G. Elliott Morris just called Texas a tossup — and I mean actual tossup, not lean-R, not likely-R. Meanwhile, I still don’t have a single in-state fundraising number for James Talarico. Welcome to Senate Pickup Watch. Today we’re poking at the Morris call, asking what has to be true for Talarico to really win, and following the money trail from Austin all the way to Maine. Morris going past Cook is real — he’s the first credible analyst this cycle to say “tossup” on Texas, and I’m not brushing that off. But a model isn’t a ground operation, and right now only one of those is actually in the field. And the Boston Globe is already writing the Maine chapter of this story, which tells you the nationalization engine is running. The question is whether that helps Democrats or just hands Susan Collins a villain. Jeff Singer, writing in The Downballot:
Texas Attorney General Ken Paxton defeated Sen. John Cornyn in a 64-36 landslide in Tuesday’s Republican primary runoff, an outcome that Senate GOP leaders fear could cost their party control of a crucial Senate seat and divert much-needed money from other battlegrounds.
G. Elliott Morris called Texas a tossup — not Cook, not a tweak from “likely” to “lean,” but the actual word “tossup” from a named forecaster. That’s a line in the sand. But I still don’t have a public Q3 in-state fundraising number for Talarico, so Morris’s model is sitting on a structural gap I can’t check. A tossup call from Morris is a model output, not a structural verdict on whether working-class Texas voters are actually ready to move. For it to hold, you need a very specific coalition shift: high-Hispanic-turnout precincts, some suburban crossover, and a nominee who fits that electorate. Talarico is a state rep from Travis County. That doesn’t make him the wrong fit automatically, but it doesn’t make him the right one either. And Tillis calling Paxton the Jeffrey Dahmer of ethics violations is great color, sure, but color doesn’t pay for field offices. Sixty-four to thirty-six in a Republican primary is the kind of number that gets forecasters’ attention — the question is whether it moves precincts in November, and that answer lives in FEC filings we haven’t seen yet. The piece says Talarico has “strong fundraising,” but it never gives us the number. Morris’s model and VoteHub’s path scenario are both built on assumptions, so what’s the actual threshold here? If we can’t name it, we’re just laundering a rating into a story. Here's VoteHub:
After fully switching the Republican nominee to Paxton this morning, the race moved from Tilt R to Tossup for the first time. Paxton is now the narrowest of favorites, with a 54.0% chance to win and a projected margin of 0.8 points.
Okay, so Morris has it as a tossup, VoteHub has Paxton at 54 percent with a projected margin under one point — that’s a real model number, not a vibe. After Cook slid it to Lean Republican, we’ve now got forecasters saying “tossup” out loud, which is the escalation I was waiting for. But here’s the catch: VoteHub’s own model says this needs a blue national environment, better Democratic performance with key Texas constituencies, and Paxton vulnerabilities all landing at once. That’s three conditions. I still don’t have a Q3 in-state fundraising number for Talarico that tells me the ground game exists to thread that needle. The 54-to-46 split is worth taking seriously, but I want to know what’s underneath it. VoteHub blended Paxton and Cornyn before the runoff because the first round was close, then switched to a pure Paxton input this morning and the race jumped a full tier. That’s a huge single-input swing, which tells you the model is very sensitive to candidate identity. So the question is: is it pricing working-class realignment correctly in a state with this much Hispanic-turnout complexity, or is it leaning on polling that tends to underrate the Republican floor in low-density Texas counties? And that’s exactly why I’m not ready to fully buy the tossup call yet. Twenty-eight points is how badly Paxton beat Cornyn — that’s a dominant primary win, not a wounded candidate stumbling into November. The vulnerabilities are real, but a guy who just crushed an incumbent senator by 28 points is not walking into the general short on energy or money. Right — Paxton’s primary margin almost argues against the toxicity story. The Republican base turned out hard for him. So the tossup call lives or dies on whether persuadable Texans who voted Cornyn in the primary either stay home or cross over. That’s a real hypothesis, but it needs to be said out loud as a hypothesis, not smuggled in as a probability. G. Elliott Morris, writing in G. Elliott Morris:
The general impression that Paxton is a worse general election candidate than Cornyn would have been is not misguided. Empirically, hypothetical general election polling showed Talarico doing about 1-2 points better against Paxton than Cornyn across many polls. That is not an earth-shattering difference — but it might be enough in a close race.
G. Elliott Morris called Texas a tossup last night, and I want to be precise about what that means. Morris going past Cook’s lean-R is a real line in the sand, not a vibe shift. But he’s building it off hypothetical general-election polling that has Talarico running one to two points better against Paxton than Cornyn. That’s the margin he’s hanging the call on. Where is Talarico’s in-state Q3 money number? Without that, I can’t tell you whether the campaign infrastructure exists to actually capture that advantage. Morris is explicit that Paxton has a “negative candidate residual” — that’s his technical term, and it’s doing a lot of work in the tossup call. The structural question is whether his model is pricing Hispanic turnout correctly in low-density Texas counties, because that’s where the polling is shakiest and where a one-to-two point swing would have to come from. A tossup built on a margin that small in a state with that polling history deserves a lot more skepticism than it’s getting. Thirty-two years without a Democratic statewide win is the sentence Morris himself buries in the piece. He’s not hiding the structural hole — he’s just deciding the Paxton residual is enough to bridge it. I’m not there yet unless I can verify the ground side of the math with a fundraising floor. From The Boston Globe:
One thing everyone seems to agree on: Cornyn would have been the stronger general election candidate, to the point that Texas would have remained a major reach for Democrats this year. But with the scandal-ridden Paxton as the GOP nominee — and after tens of millions of dollars were spent by Republicans trashing him in this primary — the state is now very much in play.
G. Elliott Morris called Texas a tossup — not Cook’s “lean R,” not a little rating tweak, the actual word “tossup.” I’ve been waiting for a named forecaster to go there, and now one has. But here’s my problem: the Globe piece is already using that call to jump over to Maine, and that’s exactly the kind of nationalization move that burns Democrats in states where voters don’t want Washington telling them what to think. The Maine connection is real, but the Globe is collapsing two separate questions into one. If Paxton’s toxicity bleeds into Collins’ race, that’s a Republican national brand problem — that’s one story. Whether Texas itself is structurally winnable for Talarico is a completely different story. Morris’s model is probabilistic, not structural, and in a high-Hispanic-turnout, low-density state where polling tends to run hot for incumbents, I want to know what assumptions get him to “tossup.” And I still don’t have a Q3 in-state fundraising number for Talarico. Morris can call it a tossup; the FEC can tell me whether there’s a ground operation behind it. Until those two things are in the same sentence, the forecaster and the campaign are living in parallel universes. The money angle the Globe raises cuts both ways. If every forecaster now prices Texas as competitive, Republican national money has to fight on two fronts at once. That three-to-one cash ceiling I flagged earlier this month gets a lot more complicated when the GOP can’t just bank Texas and redeploy. From 270toWin:
The U.S. Senate has 53 Republicans and 47 Democrats (including two independents). There are 35 seats up in 2026 - including special elections in Florida and Ohio - of which 23 are held Republicans. Democrats will need a net gain of four seats to retake control in 2027.
The 270toWin consensus map is dated April 23rd — five weeks before Morris dropped his tossup call on Texas, which means this composite is already a historical document. Ohio moving from lean-R to tossup is baked in here, but Texas isn’t, and that’s the number that actually matters for where DSCC dollars go next. And that’s the problem with consensus composites in a fast-moving cycle — they’re averaging Cook, Sabato, and Inside Elections as of a fixed date, so by the time Morris goes further than all three on Texas, the “consensus” is already behind the news. What the map does tell us is the baseline: Democrats need a net four seats, they’re playing defense on almost nothing, and the 23 Republican-held seats up include some genuinely soft targets. The open question is whether the consensus catches up to Morris, or whether Morris is the outlier. I keep coming back to the fundraising gap underneath all of these ratings — $939 million Republican to $267 million Democratic this cycle. A four-seat net gain on a three-to-one cash deficit means you cannot afford to be wrong about which states are actually gettable. Every time a composite map moves Ohio to tossup and Texas still isn’t rated, that’s donor oxygen getting pulled toward a race before anyone has verified there’s a ground operation to fund. If Senate Pickup Watch helps you stay ahead of the map, consider subscribing wherever you’re listening. And if you have a moment, leave a quick review — it really helps other people find the show.
You’ll find links to every story we covered today in the show notes. If one of them caught your ear, take a minute to read it in full.
That’s Senate Pickup Watch for today. This is a Lantern Podcast.