G Elliott Morris

3.6 points is actually lower than the average historical error on vote margin of second-round polls in Brazil

Brazil elected Lula again tonight. Poll averages appear to have had Lula and Flávio Bolsonaro effectively tied, maybe Bolsonaro up 1 on Lula — so not sure why the markets had this race at 80-20 Bolsonaro. Markets might miss big in the U.S. midterms in 9 days too.

Small-dollar donations to Republican candidates are just apocalyptic. Selected examples here. Our forecasting research found in-state individual contributions are a strong predictor of results. Way stronger than PAC spending! votehub.com/2026/10/24/2...

In this piece… 1. Some data on what undecided and low-turnout-propensity voters care about, from our latest poll 2. Another EV update: Infrequent Democratic voters are turning out at a much higher rate www.gelliottmorris.com/p/2026-10-25...

on this panel, we all came together and fixed non-response bias forever

New: Low-info voters, who put Trump in the White House, now look even more anti-Trump than high-info voters, and give Dems bigger margins on the generic ballot. 1 out of 3 Trump voters who can't name the party in charge of Congress now back Dems for 2026. www.gelliottmorris.com/p/poll-low-i...

talking about fakers open.substack.com/live-stream/...

Not sure why you would try to push expectations up for the GOP right now

Strength In Numbers is covered here www.thebulwark.com/p/a-guide-to...

Here’s a piece from me out this AM for MS NOW as part of their partnership with FiftyPlusOne. Trump’s net approval rating is the worst ever for a president heading into a midterm, and the GOP has lost its advantage on handling the economy and immigration www.ms.now/opinion/trum...

Smaller venues, empty seats, bored crowds, and a sleepy, confused headliner. The Trump rally is the defining political event of our era, repeated hundreds of times in all its grotesque glory, and now it's limping to an end. My latest at Public Notice: www.publicnotice.co/p/trump-rall...?

Last wk, Strength In Numbers polled people who previously said they voted for Trump in 2024 but will no longer cop to it. Most are now committed anti-Trump Democratic voters. This "recall error" could be biasing polls against Democrats (small fx, <1 point) www.gelliottmorris.com/p/2026-10-22...

FiftyPlusOne's investigation into "Wedgewood polls" also covered here www.theguardian.com/us-news/2026... and here www.pbs.org/newshour/pol...

it's not a mystery why this guy is up 3 in texas. whether that will last or withstand polling error is another conversation. but clearly he's got the juice

(2) Going into 2025 these same poll-based pundits promised Trump (& US) how popular his policies were despite his policies being really unpopular. They misidentified Trump's popularity on immigration with popularity of his actual policies on immigration (and so forth). They gave Trump a fake mandate

proud of this piece coming tomorrow www.gelliottmorris.com/p/2026-10-22...

no, it was not a 'polling organization' !

in this case it is really the intercept that matters most -- and also note the variance from the trend

A timely reminder. But note that while the slope of every graph is positive, the x-intercept is sometimes positive (2014; Rs outperformed on average in light blue states)... and sometimes negative (2012; Ds outperformed on average even in moderately red states).

Coverage of FiftyPlusOne investigation: “New Entrant Pollster Was Publishing All Fake Polls” talkingpointsmemo.com/edblog/new-e...

Updated the Trump approval and party trust trends with our new Strength In Numbers/ @verasight.io poll. All-time lows for Trump on prices, deportations — pretty much everything. Plus, Democrats now lead on immigration, foreign policy, and the economy www.gelliottmorris.com/p/2026-10-21...

New NYT and Strength In Numbers polls this morning showing roughly same thing regarding national environment, around D+11-12 lead for Dems in the generic ballot depending on population. Double digit margin now a 50-50 chance in my modeling of all polls www.gelliottmorris.com/p/2026-10-21...

New Strength In Number/Verasight poll conducted Oct. 13–19, 2026: - Generic ballot: D+12 LV, D+10 RV - Trump approval: 34% adults - Trust to handle most important problem: D+13 - Dems overtake Rs on trust to handle immigration, economy More (free to read): www.gelliottmorris.com/p/2026-10-21...

maybe he was just doing a social experiment ?

The New York Times editors' note on the Wedgewood Polls fiasco mentions FiftyPlusOne and links back to our investigation. We are happy our reporting on this bad actor is being recognized, and I applaud the reporting team, including Mary Radcliffe Cooper Burton, on their hard work!!

Multiple polls showed McMorrow losing ground, not just Wedgewood. But yes, the individual/s responsible should probably be held accountable for this.

If 10 more people subscribe to 50+1 because of this, we can recoup the cost of the actual polling data we bought to prove that Wedgewood was faking its polls (see chart). (If not, it was a worthwhile expense!) blog.fiftyplusone.news/p/fiftypluso...

That means you got ’em good. Kudos!