Update: Welp, here you go. One last early return before bedtime. Looking good!

Sullivan 49%, Begich 44% after the 1st 70,489 votes counted in #aksen

— Greg Giroux (@greggiroux) November 5, 2014

Update: It’s approaching 1 a.m. and there are still no results from Alaska. If the GOP was stuck at 50 seats I’d be inclined to stick it out to see if they can reach a majority, but since they’re already sitting on 52 with Louisiana an all but certain 53, Alaska is gravy at this point. Here’s your thread to comment if you’re staying up late. The HA staff will be back with election blogging bright and early.

Update: It’s 12:15 a.m. on the east coast but the 53rd Republican seat is still in the balance. Follow Alaska returns at Ace’s Decision Desk or this site.

Update: Finally, Iowa is called. 51! Now we wait to see if Sullivan can build on it. If he can, with Tillis now likely to win in NC, that’s an eight-seat GOP gain. If Cassidy wins next month, that’s … nine. And that means that even with a favorable Senate map in 2016, Dems will have their work cut out for them in trying to take the Senate back.

Update: Holy schnikes. The Upshot’s adjusted lead has Pat Roberts up by, if you can believe it, 10 points.

Update: Done deal.

PROJECTION: Republican incumbent Pat Roberts is re-elected in the Kansas Senate race pic.twitter.com/BQmCa7FRwZ

— CBS News Politics (@CBSPolitics) November 5, 2014

I thought Roberts would win, very narrowly, but if they’re making a call within an hour after the polls closed, there was nothing narrow about this. In the end, the party unified and Republicans came home. Congrats to the whole GOP team on dragging Roberts over the finish line.

Assuming Joni Ernst wins in Iowa, that’s 51. Tillis can make it 52 if he comes through and Dan Sullivan can make it 53 in Alaska in a few hours. A month from now, Bill Cassidy will make it 54 in Louisiana. And Ed Gillespie still has a prayer at making it 55 in Virginia. Oh my!

Update: The Decision Desk is now claiming that Thom Tillis is going to shock Kay Hagan in North Carolina. We’ll see. The Upshot, as I write this, is expecting Tillis to win by one-tenth of a point, but that may change. Gettin’ wavy out there, though.

Update: And no sooner do I finish writing the last update than we have this:

Channel 13 projects @joniernst has won the US Senate race http://t.co/cYao0lW2Oy #iasen pic.twitter.com/VF8ICTBCsK

— WHO 13 News (@WHO13news) November 5, 2014

That’s 51, with Tillis and Gillespie still hanging in there — presuming all of these early calls are accurate, of course. Stand by.

Update: This is a verrrrry early call by Ace’s Decision Desk, but if they’re right then the GOP’s in position to hit 51 tonight.

#KSSenate : Roberts

— Decision Desk HQ (@DecisionDeskHQ) November 5, 2014

A Joni Ernst victory in Iowa would do it. Stay tuned.

Update: For old time’s sake. Good luck coping with the condom shortage, Coloradans!

Update: At just before 10 p.m. ET, Fox News calls Colorado for Cory Gardner. Alas, you guys will have to wait for Mitt Romney to definitively rule out running before you get to see me in full eeyore freakout mode.

I’m looking forward to Senator Uterus’s concession speech, “The Threat to Vaginas.” Until then: It’s time.

Update: With just about a million votes in, Gardner’s lead is still solid at more than seven points. Time for Humpbot to warm up.

Update: Starting to feel it: Harry Enten says Gardner is (slightly) ahead of where he needs to be in three key counties.

Update: With a bit more than 800,000 votes reported, Gardner leads Udall by more than seven points. Waiting for The Upshot’s “adjustment” to see if he’s performing better or worse than expected early.

Update: Projections a little while ago had both Mark Warner and Kay Hagan eking out wins, but, er, it’s getting late in Virginia and North Carolina — and Ed Gillespie and Thom Tillis are both still leading. Stay tuned.

Update: One footnote on Texas. The schadenfreude — oh, the schadenfreude:

https://twitter.com/kherman/status/529818291603116032

Update: Results are coming in now in Colorado. Ace’s Decision Desk is down for the moment but you can follow the returns at the Secretary of State’s page.

Update: 9 p.m. ET and the polls are finally closed in Kansas and Colorado. Not surprisingly, both states are too close to call. Not too close to call: Texas, where Wendy Davis has finally been put out of her misery. See you on MSNBC.

Update: Put on some coffee.

Wow, about 75,000 votes in in Kansas and Roberts and Orman are separated by 8.

— Molly Ball (@mollyesque) November 5, 2014

Update: Almost 9 p.m. on the east coast and the only race where the GOP seems to have a real shot at an upset is in Kansas, where Roberts is barely an underdog. Taking back the Senate is the goal, of course, but it’ll be hard to call this a wave with Brown, Tillis, and Gillespie all apparently headed to defeat.

Update: Merry election Christmas from Ron Paul:

Republican control of the Senate = expanded neocon wars in Syria and Iraq. Boots on the ground are coming!

— Ron Paul (@RonPaul) November 5, 2014

Update: Another good early sign from Kansas via Harry Enten: With the first few precincts in, Roberts is running six points ahead of Sam Brownback. One big GOP fear was that Brownback’s unpopularity would drag Roberts down among Republican-leaners disinclined to split their ballot. Lots of splitting so far, though.

Update: The first returns from Kansas are tricking in as I write this. The early word from North Carolina and New Hampshire, though, is that Hagan and Shaheen are looking good early. If that holds, then the GOP needs either Roberts or Sullivan — Kansas or Alaska — to come through if they’re going to get to 51 seats tonight.

Update: Mega-troll idea for Obama: Issue that executive order on amnesty tonight as soon as the 51st seat is called for the GOP. Shazam.

Update: Contra Klein, GOP operatives in Kansas tell the Corner that they’re happy with early voting among Republicans, who made up 55 percent of all ballots. Their target is 60 percent of the electorate overall for Roberts to hold off Orman.

Update: Uh oh. From Rick Klein of ABC: “meanwhile – GOP turnout at 20-year low in #KSSEN. 2/3 say Roberts away from state too much, in prelim numbers.”

Update: A good-looking trend for Cory Gardner. Turnout in lefty areas is down, turnout in Republican areas is up:

Colo. perspective: In 2010, 125k total votes in Boulder Co.; 190k Denver, 194k El Paso. Noon today: Boulder 111k; Denver 170k; El Paso 196k

— Joshua Green (@JoshuaGreen) November 4, 2014

Update: A little extra motivation for wayward conservatives in Kansas, thanks to Uncle Joe Biden: “Pat Roberts campaign is now using Biden’s remark about Orman in 1 million GOTV robocalls hitting Kansas households.”

Update: For what it’s worth:

I'm not vouching for this. But in top Dem circles today, there's sudden optimism about Begich.

— Greg Sargent (@ThePlumLineGS) November 4, 2014

Update: Here’s that Upshot Senate adjustment tracker that I mentioned at the end of my post. Groovy.

Update: Did Mark Udall blow a chance to rally Latinos in Colorado by obsessing over reproductive issues? BuzzFeed says yes:

BuzzFeed News has learned that reliable polling in Colorado is suggesting that voters were not well-informed on the distinctions on immigration stances between the two candidates. The information suggests that Udall could have done a better job defining himself and Gardner on immigration among a sizable share of Latino voters.

Update: FiveThirtyEight and The Upshot now have nearly identical odds of the GOP taking back the Senate. The former has it at 76 percent, the latter at 75 percent.

Update: I hate to disappoint so early in the day, but you might not get to see me go Rain Man after all:

Plugged in CO source "freaking out" … Clarifies in a good way.

— Guy Benson (@guypbenson) November 4, 2014

***
Three fun races for you here — one a probable and potentially significant GOP win, one that could end in a near dead heat with control of the Senate in the balance, and one a “who the hell knows, it’s Alaska.”

The probable win belongs to Cory Gardner, of course, who’s set to bounce Mark “Uterus” Udall in Colorado. Gardner finished the race with a 2.5-point lead in the RCP poll average. He’s a 72 percent favorite in Nate Silver’s election-matic. Lotta virtual ink has been spilled on Colorado, including by me, so you should know by now what the stakes are in this one. Udall decided he was going to run an all-out “war on women” campaign against Gardner based on the latter’s previous support for “personhood” initiatives. When Gardner countered that by coming out in favor of making birth control more freely available, i.e. over the counter without a prescription, Udall decided to … keep right on going with the “war on women” stuff, come what may. No foolin’:

Holy wow. RT @rickklein: 47% of ads in #COSEN race referenced abortion, per @CMAGAdFacts and @c_good

— Reid Wilson (@PoliticsReid) November 4, 2014

As of this morning, Gardner has a lead of more than seven points among ballots in Colorado that have already been mailed in. Udall will cut into that tonight but it’s unlikely he’ll make up seven percent. If Gardner holds on and wins, it’s a big deal — he’ll have ousted an incumbent in a perennial battleground state while repelling the Democrats’ favorite demagogic line of attack. His campaign operation will be studied by Republicans nationwide as a model of outreach. With the possible exception of Joni Ernst, he’s the brightest star in this year’s constellation of GOP candidates. If we can’t win in Colorado despite all that, I’m going to start punching myself in the head and screaming like Rain Man when Tom Cruise tried to make him get on that plane. So, win or lose, you’ll want to watch this thread.

The dead heat that may decide who controls the Senate is Kansas, where, ahem, “independent” Greg Orman leads Pat Roberts by just 0.8 points. Orman is a 53 percent favorite in the FiveThirtyEight election-matic, the smallest edge of any Senate frontrunner in the country. (Kay Hagan is a distant second at 69 percent.) Enough ink has been spilled on this race that you should already know the contours here too. Thanks to a bitter primary with tea-party favorite Milton Wolf and deep disaffection among Kansas’s GOP establishment with Sam Brownback, Roberts is struggling to patch together a “50% + 1” coalition in a red state that should have been a walkover for the GOP. It was a three-way race originally, you’ll recall, but Democrat Chad Taylor dropped out when Orman began to surge so that lefties wouldn’t end up splitting their votes. Orman insists he’ll be a true independent in the Senate, to the point where his campaign website claims he’ll “seek to caucus” with whichever party holds a clear majority, but oddly enough, his campaign is run by Democrats. There are Greg Orman campaign signs on the walls at Democratic headquarters in Kansas. Despite voicing a few orthodox Republican positions, Orman conspicuously faces no attacks from Democrats because of them. And the vice president, himself a Democrat, is convinced that Orman will caucus with the Democrats. The guy’s a Democrat.

And, I assume, pretty much everyone in Kansas knows it by now. If he wins, it’ll likely have less to do with him duping Kansans into believing he’s some sort of uncategorizable centrist than with Roberts drowning in a rising tide of disgust with establishment Washington — he doesn’t even have a home in his “home” state, remember — and local disgruntlement with the GOP powers that be in Kansas. The match-up here, depressingly, is between a guy who’s carefully refused to take a position on nearly every issue lest it cost him any “Anyone But Roberts” votes and a guy who went native inside the Beltway decades ago. The one shining reason to root for Roberts is because, as I explained earlier, this race may be our only chance of clinching a Senate majority tonight. If Orman wins and, as expected, Hagan and Shaheen hold on in North Carolina and New Hampshire, then it’s Alaska or bust to get to 51 by tomorrow morning. (We’ll get to 51 via Louisiana eventually, but not until December.)

Which brings us to Alaska. Republican Dan Sullivan has led Democratic incumbent Mark Begich in nearly all of the last 12 polls, sometimes by as many as six or seven points. His lead in the RCP average settled at 2.4 points, a margin almost as wide as Cory Gardner’s. He’s a 74 percent favorite in the election-matic. It’s a reliably red state and Begich, the 60th vote for ObamaCare, has spent much of the campaign laughably insisting he’s some sort of thorn in Obama’s side. This is a gimme for the GOP, right? Well … yes, probably. But Alaska polling is so confounding that even pros like Nate Silver and Harry Enten feel compelled to break out the asterisks when modeling the state. How confounding? Within the last few weeks, one poll had Sullivan by four, another poll had the race tied, and two more polls had Begich ahead by seven points or more. Anything can happen tonight, which is nerve-wracking; the good news is that when Alaska polls miss big, they tend to miss by overestimating support for Democrats. That hopefully explains those freaky Begich blowout margins.

The polls close in Kansas at 8 p.m. ET, in Colorado at 9 p.m. ET, and in Alaska at, er, midnight. I won’t be here all night kibitzing about Sullivan and Begich but the thread will be in case you want to. Lots of updates to come at the top of this post until I drop, though. There are endless resources online for following the returns in case you’re not satisfied watching Fox News and CNN. I’ll be using Ace’s Decision Desk, which is going to try to beat professional outfits like the AP to the punch by crowdsourcing the returns on every major race. They called elections accurately before the pros did more than once during primary season. If you hate big media monopolizing returns, give Ace’s gang your business tonight. Beyond that, Silver’s site, FiveThirtyEight, and the NYT’s competitor, The Upshot, will be all tricked out for the occasion. The Upshot is promising to post real-time adjustments to the likelihood of victory in each race as results come in based on expected partisan splits, which could be interesting. In other words, if Colorado’s heavily Democratic urban centers report first and Udall jumps out to a big early lead, the Upshot’s tracker won’t show any change in Gardner’s odds of winning unless Udall is performing better than expected in those areas. Checking that tracker is a way to tell at a glance whether the race is going in a different direction than experts expected, regardless of what the vote count at a given moment says.

Cross those fingers. Updates on the way.