Assessing the reasons people vote the way they do is always fraught. After an election, we have a big batch of data that can be parsed in many ways, leading to multiple logical conclusions about why things turned out the way they did — often, not mutually exclusive ones.

Such is definitely the case with Georgia’s Senate runoff elections, which as of Wednesday morning look primed to hand control of the chamber to the Democrats. Democrat Raphael Warnock defeated Sen. Kelly Loeffler (R), and Democrat Jon Ossoff is leading Republican David Perdue, whose Senate term ended Sunday, with the outstanding votes mostly in strong Democratic areas. Ossoff declared victory Wednesday morning, although Perdue vowed to fight on.

Winning both seats would mean Democrats would control the trifecta of the presidency and both chambers of Congress — albeit with a 50-50 Senate (where ties would be broken by Vice President-elect Kamala D. Harris) and a very slim House majority.

As with everything in the Trump era, one of the big immediate questions is what role the ever-present president played.

Plenty of Republicans fretted in recent days and weeks that President Trump might cost them the Senate. They suggested Trump’s continued, baseless undermining of the legitimacy of the state’s elections would harm GOP turnout. If Republicans believed the whole thing was rigged — as most of them did, according to polls — why even show up? They also worried, on a smaller scale, about Trump’s rather needless highlighting of the congressional GOP’s opposition to $2,000 coronavirus relief checks, briefly reneging on a deal agreed to by his administration and allowing Democrats to seize on the issue on the eve of the elections.

Another plausible theory, which will be pushed by Trump’s supporters in the hours to come, will be that this was actually a validation of Trumpism — in that, without him on the ballot, Republicans didn’t have as much of a reason to vote. They’ll argue it’s a reason to keep him in the mix rather than try to turn the page.

I’m more inclined toward the former theory, although both very likely played some role.

It is indeed quite possible that Republicans didn’t turn out as much because Trump wasn’t on the ballot. Supporters of this theory will point to the 2018 midterm elections. He wasn’t on the ballot then, and the GOP lost the House.

But the thing is, that’s pretty par-for-the-course. Midterms are routinely bad for a president’s party, as plenty of history shows. What’s more, the shift in the House wasn’t borne out in the Senate. Republicans gained ground there, which Trump and his allies made pains to highlight to rebut claims that the election was a rebuke of him. (Nevermind the fact that the map of seats that were up was very favorable to the GOP.)

And there are two main reasons to believe these weren’t races that the GOP should lose, even with that history and the GOP base’s devotion to Trump in mind.

One is that this went against the historical trend. As I wrote Tuesday, Republicans have routinely overperformed in Georgia runoffs, relative to November elections. Across 10 Republican-versus-Democrat runoffs since 1992, Republicans won nine of the 10 and improved their vote shares in eight. They won both of the Senate runoffs in that span, in one case overturning a deficit on Election Day and in another case winning in a blowout.

That latter race, which was in 2008, is also instructive when it comes to my second point. That race was pivotal for the Senate — not when it came to control of the chamber, but because it decided whether Democrats would have a filibuster-proof majority with 60 votes. Democrats would have had that, a huge House majority and the presidency, by virtue of Barack Obama’s win. Georgia voters decisively rejected that proposition of overwhelming Democratic control of Washington, with Sen. Saxby Chambliss (R) winning the runoff by 12 points.

For whatever reason, though, a somewhat similar calculus didn’t help the GOP nearly so much in 2020 (and now 2021). The GOP had plenty of reason to fight hard to win these races and a very good argument to turn out the vote, given this was its last hope of holding one of those three levers of power in Washington. Even if you’re less keen on the congressional GOP than on Trump personally, that would seem to be a pretty good reason to show up. Even if you think (without evidence) that Trump actually won, it’s your insurance when it comes to protecting his legacy.

Despite all that, though, the GOP did worse than two months ago. In both Senate races, Republicans took more votes the first time around — by about 1 point in the Warnock race and 1.8 points in the Ossoff race. The races were close — as was the presidential contest — but they were fighting from a winning position. And they underperformed Election Day in a way they almost never do.

It’s quite possible Trump’s actions weren’t determinative. Maybe Democrats were just more jazzed about voting relative to Republicans than last time around, even if that hasn’t generally been the case in the state. Perhaps the state’s newfound purple hue has changed things. Maybe this was a fluke attributable to expanded mail-in voting.

But what’s clear is that the GOP had the opportunity to focus on a compelling argument in these races. Even if Trump didn’t dissuade enough GOP voters with his baseless voter fraud claims to make the difference, he certainly muddied that message by keeping the focus on something else: himself. Even Monday night, in his 11th-hour rally in the state, he pitched the state’s GOP senators as “truly the last line of defense” but then quickly corrected himself by suggesting he might still be in office to be that last line of defense.

We’ll never know what would have happened if Trump had conceded and kept things focused on that more-compelling, Senate-focused message. At the least, he turned his supporters’ attention to something far less productive. This, at best, could have given them a false sense of security about the true immediacy of electing Republicans to these seats and, at worst, might have convinced enough of them that they couldn’t trust their votes to be accurately counted.

(Given the small margins, the latter is among many plausibly decisive explanations, given it need only have affected a small portion of voters. Trump allies, including his former lawyer Sidney Powell and L. Lin Wood, expressly urged Republicans to boycott if GOP officials in Georgia didn’t overturn the presidential result.)

All in the name of a doomed-to-fail and self-serving attempt to change the results of a presidential race that has been over for many weeks — a fantasy that his party, perhaps now to its significant detriment, decided it had to entertain.

And if, as appears quite likely, the GOP loses the Perdue seat, one big thing will be tough for the GOP to look past: Trump came into office with all three levers of power in Washington under GOP control. He will leave with none of them. Whatever specifically cost Georgia for the GOP, it will have lost Washington on Trump’s watch.