LA Times Poll (July 30th) - Trump Leads Clinton 46.1 to 41.9

[TomCADem]

This strikes me as a weird poll given that it seems to have Trump leading since July 10th with his lead growing even as he has gone more and more crazy. Perhaps the key really is just to such all the oxygen out of the room by engaging in racist, barbaric behavior.

http://96.127.53.23/election/

Here is the LA Times' story from July 25th discussing their poll results back then. The updated results from the link above do not show any change from the DNC convention.

http://www.latimes.com/politics/la-na-pol-convention-bounce-20160725-snap-htmlstory.html

A Republican convention that got off to a rocky start and endured several distractions nonetheless generated a modest increase in Donald Trump’s poll standing, moving the New York businessman back into a lead over Hillary Clinton.

Through Sunday, the USC Dornsife/Los Angeles Times Daybreak tracking poll of the race shows Trump gaining about 3 percentage points in the aftermath of the convention. That would be roughly in line with the convention bounces enjoyed by Democratic and Republican nominees in the last three election cycles.

As of Sunday, the poll, which is updated daily, showed Trump leading Clinton 45% to 41%. The lead is within the poll’s margin of error of 3 percentage points in either direction, meaning that the apparent lead could be the result of chance.

[Hav]

1. This is one of a few polls

that had Trump up consistently and sometimes by quite high margins. I'd guess their expected demographics differ from the majority of the other pollsters.

[pnwmom]

9. And note how they report their figures with decimal points, as if they could be that precise.

Though they didn't use random samples so they can't even report a calculated margin of error.

[writes3000]

2. They're polling from the same pool of people every time. All this tells us is the electorate is...

Divided and entrenched.

[Democat]

3. This poll's methodology is odd, it questions the same people

This is the one poll that Trump has been using to promote that he's ahead. The poll does not use a random sample of people each week, instead it uses the same group of people for the entire election.

See this thread:
http://www.democraticunderground.com/12512326093

[Bernardo de La Paz]

4. Remain calm. It's a rolling poll, so it still has pre-Convention data in it.

Besides, there are still 100 days to the only poll that counts.

[Doodley]

5. There is something wrong with the selection of their voter panel.

They have only had Trump behind once and he was barely behind then.

[pnwmom]

6. This is an online "panel poll" not a traditional poll using a random sample of voters.

And it includes anyone over 18, not just registered voters, and is only conducted in English.

So it's just a toy, basically.

Here is one analysis that makes some interesting points:

http://www.dailykos.com/story/2016/7/28/1553533/-WTF-is-up-with-the-USC-Dornsife-LA-Times-tracking-poll

The USC “Daybreak” poll is the new Rasmussen, and something very fishy is going on which doesn’t seem to jibe with the rest of the polling world.

If you’re not aware of Trump’s new favorite poll, their approach is this. USC selects a nearly fixed panel of 3000 (they actually slowly build over time), not unlike the RAND panel from the last election, and they repoll ~1/7 of their group each day. Consequently, the numbers represent something like a 1 week rolling average of respondent sentiment.

In principle, this should be great for looking at trends. While the panel may be biased one way or another (and boy it sure is!), changes in the numbers from day to day or week to week represent actual individuals changing their minds, rather than statistical sampling fluctuations.

But there are some very weird features. For one, the poll starts with an extremely R-leaning bias. On July 10, the first day of their data, they have C-3, when the Huffpo pollster had it at C+4 — that’s a 7 point Republican bias. As you’ll see in the snapshot above, according to USC this morning, it’s C-7.3, which points to a nationally tied race once we “unskew the polls.

[Drunken Irishman]

7. RAND was a poll I followed four years ago.

This is the same poll, but now run by USC-LA Times. RAND had Obama consistently ahead during the post-convention polling in the last election:

https://alpdata.rand.org/index.php?page=election2012

Romney led a few times but ultimately, Obama held the lead throughout - even when most polls had him either tied or down.

Their final results were 49-46.

So, this poll does give me pause but then their track record is just four years ago and who knows?

[pnwmom]

8. The RAND poll is still in existence. This one says it uses the same methods,

not that it is the same poll.

And the RAND poll is still listed on 538. 538 doesn't rate the USC poll.

And isn't an online poll conducted only in English likely to under-represent Latino voters and people without computers?

http://magellanstrategies.com/online-sample-for-voter-surveys/

The Truth: Why Online Sample Sucks for Voter Opinion Surveys

As a political pollster I often get asked by clients and the media if Magellan Strategies uses online sample for voter opinion surveys. This is a very good question, and shows that people are aware of the rapid decline of voter households with a landline telephone, and the growth of “cell phone only” or “cell phone mostly” households. Many people think it should be relatively easy for a political polling firm to simply email a bunch of voters and ask them the same questions that we do over the phone. Unfortunately that is not the case, and believe me when I say we have tried many different approaches to get registered and likely voters to participate in a survey by emailing them an invitation. The truth is that the available online sample sucks for a voter opinion survey. Let me explain the reasons why.

[Drunken Irishman]

10. From what I can see, RAND is not doing a poll this year.

It's been pushed over to this one - there is no indication of any daily tracking poll on their website (tho, if you find one please show me, I am curious).

As for online polls, yes, they're not very reliable but this poll isn't a traditional online poll.

All I'm saying is that four years ago, when most major polls had Romney and Obama tied, or a slight Romney lead, RAND correctly projected Obama was ahead - up until the final day with results that were very similar to the outcome.

[pnwmom]

11. Their website says they're doing one. With a poll like this, everything is dependent

on that panel the USC created. Just because RAND was lucky four years ago doesn't mean Dornsife will be equally lucky this year.

https://www.rand.org/labor/alp/2016-election-panel-survey.html

[Drunken Irishman]

12. Which I pretty much said in my original post.

As for RAND, I can't find any tracking poll on their site. I'd be curious to see their results.