# Trump could win the election in a nowcast by FiveThirtyEight **URL:** \ **Category:** Politics & Elections\ **Created:** [July 25, 2016, 1:36pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846 "2016-07-25T13:36:00Z")\ **Posts on this page:** 20\ **Page:** 45
**Author:** ![Richard\_Parker](https://avatars.discourse-cdn.com/v4/letter/r/35a633/32.png) [@Richard\_Parker](https://boards.straightdope.com/u/Richard_Parker)\ **Post date:** [September 26, 2016, 8:31pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/882 "2016-09-26T20:31:02Z")
> [@elucidator](#): > > Who can gauge the changes in the Apathy Party? The people who either don’t vote or don’t really think about it until crunch time, and maybe not even then? Don’t the polls mostly target people who vote, as a regular thing? “Likely voters”, “registered voters”, so on and so forth? Adults who are not registered, and registered voters who are not likely to vote, both tend to lean toward Democrats. But you have hundreds of pollsters using many different methodologies to determine who will actually vote, and collectively they’re pretty darn good at it. Plus, unregistered and unlikely voters are disproportionately found in noncompetitive states. So it’s pretty unlikely things will be any different this election with respect to the slumbering giant that is the Apathy Party. ---
**Author:** ![Corry\_El](https://avatars.discourse-cdn.com/v4/letter/c/ecae2f/32.png) [@Corry\_El](https://boards.straightdope.com/u/Corry_El)\ **Post date:** [September 26, 2016, 8:34pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/883 "2016-09-26T20:34:21Z")
> [@Happy\_Lendervedder](#): > > Bill didn’t campaign for Gore. I said “Obama and Biden are good surrogates for Clinton to some audiences. However as to the general tone of the election and current tie, the reference to Bill Clinton 2000 shows the lack of any paradox in Obama’s approval rating v (Hillary) Clinton’s struggles. (Bill) Clinton’s approval avg over first half of 2000 was higher than Obama’s first half of this year, but Gore struggled.” As a couple of times previously, I’m struggling to see the relevance of that response. I know Bill Clinton didn’t campaign a lot for Gore, though it’s not true that he didn’t at all. This was discussed in the recent thread ‘Is Obama’s stumping for Clinton unprecedented?’ > **[ABCNEWS.com : Clinton to Stump (a Little) for Gore](http://web.stanford.edu/class/polisci179/ABCNEWS_com%20%20Clinton%20to%20Stump%20(a%20Little)%20for%20Gore.htm)** > > \- President Clinton will finally hit the campaign trail next week on behalf of Vice President Al Gore - but is being steered away from the swing states that are likely to decide this year's election. But I don’t suppose you’re actually saying you know that Gore would have had smooth sailing if he’d just chosen to use Bill Clinton more (I hope you don’t claim such an ability 🙂 ). As that article said at the time, Bill Clinton was viewed as radioactive with a lot of voters despite the reasonably high overall approval. That could be disputed, or the same comment made about Obama, or that one disputed too. The actual point is the lack of much reason to think a second term president’s slightly above 50% approval means a lot in favor of a would be successor from the same party in the big picture. ---
**Author:** ![asahi](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/asahi/32/8693_2.png) [@asahi](https://boards.straightdope.com/u/asahi)\ **Post date:** [September 26, 2016, 9:13pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/884 "2016-09-26T21:13:26Z")
> [@Lantern](#): > > Yeah some scary polls out today. If you look at RCP their no-tossup count gives it to Hillary 272-266 and that includes Colorado where she is up by 0.2%. In the last couple of weeks her “blue wall” has crumbled into dust. > > I would still give her the edge because I think her superior ground operation is worth a point or two and I am still counting on Trump to make big mistakes especially during the debates. Hillary also has a massive advantage from popular surrogates like Obama, Bill and Biden and you can bet that they will step up their campaigning in the final weeks. > > The first debate couldn’t be set up more nicely; it’s truly going to be an epic moment in US political history. I think her superior ground game is a real advantage if Clinton is holding on to slim advantages across several states. However, if Trump starts to pull away in Florida, Ohio, and perhaps another state like Colorado, the ground game is there just to keep her competitive. But at that point, the writing would in a sense already be on the wall. Hillary Clinton is (pardon my bias) an inherently superior candidate to Trump, but if she’s still struggling to make that case to voters in early November, then her ground game probably won’t matter. Clinton needs to have a strong performance in this debate in the eyes of voters, which might be different than how we at SDMB might view a strong performance. Whatever these voters want, Hillary needs to give it to 'em tonight. She has to stop the bleeding now, or people will start accepting and possibly even getting comfortable with the idea of President Trump – which we absolutely don’t want. ---
**Author:** ![skdo23](https://avatars.discourse-cdn.com/v4/letter/s/a5b964/32.png) [@skdo23](https://boards.straightdope.com/u/skdo23)\ **Post date:** [September 26, 2016, 9:14pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/885 "2016-09-26T21:14:08Z")
I haven’t noticed much interest in Sam Wang on the SDMB as of late, but HRC just made some encouraging gains in his predictions in [his most recent update](http://election.princeton.edu/). Predicting 307 EV, Meta-margin at 2.4%, up from 1.5%, and roughly 10% gains in both random drift and Bayesian win probabilities (currently 79 and 87%, respectively). His newest article, _Three Reasons To Ignore Debate “Expectations,”_ currently featured on the linked page, hopefully has some truth to it. ---
**Author:** ![CarnalK](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/carnalk/32/486_2.png) [@CarnalK](https://boards.straightdope.com/u/CarnalK)\ **Post date:** [September 26, 2016, 9:17pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/886 "2016-09-26T21:17:31Z")
> [@iiandyiiii](#): > > I’m pretty sure those Ipsos state polls are a subset of their national poll, and not corrected for things like demographics, which means they are just barely better than useless. A new thing (I think) this year is the big pollsters also releasing state subset numbers when they do a national poll, but the samples are too small for many states to correct for demographics and other things. Do you have a cite for this? It would be pretty unethical for them to not put massive caveats/MOE for these sub-polls and for other people to report same without mentioning those caveats. ---
**Author:** ![suranyi](https://avatars.discourse-cdn.com/v4/letter/s/e36b37/32.png) [@suranyi](https://boards.straightdope.com/u/suranyi)\ **Post date:** [September 26, 2016, 9:35pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/887 "2016-09-26T21:35:59Z")
> [@skdo23](#): > > I haven’t noticed much interest in Sam Wang on the SDMB as of late, but HRC just made some encouraging gains in his predictions in [his most recent update](http://election.princeton.edu/). Predicting 307 EV, Meta-margin at 2.4%, up from 1.5%, and roughly 10% gains in both random drift and Bayesian win probabilities (currently 79 and 87%, respectively). > > His newest article, _Three Reasons To Ignore Debate “Expectations,”_ currently featured on the linked page, hopefully has some truth to it. In fact Sam Wang had a comment in that article in which he said, essentially, that Nate Silver’s current figures make no sense. ---
**Author:** ![asahi](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/asahi/32/8693_2.png) [@asahi](https://boards.straightdope.com/u/asahi)\ **Post date:** [September 26, 2016, 9:36pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/888 "2016-09-26T21:36:01Z")
> [@marshmallow](#): > > We’re one terrorist attack or dank meme away from living in some sort of Upton Sinclair fanfiction. I’ll blame the Dems for putting up the most establishment candidate ever in an anti-establishment year. Heck of a job. Pfff…frankly this kind of thinking is exactly why we have the problems we now have in this election. If people would just fucking vote based on who’s qualified instead of whoever is capturing our imagination we wouldn’t have to worry about Trump. Bernie Sanders, however kewl and hip he might have been, would not have been the right choice. He might have defeated Trump but he would end up being nothing but a gadfly as president just like he’s been a gadfly as a senator. This country doesn’t need a gadfly anymore than it needs a reality TV buffoon. What it really needs, what it really lacks, is a population of people who are demonstrably competent at voting. ---
**Author:** ![asahi](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/asahi/32/8693_2.png) [@asahi](https://boards.straightdope.com/u/asahi)\ **Post date:** [September 26, 2016, 9:37pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/889 "2016-09-26T21:37:38Z")
> [@suranyi](#): > > In fact Sam Wang had a comment in that article in which he said, essentially, that Nate Silver’s current figures make no sense. Ooooooohhhhhh, looks like the stat geeks are talkin trash and getting ready for an online bro-down. ---
**Author:** ![skdo23](https://avatars.discourse-cdn.com/v4/letter/s/a5b964/32.png) [@skdo23](https://boards.straightdope.com/u/skdo23)\ **Post date:** [September 26, 2016, 9:45pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/890 "2016-09-26T21:45:33Z")
> [@asahi](#): > > Ooooooohhhhhh, looks like the stat geeks are talkin trash and getting ready for an online bro-down. Yeah, and with all due respect to Nate Silver, I really hope that Wang is right. ---
**Author:** ![iiandyiiii](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/iiandyiiii/32/7924_2.png) [@iiandyiiii](https://boards.straightdope.com/u/iiandyiiii)\ **Post date:** [September 26, 2016, 9:51pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/891 "2016-09-26T21:51:42Z")
> [@CarnalK](#): > > Do you have a cite for this? It would be pretty unethical for them to not put massive caveats/MOE for these sub-polls and for other people to report same without mentioning those caveats. It’s from my memory of a 538 article (by Nate, I think) explaining all those new state “polls” with small sample sizes from the last few months. I did a 2 minute search and couldn’t find it – I’ll try again later. Maybe your google skills will be better, if you feel like looking. ---
**Author:** ![asahi](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/asahi/32/8693_2.png) [@asahi](https://boards.straightdope.com/u/asahi)\ **Post date:** [September 26, 2016, 10:01pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/892 "2016-09-26T22:01:07Z")
> [@skdo23](#): > > Yeah, and with all due respect to Nate Silver, I really hope that Wang is right. I don’t think anyone truly knows at the moment what the real probability for this race is – too much polling and not enough of it settled. But honestly, given the recent trend line, Wang’s probability seems a bit overconfident. I’m assuming I’m reading his site correctly, which I admittedly might not be. I think Clinton has a very slight electoral college advantage, but even with the more optimistic interpretation of polling, the trends probably suggest that Trump has enjoyed a good 6-week run of success and that, at best, Clinton has probably only managed to slow his momentum somewhat. Worst case scenario? Trump’s actually even or winning. ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 26, 2016, 10:16pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/893 "2016-09-26T22:16:22Z")
[Electoral-vote.com](http://Electoral-vote.com) is making the argument for “gut” analysis of the race. That Trump has a ceiling on his support, which is why he can’t win. Except that’s what I assumed in the primary. Plus Gary Johnson is an outlet for any voter who doesn’t like either candidate, so Trump’s ceiling is irrelevant. For what it’s worth, my gut tells me [electoral-vote.com](http://electoral-vote.com) is right. Clinton probably wins most Stein voters on election day. But that’s a skewed polls argument and you know how good I am with those. ---
**Author:** ![Euphonious\_Polemic](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/euphonious_polemic/32/3002_2.png) [@Euphonious\_Polemic](https://boards.straightdope.com/u/Euphonious_Polemic)\ **Post date:** [September 26, 2016, 10:35pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/894 "2016-09-26T22:35:33Z")
> [@adaher](#): > > But that’s a skewed polls argument and you know how good I am with those. I have to say, I like the fact that you stuck around after the last election and took your lumps, and that you can poke fun at yourself. ---
**Author:** ![iiandyiiii](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/iiandyiiii/32/7924_2.png) [@iiandyiiii](https://boards.straightdope.com/u/iiandyiiii)\ **Post date:** [September 26, 2016, 10:39pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/895 "2016-09-26T22:39:26Z")
One thing to think about – Hillary Clinton has the same data team as Obama in 2012, and Obama’s 2012 data team and internal polling were [much, much steadier](http://www.bloomberg.com/news/articles/2013-05-31/obamas-data-team-totally-schooled-gallup) and more accurate than the national polls. So we can keep an eye on which states she’s focusing on – and IIRC, she’s not putting money into CO, PA, WI, and MI, which might be a really good indication that her internal polling shows that those states are in the bag. ---
**Author:** ![DSeid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dseid/32/20194_2.png) [@DSeid](https://boards.straightdope.com/u/DSeid)\ **Post date:** [September 26, 2016, 10:53pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/896 "2016-09-26T22:53:42Z")
> [@CarnalK](#): > > Do you have a cite for this? It would be pretty unethical for them to not put massive caveats/MOE for these sub-polls and for other people to report same without mentioning those caveats. Wang alludes to it [here](http://election.princeton.edu/2016/09/20/ipsosreuters-rips-off-the-band-aid/). Not so good for individual states but in aggregate very useful. > [@](#): > > … For us, the large error bars for the Ipsos polls are acceptable because we aggregate them into a single snapshot, which effectively reduces the overall error. For example, if each poll error is +/-10% and 20 states are competitive enough for the polls to be useful, then the aggregated error is +/-10/sqrt(20), or +/-2.2%. Think of each poll as being a mystery body part; what PEC provides is the head cheese. … From comments in that thread we can link to [this discussion](http://www.electoral-vote.com/evp2016/Pres/Maps/Sep21.html#item-11) on [electoralvote.com](http://electoralvote.com): > [@](#): > > We ran an analysis yesterday as follows: We first removed all the Ipsos polls and then ran the software to calculate the map. Then we removed all the non-Ipsos polls and did it again. Ideally, they should be quite similar, indicating that Ipsos is getting more-or-less the same results as the other pollsters. That’s not what we found. In nine of the 44 states Ipsos has polled, the Ipsos and non-Ipsos results differ by 10 or more points. That is way outside the margin of error (usually about 4%). In 21 states, the difference between Ipsos and non-Ipsos was 5 points or more. Based on this analysis, we have decided to retroactively remove all the Ipsos polls from the database and not use its Internet polls going forward. > > Interestingly enough, Ipsos is not biased. Averaged across all states, Ipsos favors the Democrats by a very small margin: Only 0.70%. That is fine. Except that it gets this small margin by having large errors cancel out. For example, Ipsos has Arkansas 18 points less Republican than the others, but has Texas 15 points more Republican than the others. Which seems consistent with Wang’s take. _Individual state results are garbage_ but for aggregation purposes they can have a place. > [@skdo23](#): > > I haven’t noticed much interest in Sam Wang on the SDMB as of late, but HRC just made some encouraging gains in his predictions in [his most recent update](http://election.princeton.edu/). Predicting 307 EV, Meta-margin at 2.4%, up from 1.5%, and roughly 10% gains in both random drift and Bayesian win probabilities (currently 79 and 87%, respectively). > > His newest article, _Three Reasons To Ignore Debate “Expectations,”_ currently featured on the linked page, hopefully has some truth to it. What I thought was most cogent was this observation: > [@](#): > > Via Will Jordan and CNN/Gallup, here is survey data on who was expected to do better in past debates. … Notice a pattern: all of these candidates ended up winning the popular vote. The explanation is simple: people can tell who the stronger candidate is, and expect that person to do well. As you can see, Hillary Clinton is narrowly expected to perform better tonight. > > Now, think about this for a moment. What does it mean for a candidate to “exceed expectations”? It usually means that he/she was the weaker candidate to begin with. So if some commentator emits hot gas about either candidate doing better than expected, there is a good chance that candidate will be the loser in November. ---
**Author:** ![elucidator](https://avatars.discourse-cdn.com/v4/letter/e/8edcca/32.png) [@elucidator](https://boards.straightdope.com/u/elucidator)\ **Post date:** [September 26, 2016, 10:55pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/897 "2016-09-26T22:55:54Z")
nm ---
**Author:** ![Euphonious\_Polemic](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/euphonious_polemic/32/3002_2.png) [@Euphonious\_Polemic](https://boards.straightdope.com/u/Euphonious_Polemic)\ **Post date:** [September 26, 2016, 10:57pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/898 "2016-09-26T22:57:12Z")
> [@iiandyiiii](#): > > One thing to think about – Hillary Clinton has the same data team as Obama in 2012, and Obama’s 2012 data team and internal polling were [much, much steadier](http://www.bloomberg.com/news/articles/2013-05-31/obamas-data-team-totally-schooled-gallup) and more accurate than the national polls. So we can keep an eye on which states she’s focusing on – and IIRC, she’s not putting money into CO, PA, WI, and MI, which might be a really good indication that her internal polling shows that those states are in the bag. That is quite an interesting article. I wonder what sort of data team Trump has assembled? Of course, he has only the VERY BEST people. Apparently Cambridge Analytica has the job - some people are saying that other, more exprienced firms did not want the job at all. Cambridge Analytica’s claim to fame is that they worked on Cruz’s failed campaign, so they have that going for them. ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 26, 2016, 11:31pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/899 "2016-09-26T23:31:52Z")
Well, one thing that can be said is that if she loses, it won’t be because her campaign failed her, but because she failed her campaign. ---
**Author:** ![asahi](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/asahi/32/8693_2.png) [@asahi](https://boards.straightdope.com/u/asahi)\ **Post date:** [September 26, 2016, 11:33pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/900 "2016-09-26T23:33:36Z")
Good campaign teams will typically have better polling than campaigns – getting accurate data and knowing, down to the precinct, which voters are enthusiastic about their candidate and which aren’t is their business. So I would have to agree that even looking at RCP averages may tell us a lot, but the campaigns probably have better data – or at least the better campaign does typically. ---
**Author:** ![asahi](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/asahi/32/8693_2.png) [@asahi](https://boards.straightdope.com/u/asahi)\ **Post date:** [September 26, 2016, 11:42pm UTC](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846/901 "2016-09-26T23:42:14Z")
> **[Reuters/Ipsos States of the Nation: Clinton leads in Florida](https://www.reuters.com/article/us-usa-election-poll-idUSKCN11W242)** > > Democrat Hillary Clinton has pulled ahead of Republican rival Donald Trump in the traditional battleground state of Florida, strengthening her position in the race for the White House, the Reuters/Ipsos States of the Nation project showed ahead of... I think there are two polls out now showing that Florida has flipped back to Clinton. I’ve been wondering whether or not Trump’s foundation problems might hurt him - and if there was anywhere that it might, Florida would be the likely place. Hard to know if this is just a blip or something deeper, but it’s a good sign for Clinton. Hopefully she can build on this by having a strong debate and not letting other states slip. But even so, if she can win in FL, that would be pretty close to a mortal blow to Trump’s chances. He’d really have to get out the vote across a lot of geography and that’s not something his campaign is in a position to do. [Previous page](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846.md?page=44) [Next page](https://boards.straightdope.com/t/trump-could-win-the-election-in-a-nowcast-by-fivethirtyeight/760846.md?page=46)