Taurus.in.a.China.Shop

TLDR; Forecast: R1-4 estimate: 3,114,070 D1-4 estimate: 2,991,559 Good for Republicans at first glance but probably not. A key takeaway from the turnout data that I haven't seen anyone talk about yet is that the Democrats have essentially created an extra week of early voting (EV) for themselves. As a percentage of their category of voters, D1 voters surpassed their 2022 12 day EV turnout rate this year on Monday, (day 8). D2 and D3 segments are on track to match or exceed their 12 day EV turnout rate when we see numbers for Tuesday (day 9). D4 probably exceeds 2022 EV turnout rate Wednesday (day 10). Republicans, on the other hand, seem to be tracking a day or two ahead of their 2022 turnout percentages. So what does that mean? Assuming that there is no ticket splitting and that people with general election only and new voters split evenly (two absurd assumptions but that's the copium that MAGA seems to be smoking so let's go with it), and if we use 2022's daily turnout rate changes as a benchmark for the remaining days, the categories break down as follows: R4 ~100% (let's call it 99% because 100 is absurd) R3 98% R2 90% R1 74% D4 ~100% (again, say 99%) D3 ~100% (99%) D2 98% D1 83% The excitement from Dems doesn't do much for their final turnout for D4, but makes a big difference for D3, D2, and D1. If you think that 83% D1 turnout is crazy, note that's in line with R2 turnout in 2022. Currently D1 turnout rate is 45% and R2 turnout rate is 47% so Democrat enthusiasm would have to suddenly take a sharp turn to make D1s return to deviate significantly from tracking with R2s. (I wish that D2s were tracking with R2s so I could make an R2-D2 joke). So what vote totals do those percentages estimate? R1-4 estimate: 3,114,070 D1-4 estimate: 2,991,559 In other words, if turnout trends continue then people with GOP voting history (not counting mixed voting history) will have ~120,000 votes to play with (which is oddly close to the 150,000 that I spitballed on the phone with a friend two nights ago). When we're looking at around 10 million votes that's 1.2%. Now my gut tells me that D2 will not hit 98% but instead be in the low 90s. That's not a huge number of votes, but, as we'll see in a bit, those 70,000 or so votes are important. I do expect that D3 and D4 will get as close to 99% as is possible. Now let's look at some of the assumptions starting with ticket splitting (and this is where the Paxton stans and MAGA diehards need to be honest with themselves). If the margin ends up being only 120,000, do you really think there aren't 60,000 Texans who will vote for Abbott for governor and for Talarico for senate? I personally know 20+ people who have NEVER voted for a Democrat who are voting for Talarico. MAGA will just deride them as crypto-Democrats but that's insanity when these people have spent their entire lives voting over and over for Republicans. There were half a million people who showed up to vote for Cornyn in the runoff. Paxton's people and MAGA in general have shown that portion of the party utter contempt for years and did jack shit to try to bring them in since Paxton won the runoff. Do you really think that ALL of those people are going to show up for Paxton? From my perspective it seems like the doubling down on "Bush/Rove/Cornyn Republicans are all RINO traitors" was a really bad message when you need these people to vote for you. And that all assumes that the people who are general election only voters and first time voters are completely immune to the national mood. It strains credulity to think that there's been this enormous uptick in Democrats voting but the people without primary voting history are just evenly split. Given how large of a group that is, a 3% advantage for Democrats in that group would be the entirety of the GOP advantage using the percentages I calculated above. And before you say "but historically that group has been 60/40 Republican," Abbott crushed Beto in 2022 with 4,435,558 votes. There were 2,817,822 R1-R4 who voted in that election and 3,063,760 general only and no history voters. There were 135,624 mixed-primary-lean-Republican voters. 4,435,558 − (2,817,822 + 135,624) = 1,482,112 So if Abbott got approximately 1,482,112 votes from the general only and no history voters then the percentage of that group he got was: 1,482,112/3,063,760=48.38% In other words, a 3% advantage for Dems in that general only and new voter makeup is not only reasonable, it might be a conservative estimate. If Democrats have a 3.3 advantage and Republican ticket splitting minus Democrat ticket splitting is only 1% in favor of Democrats, Talarico wins by around 70,000 votes. Based on what I've seen from some other data, it looks more like that 1% is actually about 3%. If it is, then Talarico wins by around 200k votes. To be fair, a lot of what I hear about the ticket splitting is anecdotal. I've only seen a couple data driven pieces on ticket splitting with the most egregious being that 1 in 9 Abbott voters are voting for Talarico and that 1 in 16 Hinojosa voters are undervoting Talarico. Note the difference--An Abbot voter switching to Talarico at the senate race is -1R and +1D to the totals while undervoting Talarico is -1D. I personally think that the ticket splitting issue is much worse for Paxton than his supporters do (although that's a low bar since they don't seem to think ticket splitting exists). I'm also not saying it's impossible for the general election only and new voters to break for Republicans, but I don't really see an explanation for it when their turnout is as elevated as it is relative to 2022. Somehow there would need to be an explanation for why Republicans who never vote in primaries are suddenly motivated to vote at higher rates than those who do vote in primaries. An interesting possibility is whether or not the election turnout totals are actually more inelastic than the current high turnout indicates. If final turnout totals actually trend towards a single point per segment of voter then we should expect to see a drop off somewhere soon. I suspect that is NOT the case as if it were then previous elections would show a variety of shapes of daily voter turnout rather than the vaguely tan(x) shape we see over and over again. Prediction time: Talarico wins by more raw votes than Cruz beat Beto in 2018. I think my R1-4 and D1-4 numbers are in the ballpark but I think that ticket splitting at the senate race costs Paxton about 100,000 votes (around 2.5% what should be his voters) about half of which get added to Talarico and that the elevated general only and no history turnout indicates the same enthusiasm amongst Democrats with no primary history as with D1-4 and the actual split in that group significantly exceeds a 3% tilt to Democrats. To sum up, if you think that Paxton is winning you have to stick to the following assumptions: -Republican voter turnout is going to accelerate at rates higher than they are already -There is no such thing as an Abbott/Talarico voter -The people who are general election only and new voters are not only immune to trends but in fact going to shift more Republican than they were 4 years ago in a year where Democrats are currently nearly doubling their turnout from 4 years ago Keep in mind that this is after more than a quarter BILLION dollars have been spent to prop up Republicans in Texas. So my question to the hardcore MAGA voters is this: is picking the literal worst person possible in the primary and losing with them after spending millions which could have been spent in other races worth removing someone who votes with you 99% of the time? I think that's the worst trade deal in the history of trade deals. All of this analysis is from looking at @longhornderek's excellent work: mcusercontent.com/d3064a2fad…