Share this page!

Lakshya Jain Profile picture
Twitter logo
Jan 10 • 5 tweets • 1 min read
Both Katie Porter and I have something in common (in that neither of us will ever be a US Senator).
OK a more detailed read on #CASen that isn't just a flippant comment:

Porter's odds, I'd say, are really around 10%. Schiff is going to be *the* SoCal candidate, beats Porter with every one of her strong demographics, and has *way* more money than anyone realizes.
Lee and Khanna will run from NorCal. I do not think Breed runs, but if she does then she becomes the favorite (or co-favorite, with Schiff). Ultimately, I think that Porter would have a way better shot if Adam Schiff wasn't running, but he is, and that's a problem for her.
Porter is going to get very few endorsements compared to Schiff and doesn't have a disproportionately higher name ID to offset her cash disadvantage against him. They're competing for the same geographic base, and Porter isn't going to do great in the Bay Area.
Similarly, the geographic factors I've mentioned also make me think people are too quick to discount both Khanna and Lee. Neither of them wins if Breed runs, IMO, but I don't expect her to. Let's see if anyone else jumps in.

• • •

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
Jan 9
FWIW basically any objective measure has John Fetterman as a significantly above replacement candidate and Dr. Oz as a horrible, *horrible* one — Fetterman would have won even if the gubernatorial matchup was Shapiro vs a generic R (if that could ever emerge from a PA GOP field).
I generally don't do class-based analysis because I'm not that great at punditry but I do think there's this thing where a lot of the college-educated folks are extremely annoyed that someone who refuses to dress or talk like them despite actually being One Of Them managed to win
yeah Fetterman had a stroke that somewhat limited him physically, but it's been well-documented that this doesn't have any lasting *cognitive* issues and I think voters recognized this sort of thing better than many consultants do (plus, it likely helped people relate to him).
Read 4 tweets
Lakshya Jain Profile picture
Dec 30, 2022
Probably makes it clearer now that we have the margin included with the WAR, but these are all the districts that would have flipped if the matchups were "generic R vs generic D", as per our @SplitTicket_ model.
@SplitTicket_ So, I want to talk a bit about what that actually means.

You can't really measure how good one candidate is without examining how good or how weak their opponent was. From his previous elections, we know Rudy Salas was an excellent candidate. The problem is that Valadao is too.
@SplitTicket_ So, what you really do is to measure the candidate quality differential of a matchup and compare it to the modeled generic matchup, as this is the only baseline you can have. So, Valadao performed 5 points better against Salas than a generic R would have vs a generic D.
Read 5 tweets
Lakshya Jain Profile picture
Dec 30, 2022
🥁🚨🚨🚨🥁

PRESENTING: The @SplitTicket_ 2022 House Wins Above Replacement model — our attempt at boiling candidate strength with one number. If you've ever wondered how candidate quality impacted a race, check out what our metric suggests!

split-ticket.org/2022/12/30/our…
@SplitTicket_ There are three main themes we touch on.

(1) Democrats ran stronger candidates where it mattered, in general. 45 battlegrounds saw Dem overperformances, while 37 battlegrounds saw GOP overperformances. This fits with what we've seen in other elections.
@SplitTicket_ (2) Democrats probably won more seats than they should have, even considering the overall environment. We think they ended up winning 9 seats that would have gone the other way with a generic nominee, compared to the GOP's 6.
Read 5 tweets
Lakshya Jain Profile picture
Dec 7, 2022
OK, very brief methodological thread on how we did this. We began with precinct early vote shares and splits from November and registered voter numbers, split by race and precinct. Then we took statewide turnout rates and shares and adjusted them by precinct using rake weighting.
The key variable here is white vote in support share. You begin by holding Black/Hispanic/AAPI support shares constant, applying statewide turnout shares by race per precinct, and calculating what the white vote would have to be as step 1.
This isn't perfect. You're going to fall short in some places. Here, it's key to look at precinct composition and see the margins you've put in. If your vote shares exceed the bounds you've set in a precinct (e.g. whites in GA aren't 90%), you start playing with electorate shares
Read 9 tweets
Lakshya Jain Profile picture
Dec 6, 2022
Absentee file updated last night, so there'll be one more update to the Georgia model, but in general I'm not quite sure there's much good news for Herschel Walker heading into this election, from everything I can model (and maybe I'm missing something massive, who knows...)
Like I said, in 2021 I could construct several cases in which Perdue narrowly squeaked by Ossoff in my model at the end, though I found them wholly unconvincing. Walker could win, but he's facing a deficit of 290K votes after the last update in our model.
Even with 1.56M turnout on election day, if Walker gets November splits of R+15 in the general, we're looking at Warnock +1 in my estimation. Walker needs to blow the doors down with the splits. Very possible, but it shouldn't be understated as to how tough it is.
Read 4 tweets
Lakshya Jain Profile picture
Dec 5, 2022
Our final @SplitTicket_ #GASen runoff model has Warnock +3 as the predicted outcome and a race rating of Leans Democratic. Warnock has an early vote lead of ~280K, by our estimation, and Walker needs a whopper of an election day to reverse things.

split-ticket.org/2022/12/05/geo…
@SplitTicket_ I will say that the only reason I have this at leans Democratic instead of likely is because I have no idea what election day is going to look like, nor do I find it useful in projecting a false sense of confidence about it. This is worse for the GOP than Loeffler/Perdue 2021 IMO
Model made by me and @ADincgor (hey, sound familiar?), with data cleaning assistance from @HWLavelleMaps and @Thorongil16 (who also made the map). An interactive county map can be found here rpubs.com/splitticket/ge….
Read 5 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

Don't want to be a Premium member but still want to support us?

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

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

Send Email!

:(