The researcher behind Thematic Markets and a returning guest on The Grant Williams Podcast. In episode 01, recorded on the eve of the 2026 midterms, he defends a forecast that the polls were missing low-trust populist voters, explains why he calls it a judgment rather than a statistical finding, and works through what a Republican hold would mean for Donald Trump's last two years. In episode 02, recorded four days after the Democrats won both chambers, he comes back to take the forecast apart: three of his four assumptions held, and he puts the miss down to his own bias rather than to the framework. Start with episode 01, at about 48 minutes for the call, then episode 02, at about 4 minutes for the post-mortem.
Across the episodes
Enters as the forecaster whose out-of-consensus midterm call the episode examines, a market researcher who trusts judgment where the statistics run out
Barth casts himself as an outsider to election forecasting: the calls are a by-product of research written for investors.
"My research is really economic and market research, right? So what am I doing forecasting elections out there?"Marvin, The Grant Williams Podcast: Forecasts and Fortunes, Episode 1 (2 Nov 2026), 8:01
His standing as a forecaster rests on the cycles where he broke from the polls, not on the ones where he agreed with them.
"And in particular, it's been in calling them against the grain of what the consensus was."Marvin, The Grant Williams Podcast: Forecasts and Fortunes, Episode 1 (2 Nov 2026), 8:17
The midterm forecast rests on older work: his 2016 survey of populist manifestos supplies the definition of populism the whole argument uses.
"What populism really comes down to, and so I did this study back in 2016, which I think we've talked about on some of our previous conversations, The Politics of Rage, where I went through populist party after populist party across 22 different democracies, what their manifestos actually said."Marvin, The Grant Williams Podcast: Forecasts and Fortunes, Episode 1 (2 Nov 2026), 19:17
Barth answers the charge of partisan bias with his incentives: as an independent researcher he is paid for accuracy, not for a side.
Grant had raised Barth's past service as a Bush administration appointee as something people would read as Republican bias.
"Today, my clients pay me to get the right answer. I don't give a damn, right?"Marvin, The Grant Williams Podcast: Forecasts and Fortunes, Episode 1 (2 Nov 2026), 36:15
Returns as his own auditor, keeping the method of the first conversation while giving up its conclusion, and naming his own leanings as the fault
Barth opens the post-mortem on his own reasoning rather than on the result, the order the whole episode keeps.
"Let's start with where I came from. What were the things that were driving my view?"Marvin, The Grant Williams Podcast: Forecasts and Fortunes, Episode 2 (7 Nov 2026), 4:08
He locates the miss in the analyst, not the data: the error is the prior he brought, the very charge Grant had warned critics would make in episode 01.
Barth has just said the one assumption that failed was that the swing voters would vote Republican.
"And I think that in deep introspection comes down to my own biases here."Marvin, The Grant Williams Podcast: Forecasts and Fortunes, Episode 2 (7 Nov 2026), 8:54
Method and call are separated: he keeps the episode-01 framework and discards only the assumption he layered on top of it.
"But the framework, which is really what I try and focus on, what I think is the most important thing for an analyst to look at, actually did seem to work. I just imposed a bias on it that gave me the wrong answer."Marvin, The Grant Williams Podcast: Forecasts and Fortunes, Episode 2 (7 Nov 2026), 8:59
The bias he blames, stated outright: what episode 01 answered with "my clients pay me to get the right answer" is here conceded as the likely source of the error.
Barth recalls, as in the last episode, that he served in a Republican administration.
"And in general, my political leanings are towards the Republicans."Marvin, The Grant Williams Podcast: Forecasts and Fortunes, Episode 2 (7 Nov 2026), 11:32
The reason he came back, put as a working rule: the post-mortem is part of the process, not an apology for it.
"And that's what I try and focus on is make sure that I learn from my mistakes."Marvin, The Grant Williams Podcast: Forecasts and Fortunes, Episode 2 (7 Nov 2026), 52:56
Connections
Barth's 2016 study of populist party manifestos across 22 democracies.
Barth publishes his institutional research, including the midterm piece, through Thematic Markets.
Grant introduces him as a returning guest and a friend.
Barth returns after the vote to explain why his call failed.