Share this page!

Lakshya Jain Profile picture
Twitter logo
14 Sep, 3 tweets, 1 min read
On the eve of #CARecall, we can probably say that this race is safe D, because there is genuinely a < 5% chance Newsom loses tomorrow. It would take a polling error of a magnitude large enough to end the industry in California altogether.
My prediction is Remain +21 (with a rather high error range — nothing from Remain +15 to Remain +23 would surprise me), but we probably won’t know the full results for a while. But the race should probably be called tomorrow night. I’d be surprised if it isn’t.
(someone better catch the Brexit meme here)

• • •

Missing some Tweet in this thread? You can try to force a refresh
　

Keep Current with Lakshya Jain

Lakshya Jain Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @lxeagle17

Lakshya Jain Profile picture
11 Sep
If I had to take a guess at the #CARecall margin, I’d probably say No wins by 17, based on the available evidence. There’s been some very serious late movement towards Newsom and it’ll take a Herculean lift and a massive polling error for Yes to have a whiff of a chance here.
And once again, it was exceptionally smart of Newsom’s camp to not field a replacement candidate. That’s the only way you keep the Democrats all in line, and if you do that, recall gets squashed.
Anyways, if that does happen, Newsom would emerge stronger than ever before, because the press insistence on making it a horse race would backfire badly and cost the CA GOP their best 2022 candidate in Faulconer, and Newsom would have overperformed public expectations by a lot.
Read 4 tweets
Lakshya Jain Profile picture
9 Sep
No such question has been settled, considering that non-college rural whites have one of the lowest voting observable propensity rates, and if you think people will let a vaccine mandate from 2021 influence their 2022 voting likelihood, you might be overreacting again.
I feel like the last five years should have taught people not to overreact to a single news cycle and yet here we are every time something happens that conservatives or liberals think will be the death knell for the other party.
FWIW, racial minorities, who still break heavily Dem, have fairly low voting propensity rates too, so the "will the 2022 electorate be more or less Biden-supporting (in terms of presidential vote cast) than 2020?" question probably hinges to some degree on their turnout also.
Read 4 tweets
Lakshya Jain Profile picture
30 Aug
🚨#VAGOV MODEL ALERT🚨

@Thorongil16 and I have made a model for the 2022 Virginia governor race, based on polling and state fundamentals.

Our current forecast is Likely Democratic, with a Democratic win probability of 79%, and a D+8.2 forecasted margin.

striderforecasts.com
50% of the model is based on polls, weighted for sample size and recency. 25% uses a 2013/2017 composite educational turnout screen and applies polling margins by education. 25% uses 2020 margins, split by education, for voters and applies it to the poll electorates.
For the county forecast, we calculate the expected lean of each county by taking a composite average of county PVI relative to statewide average for 2020, 2017, and 2013, and we apply our model's overall forecast to the county leans.
Read 6 tweets
Lakshya Jain Profile picture
24 Aug
California recall is under a month away, and we have our first sets of mail return data! Let’s break down the numbers, per @paulmitche11.

My initial takeaway: This is beginning a lot like 2020's mail returns did, and that's pretty good news for Democrats. But still early.
In CA, 68% of November 2020 ballots were cast by mail. AP estimates that Biden won them 72-26. The Dem/Ind/GOP split was 50/25/25 in share of returned ballots. Currently, it’s 56/22/22 in the first batch of data (5% returned; November saw 61% returned).
Yesterday, the report we got indicated 3% of ballot returns. From there, Democratic share has dipped a percent or two, which makes sense considering how few ballots have been returned and how Democrats tend to return their ballots earlier.
Read 13 tweets
Lakshya Jain Profile picture
6 Aug
The biggest argument I see against Conor Lamb's candidate quality is how he "underperformed Biden".

This is not nearly as clear cut as everyone seems to think. Lamb's vote share was 51.1%. Biden's was 50.7%. The thing is, Lamb's opponent got 48.8%, and Trump got 48%.
In 2020, closer districts (particularly suburban ones that are ancestrally Republican) tended to see Biden overperform the House candidate by a fair amount. Lamb underperformed Biden by only 0.4 points in margin, which is about a point or two *better* than we may have expected
Add in 2018, where @EScrimshaw notes how Conor Lamb blew through expectations and thrashed his opponent with a 12.6 point victory despite having $8M in outside spending thrown at him, and I think it's clear -- Conor Lamb is a very, very good candidate. scrimshawunscripted.substack.com/p/pennsylvania…
Read 5 tweets
Lakshya Jain Profile picture
5 Aug
It's well known that SurveyUSA can't poll California -- they've had several ridiculous polls over the last several years that border on polling malpractice.

But this result is fairly shocking. Let's dive in further to see what's going on. 🧵
To begin with, their sample of *registered voters* is kind of off in partisanship and race.

The D/R/Ind. splits in February 2021, per the CA SOS: 46/24/24

The SurveyUSA poll: 46/29/22
The racial splits are kind of strange too. We know, from AP voter analysis, that the CA electorate was ~54% White, 7% Black, 24% Latino, 15% Asian/other.

SurveyUSA has the *registered voter* sample at 61% White, 4% Black, and 15% Latino. That's quite off.
Read 8 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!

Like this author's thread?

Get emails when @lxeagle17 has new unrolls!

Check out other threads from @lxeagle17!

Read all threads by @lxeagle17

:(