nomination, a contested convention, or leaves room someone else (JB/PB/EW/AK) having a big night.
this is Salisbury OF Justin Meekins, one of the best hitters in D-III baseball
the progression of his roster photos from freshman year to now is like the galaxy brain meme but for facial hair and glasses
Bloomberg has certainly beaten my expectations so far and it's possible that he'll be at 25% nationally by Super Tuesday. But he could also decline as a result of a bad debate or say Biden gaining momentum by winning SC. Media-driven poll surges and declines are not uncommon.
Ehh, he seems to have gotten a little stuck in the polls at ~16% and this is happening at a time when the opposition research is just starting to escalate and the tone of media coverage has shifted from surprise to skepticism. twitter.com/asymmetricinfo…
Well, I'd say it's more than that. The point of a long primary campaign is that you have to survive lots of tests. Debates, opposition research/rough news cycles, polling slumps—and then perform respectably well in the first 4 states. Bloomberg hasn't done almost any of that yet. twitter.com/asymmetricinfo…
Interestingly enough, the Bloomberg campaign seems to be acknowledging this. I kind of doubt that it will persuade other candidates to leave the race though. axios.com/bernie-sanders…
Changes in our nat'l polling average since 2/13 (when we started to get the first hints of post-NH data). Basically everything is stagnant EXCEPT Sanders may be gaining further.
My sincere apologies for not catching this sooner. Bugs can happen but I usually have a good eye for when something doesn't look right. Thought the changes were just the model reacting to IA; didn't realize I'd also introduced a problem with the demographic regressions. Shoot. /7
The fact that you can pick up district delegates even in states where you don't get 15% statewide makes it slightly harder for candidates to win by runaway delegate margins. So reintroducing the proper amount of district-by-district variance tends to help "no majority". 6/
The district-by-district forecasts were also underdispersed. That hurt candidates who projected to get <15% statewide, but who were potentially strong in some districts and thereby could expect to pick up some district delegates. 5/
So, for example, the model now has Biden doing better than before in South Carolina and Alabama, but Sanders doing better than he was before in Nevada and California. 4/
Without demographics the model basically defaulted to a geographic prior, which tends to underrate Biden/Bloomberg in the South and overrate them (especially Biden) outside the South. It also tended to overrate Sanders in Southern states and underrate him in Western states. 3/
Specifically, the demographic regressions we use as a compliment to the polls weren't calculating properly since some code changes we introduced on 2/5 (after IA). As a result, the state forecasts were underdispersed (too conservative) especially in states with little polling. 2/
Welp this is annoying, but we found an issue with how our model was making state-by-state forecasts. The error is fixed. The topline effect is not very large and mainly helps "no majority" slightly. A few state forecasts have changed more noticeably. 1/
Obviously Bloomberg could see his #'s keep growing, or they could shrink, but the scenario where he stalls out at ~16%, enough to make Biden's fall much worse and to blunt momentum for Buttigieg, but without coming particularly close to Sanders, is a really good one for Bernie.
State polls were mostly pretty good for Bloomberg but national polls were meh.
Obviously Biden has lost a lot of ground since Iowa but he may have found a bottom (for the time being) at ~16% nationally.
Welp, there were a whole LOT of polls today and the upshot was basically:
* Mostly good news for Sanders
* Mixed news for Bloomberg and Biden
* Not much sign of a major surge for Buttigieg or Klobuchar
* Warren could use some kind of momentum swing
I should say: it's unlikely we get a scenario *exactly* like this. More often, 1-2 of the candidates below Sanders will have dropped out or faded in the polls to the point of oblivion. But this general family of scenarios, with Berrnie having a clear plurality, is fairly common. twitter.com/davidplouffe/s…
What it illustrates, though: We're *not* in a 2-candidate race *right now*. We're in a race that has 1 clear but not overwhelming frontrunner, Sanders, and a lot of candidates chasing him, of whom Bloomberg may or may not be the strongest.
It is important to note that this this reflects an average of many different scenarios and the ranges are very wide.
Average projected delegates through Super Tuesday:
Sanders 608 (41% of delegates thru March 3)
Bloomberg 273 (18%)
Biden 270 (18%)
Buttigieg 157 (10%)
Warren 127 (8%)
Klobuchar 55 (4%)
Sanders now with a 9-point lead (over Bloomberg/Biden, who are essentially tied for 2nd) in our national poll average. projects.fivethirtyeight.com/polls/presiden…
Shifts in Democratic primary polls don't seem to be correlated with shifts in general election polls, i.e. Biden's numbers vs. Trump are holding up fine. twitter.com/mmurraypolitic…
, there are currently 8 stories on the NYT politics page about Bloomberg, and 0 about Buttigieg who (i) is actually on the ballot in the next two states and (ii) is in first place in pledged delegates so far.
FYI, this is a Steyer internal poll in Nevada. We list internal polls on our polls page but they do not figure into our averages/model. (FWIW, we have tried to incorporate internal polls into our general election forecasts but the primaries are trickier.) projects.fivethirtyeight.com/polls/presiden…
Our surveys showed that as impeachment wore on, rank-and-file Republicans became more supportive of President Trump and of Rs more broadly. 53eig.ht/2uNsF50
At the end of our 4-month panel survey, @ameliatd
interviewed some of the participants to get a better sense of how they felt. The answer?
"We’re divided. We’re stuck. I knew that before impeachment, but it’s even clearer to me now."
Meanwhile, I've got a long story up this morning on the evidence for and against Sanders having a "ceiling", and why it might not matter anyway. fivethirtyeight.com/features/sande…
With a couple of strong polls last night and this morning, Sanders has re-overtaken "no majority" as the Democratic frontrunner. projects.fivethirtyeight.com/2020-primary-f…
Seems like attempting to punish the players, knowing you'll probably lose the grievance, could still have been politically wise? At the very least, other players around the league (not wanting to undercut the MLBPA) might have been more reluctant to speak out against the Astros. twitter.com/EvanDrellich/s…
I kinda wonder if Bloomberg, by escalating fairly minor sh*t with Sanders, is—in addition to distracting from negative storylines about himself—also creating a pretext for dumping a whole lot of negative ads on Sanders later on.
Klob and Pete have plenty of differences. But they have enough overlap in terms of who tends to support them, which states they perform well in, and how they figure into the media narrative about the race that you wonder what things might look like if there were 1 of them, not 2.
Bloomberg is comfortably above 10% in the national poll average, and Marist polls tend to come in pretty close to the averages, so it would be a pretty big upset if he cannot get to 10% in the poll tomorrow AM. twitter.com/TrackerDebate/…
Our Nevada polling average after adding a couple of polls today that showed very... different results from one another.
Now, it's true that Bloomberg and Biden are only barely above 15% in many states. On the other hand, Buttigieg and Warren are only barely below 15% in many states. Even with the field as scrambled as it is, it's likely that most states have ~3 or 3.5 delegate qualifiers.
Super Tuesday states where we currently have candidates projected to get at least 15% of the vote:
Sanders: All Super Tuesday states
Bloomberg: All states but VT
Biden: All states but VT, UT, MN
Buttigieg: OK, AR only
Warren: MA only
Klobuchar: MN only
The post-NH polls, though a weird/eclectic bunch, don't seem to show many bounces from NH. Bernie hasn't risen much further. Biden, following a big decline after IA, hasn't fallen much further. Buttigieg and Klobuchar may be gaining, but it's a fairly slow pace.
Now, if you've done the research and you have some empirical reason to expect the trend to *continue*, maybe you take a different view. But there are also lots of time series that are mean-reverting, i.e. where a value that drops from 25 to 17 is likely to bounce back to 25.
Biden probably does have an easier path and perhaps a much easier path. People overrate the importance of *trends* and underrate the importance of *levels*. Being at 17% (down from 25%) is generally still much much better than being at 7% (up from 4%). twitter.com/jimgeraghty/st…
I still think a rejuvenated version of Biden would be Bernie's toughest matchup. A lot of working-class support and a lot of African-American support. twitter.com/daveweigel/sta…
Can't tell which of these takes I like better:
1) The polling in Nevada is screwy, so one should default toward one's priors that Bernie is a heavy favorite
2) The polling in Nevada is screwy, but the Nevada caucus is also screwy so one shouldn't make too many assumptions at all
I did something similar a loooong time ago: Preseason World Series odds predicted attendance better than actual wins. baseballot.blogspot.com/2011/12/wins-d… twitter.com/natesilver538/…
I need to find the old study I did that found that merely singing free agents was associated with higher ticket revenues, holding actual wins constant. Fans reward teams that seem to be making an effort to be competitive and punish the ones that aren't. twitter.com/jareddiamond/s…