CBS News / YouGov: Braun +3 (IN), Sinema +3 (AZ), Tied in FL

[Zaybay]

It's not really fair to immediately trash 538's model just because Braun has taken a small lead in a handful of recent polls. Data-driven models are intentionally cautious in reacting to new polling data. They are meant to have an element of stability based on other factors independent of polling, such as fundamentals (although it's fair to debate what should actually count as a fundamental).

A very clear trend over several credible polls is required to counteract those fundamentals. It's a feature, not a bug. If your probabilities swing wildly with each new poll, you don't really have a model. You have just a regular polling average, and a very noisy one at that. Even most ordinary polling averages collect polls over an extended period of time to minimize statistical noise.

YouGov is the first non-partisan pollster in a long time to show Braun ahead, so any expectation that the models will swing dramatically just of because of that is absolutely unrealistic and uninformed.

This is the correct take right here.

[Zaybay]

Gonna be honest, when I heard YouGov was gonna release some polls, I thought AZ would be R, IN would be R and FL would be a tie/a tiny bit D. I was surprised, but not in the way I thought I would be.

My reasoning for this was because, and this is important, YouGov is an online pollster. Their polls are not really that great put next to Fox, Q, Monmouth or other live callers/interviewers. Because of this, its likely they will get odd results, as other online pollsters(such as Ipsos) have gotten.

Its funny, looking at Ipsos and YouGov, both are really the inverse of each other. Ipsos constantly overestimates D chances, but has also been rather swingy, having them lead by 15 one day and 2 by the next. YouGov, however, constantly underestimate D chances, but is much more static(insert every GCB poll YouGov has done).

Anyway, its a datapoint, but not one to treat as the end of the world, and that goes for both sides.

But if there is one thing to discount, its that FL poll, that thing is complete and utter garbage.

[Zaybay]

George Khalaf, the Local Pollster of Arizonas "Data Orbital" who admitted Friday that their own Poll was wrong has come out swinging at CBS/YouGov for the trash they did with their AZ Poll...read here:

https://twitter.com/George_Khalaf/status/1056582327017472000

Based on CBS Turnout Model without D-Surge McSally is ahead.

I am not a fan of these polls, but this is a terrible excuse. All it is is looking at the current early vote numbers, which, in AZ, is a stupid idea. The state of AZ is still heavily R registered, but the actual voters are not heavily R. By looking at the numbers in AZ, you miss who the voters are. While the early vote is pretty R, about 8% more last I checked, it misses who those Rs are, and who the Is vote for.

And this isnt helped by the fact that that R% has been going down this whole time.

Should also note that this guy may have a reason to dispute a Sinema-leading poll:
https://twitter.com/George_Khalaf/status/1056617500064985088

[Zaybay]

snip(way too long)

His complaints about the poll are that its online, the same as mine, and literally doing exactly what I said was a bad idea, taking early vote at literal face value. He is only assuming that Kavanaugh has caused the R voters to coalesce, which both makes no sense(Trump has not seen an approval jump in AZ, nor has polling reflected a shift since Kavanaugh in AZ, an event that has literally been forgotten about in most states) and has no basis in data. And if you have such complaints, I will refer you that the poll also has Indies going to McSally, which is likely not going to be the case.

[Zaybay]

So the AZ part is trash but IN and especially FL are fine? That’s really being hackish

AZ doesn't make sense. You really believe AZ will be only 70% White? I do not believe that for a minute. Too many young people as well.

Possibly. Pollsters have been saying the Hispanic/Latino vote is starting to get a noticeable uptick. Also again the FL poll looks like an outlier but you have no problem with that and I dang sure know you have no problem with IN

Not to mention in the AZ poll, Indies are going to McSally, which is rather ridiculous as all other polling has shown otherwise, but nope, no complaints there.

[Zaybay]

Listen, everyone is trying to guess the right turnout model, which is why all the polls are all over the place. The averages will probably end up being fairly accurate, as usual, but this election is pretty unpredictable.

I agree with YC here. The turnout model is a big reason why the polls could be spot on or fail completely, and thanks to the few amounts of polling, we could be taken by surprise, either by the D incumbents winning handily, or the Rs taking more than 4 seats in the senate race.

(Personally, based on the early vote, I would lean towards the former, but I would actually have to wait for the election to confirm)