Part of former Rep. Beto O’Rourke’s appeal, should he run for president in 2020, is how well he performed for a Texas Democrat in his 2018 campaign for US Senate.
The pro-O’Rourke argument is pretty easy to make. He lost by just 3 points in 2018. No Democrat has come that close to winning a Senate seat from Texas since 1988. O’Rourke outperformed the political lean of the state by 10 to 15 points (depending if you use FiveThirtyEight’s or my “weighted average partisanship” calculation) when you look at a slew of elections since the beginning of the decade. That’s especially impressive for someone running against an incumbent.
What these statistics may not take into account is how quickly Texas has undergone a political shift thanks, in part, to President Donald Trump.
When you examine O’Rourke’s performance in the context of only recent political developments, it comes across as less extraordinary, in my opinion.
One of the better metrics to judge O’Rourke’s 2018 performance is comparing him to how the average Democrat did in Texas in 2018. These are candidates who all ran under the same political environment in the same state. O’Rourke did better than most, though not considerably so.
Yes, O’Rourke did about 11 points better than Democrat Lupe Valdez in the governor’s race. In the other two major statewide races (lieutenant governor and attorney general), Democrats did quite well relative to recent years. Mike Collier lost by 5 in the lieutenant governor’s race. O’Rourke’s 3-point defeat was only 1 point better than Justin Nelson’s in the attorney general race. Both Collier and Nelson were up against Republican incumbents like O’Rourke was.
Of course, these are all state offices, not federal offices. The congressional races in the state suggests something similar, though. O’Rourke 3-point loss in the Senate race was only 3 points better than Democrats did in the cumulative House vote in the state (taking into account races where one party didn’t run a candidate). In races involving Republican incumbents, O’Rourke did 4 points better than the average or median Democrat.
Nationally, Democratic Senate candidates did on average equally well as the average House candidate in their state in races involving Republican incumbents. Democratic candidates, on average, did about 2 points better than House Democratic candidates in open races as well as those involving Democratic Senate incumbents.
By this stat, O’Rourke did a little better than the average Democrat nationally, though not exceedingly so. If you want to see a truly impressive performance, look at Sen. Amy Klobuchar in Minnesota. She outperformed the average Democratic House candidate in her state by 13 points.
Still, you could make the argument that O’Rourke’s candidacy drove Democratic gains in Texas, which means these stats undersell his strength. That’s totally plausible.
A more likely explanation in my mind is that politics is becoming increasingly about opinions toward the President.
Hillary Clinton lost Texas by 9 points. O’Rourke did about 6 points better than that. In the five other races involving Republican Senate incumbents, the average Democratic candidate did 5 points better than Clinton. The median did 6 points better.
In other words, O’Rourke did no better than the average of this admittedly small control group.
Expanding it out, you see that the median Democratic candidate did 9 points better than Clinton in races with no elected incumbent and 12 points in races with a Democratic incumbent. That makes sense given that incumbents tend to do better than non-incumbents. On this measure, Klobuchar did over 20 points better than Clinton. (Notably, a regression controlling for Clinton’s performance, incumbency and interaction between the two does find a slight, though not great, O’Rourke overperformance like the House vote.)
Additionally, O’Rourke didn’t do considerably better than Trump’s approval rating suggested he should have, either. Trump’s net approval rating (approval rating - disapproval rating) was 0 points in the exit poll. If every Trump disapprover went for O’Rourke and every approver went for O’Rourke’s Republican opponent Ted Cruz, the race should have been even.
It’s makes sense that O’Rourke slightly underperformed what Trump’s approval rating would suggest because he was facing an incumbent. The median Democrat in Senate races involving Republican incumbents did either slightly worse or better than O’Rourke did relative to Trump, depending on how you measure Trump’s approval rating. Again though, there’s no sign O’Rourke did considerably better than Democrats nationwide in a similar position.
One thing I haven’t mentioned is money. O’Rourke may be able to raise a lot of it and fast. It’s certainly a strength that could help him in the primary. It won’t help him in the general election where both sides will have enough money no matter the feelings toward Trump.
And if you still don’t believe that O’Rourke’s 2018 and potential 2020 general election performance has a lot do with Trump, examine the latest poll from Texas. Quinnipiac University found the three Democrats with name recognition of above 70% in very similar positions against Trump for 2020. Former Vice President Joe Biden, O’Rourke and Sen. Bernie Sanders all trail Trump 47% to either 46% or 45%. Other lesser known candidates receive less support, but Trump is consistently at either 47% or 48%. Not surprisingly, Trump’s 47% is equal to his favorable rating in the poll.
Overall, O’Rourke is certainly stronger in Texas than you’d have expected of a Democrat five years ago. According to this data, so too are other Democrats. O’Rourke’s 2018 performance suggests he is certainly no worse than the average Democrat. He’s probably better. O’Rourke, though, isn’t likely some electoral magic bullet for Democrats.