G Elliott Morris

data > demagogues

“The party told you to reject the evidence of your eyes and ears. It was their final, most essential command.”

The Trump administration is trying to block politically inconvenient government data Hands off the public's data!

www.gelliottmorris.com/p/the-trump-...

Terrible, but sadly, expected — and the latest remark in an Orwellian laundry list that suggests the president views gov economic data as a source of propaganda to make him look good, not to guide policy for the best interest of all Americans. Facts don’t matter, only Trump does.

Support @gelliottmorris.com
Strength In Numbers.
I do. It's invaluable. Support this team. Read the below post. Be informed.

We'll be updating our new generic ballot average every day until the midterms. Follow along at www.gelliottmorris.com/p/data

(a) i did not say that
(b) you are more than welcome to propose a better predictor of House election results than presidential partisan lean over the last two cycles + a national intercept. we can see which one backtests better historically and use that

Updated our economic index with today's new data. Not a good trend! www.gelliottmorris.com/p/data

<3

i didn't touch on this in the article (it is supposed to be an announcement article, rather than an analysis) but worth noting Reps are down 5 points in the generic ballot so far this year, and took big hits after tariffs, abrego garcia, and LA. Dems up modestly since OBBBA.

On average, the party not in control of the White House gains 6 points from this point in the midterm cycle to Election Day (next Nov). Sometimes they gain more (11 points in 2009-2010!), sometimes less (2018 — started high).

Dems up +2.3 today.

www.gelliottmorris.com/p/democrats-...

Democrats lead House generic ballot polls by 2.3 points How our new Strength In Numbers generic ballot average works, and what a 2-3 point win would look like in Congress

Democrats lead in our new average of House generic ballot polls by 2.3 points.

This puts them ahead of pace for the average out-party in midterms going back to 2006, though they lag the big blue wave of 2018 (for now).

Please share and subscribe ⬇️

www.gelliottmorris.com/p/democrats-...

The Democratic image is underwater — but they're not losing The party has a near-historically low favorability rating. This is mostly due to disaffection on the left, where voters still say they'll vote Democratic for the midterms

I got close to the same number when I did this math on my own polling data, on Tuesday: www.gelliottmorris.com/p/democratic...

Democrats Regain Advantage in Party Affiliation Americans' party identification has shifted from a Republican advantage in 2024 to a Democratic advantage in the second quarter of 2025.

If you do some light math on the new Gallup poll, Democrats are taking a 9-point hit to their favorability rating (18 points on net!) due to having a lower favorability than the GOP *among their own voters.* news.gallup.com/poll/692978/...

thermostat go click-click-click

i just thought it'd be faster and i didn't need to save the code bsky.app/profile/rbrt...

generic ballot average, coming out tomorrow, 7am at gelliottmorris.com/subscribe/

he's coding directly in the terminal. somebody stop him!

Key Inflation Measure Rose in June, Adding to Pressure on Fed

The price index was the latest indication that President Trump’s tariffs are beginning to show up in consumer prices.

Via @bencasselman.bsky.social

www.nytimes.com/2025/07/31/b...

NC Gov Cooper is up by 1 point in the NC Senate race in this new poll — with an R+8 party ID breakdown. For context it was probably like R+6 in 2024 (someone with the voter file can check that for me), and the turnout environment will favor Ds next year. So…. irp.cdn-website.com/07e08705/fil...

If you swing the Rio Grand Valley by double you are still looking at roughly a D+8 pop vote to get back to 13 seats. Under the current map they’d be at 14 at D+10. Again I think this is beside the point given the impact at the map in the R+2-D+5 range

D to R. Damnit man

I meant from R to D. That seems obvious so I’ll leave up the tweet

and remember that the low-probability events get boosted in the markets by gamblers trying to get a huge payout…

Texas Republicans are trying to cheat Democrats out of midterms victory That, and other topics from the July 2025 Strength In Numbers reader Q&A

The new TX GOP congressional map would flip 5 House seat from R to D. It would stay at 30R - 8D seats until Dems hit a 7-8pt national popular vote win. Then, they’d still be behind current 13 seats until they hit roughly +15 in the nat popular vote (not happening).

open.substack.com/pub/gelliott...

DRA 2020 A free app to create, analyze and share redistricting maps for any state.

If you shift this new map 10 points toward Dems on margin (increase D vote share by 5 points), it goes from 30 R 8 D to… 30 R 8 D. davesredistricting.org/maps#analyti...

Pls stop calling the new TX map a “dummymander.” It’s not. The GOP was careful not to dilute R votes in Lean R seats.

In a normal redistricting cycle this would be a biased map. Mid-cycle, at the president’s request, it’s an insidious partisan plot to steal seats. That’s what’s worth focusing on.

The way Rs redrew the map, there is no meaningful weakening of Likely R seats to Lean R/Tossup to turn those Dem seats red. It’s entirely a shift to the right. They were smart about this, and carved out voters from deep, deep red seats to move the needle. No “dummymander”

bsky.app/profile/davi...

The way Rs redrew the map, there is no meaningful weakening of Likely R seats to Lean R/Tossup to turn those Dem seats red. It’s entirely a shift to the right. They were smart about this, and carved out voters from deep, deep red seats to move the needle. No “dummymander”

bsky.app/profile/davi...

Having spent all day deep in the weeds on the new TX GOP map, I concur—it's in no way a dummymander. The "bluest" GOP-held district, #TX23, would still have gone for Trump by 15 points.

To put that in context, the reddest Dem-held seat, #ME02, went for Trump by 9 last year.