Even if you don't trust polling, you should at least trust polling TRENDS

[RidinWithHarris]

Of course, if you don't trust polling because you think the numbers are just totally made up to promote a story line or an agenda, I can't help you.

But if your lack of faith in polls come from doubting the likely voter models, sample size, "no one has land lines anymore!", etc., then at least realize this: whatever skew a particular pollster might have, it will be roughly the same skew every time a poll is taken.

When lots of major polls show a sizable shift toward Harris and away from Trump, you can be pretty damn certain that shift is REAL.

This election is truly shifting our way, big time.

Will that hold? That's not something polling can tell you. My own gut feeling (for whatever that's worth) says YES!

[Walleye]

1. I believe the enthusiasm, and I think the shortened campaign season is really appealing to voters

We are not used to planning months and months ahead anymore

[Fiendish Thingy]

2. Note: averages are not the same as trends

Trends reflect movement and momentum, and the majority, if not all polls, show significant movement in Harris’ favor, regardless of the specific numbers reported.

[RidinWithHarris]

4. The polling average is trending, and individual polls themselves are also trending, in Harris's favor

Yes, trends and averages aren't the same thing, but they aren't mutually exclusive either.

[ColinC]

3. Always

The polling trends have been substantially reversed which is wonderful

[VMA131Marine]

5. Also, while I would argue polling averages are more accurate

they are lagging indicators when there are major changes in the race as we are seeing now. The current polling averages still contain results from just after President Biden withdrew and things have clearly moved a lot in VP Harris’s favour since then. There will be more movement during the convention.

IMO this is what is freaking Trump’s team out. They know Trump is going to be down substantially in the polling average after things settle down and they will have very little time to change the momentum before things set in concrete.

[RidinWithHarris]

10. There's no fixed window of time that you have to use for polling averages

But yes, polling averages that include Biden/Trump numbers mixed in with Harris/Trump numbers will naturally track more slowly in Harris's favor.

If you average post-Biden-dropping-out polling, that average, the only average that's really going to matter, is definitely going our way.

[VMA131Marine]

18. That's why I say they are a lagging indicator.

The averages will take time to catch up with events on the ground. Look at the trend in the individual polls to see which way the average will go.

[Rubyshoo]

6. THANK YOU !!!

[RedSpartan]

7. LFG

[getagrip_already]

8. Trends are only valid if they dont change their turnout models

But they are constantly changing those, and they should.

But in reality, they have no way to know who, or in what numbers, will show up at the polls.

Legitimate pollsters lose a lot of sleep over this.

Will more young woman turn out? More white men? More blacks/hispanics? More young voters? Etc, etc.

Changes in those model inputs will change the outcome of a poll.

Are they changing them based on recent enthusiasm, or keeping pre-kamala ratios?

Are rasmusen like pollsters going to lean more towards a red wave?

We dont know. We do know they can easily impact results that will impact trends.

So dont be shocked if the trends change.

All that said, im much happier with good numbers and trends than bad, i just know that can change.

[RidinWithHarris]

14. It's highly unlikely that the strong positive trend we're seeing is a whole bunch of separate pollsters...

...all changing their modeling at the same time in more or less the same way.

[getagrip_already]

16. Im not saying that..

The recent polls were conducted under the old models. That in itself is very good news for us because it measures real change in feelings.

Its future polls they will adjust that will impact the trends.

Right now some of the companies are wondering what went wrong in the polling. The results were unexpected to them.

Now they can make adjustments.

[WarGamer]

9. People tend to trust polls that reinforce their internal biases and reject those that go against them...

I can show many links...

[RidinWithHarris]

11. I personally don't do that, which is why I was getting pretty damned scared...

...until the last couple of weeks.

[WarGamer]

12. Basically polling now is the same as it was pre-debate

[RidinWithHarris]

15. The same? Where do you see that?

Are you (and this is not mathematically valid) saying polling averages are "basically the same" until the average change is larger than each polls margin of error? Perhaps even using double that margin of error as the metric for how far apart the numbers have to be?

[WarGamer]

17. I'm saying that before the debate, it was a 50/50 race with some polls showing Biden ahead.

In other words... the same as now.

[RidinWithHarris]

19. Fair enough

Even getting back to the pre-debate state of things is good thing, however.

A less rigorous argument, but I still think worthy of consideration, is that the huge burst of enthusiasm for Harris, if sustained, means likely voter models are likely underestimating Democratic turnout which (based on many special elections where Democrats over-performed) might have already been weighted too much against Democrats.

[Cosmocat]

20. Eh

Before the debate I felt Joe was behind generally a but and clearly electorally. I hate what they did to him after the debate, but he was bleeding out more and more overtime from that point.

[ImNotGod]

13. yes, i only look at trends, everything else is just noise. nt