As you watch the election results come in tonight, you’ll notice that they don’t actually come in a predictable order. Loose races can be called earlier than tight races, and the ballot-counting is more efficient in some states than others. So while I’ve arranged the following guide to today’s race in order of when the polls close, the results in the races discussed will not necessarily be reported in this order.

Calls for House results are in bold, while calls for Senate results are in bold and italics. All Times Eastern Standard.

6:00 PM — Polls close in the portions of Indiana and Kentucky that are in the Eastern Time Zone. Kentucky’s 3rd and 4th District are both entirely in the Eastern portion of the state, and both are held by vulnerable Republican incumbents.

In Kentucky-4, conservative Democrat Ken Lucas is back in the race after leaving the seat to fulfill a term-limits pledge. Essentially, he and Republican Geoff Davis both have some of the advantages of incumbency. Polls have pointed in both directions, but I give a slight edge to Davis (Rep. hold). The polls have really been all over the place in Kentucky-3, which includes the Louisville area, where Anne Northup (R) is facing a challenge from John Yarmuth (D). The Louisville Courier-Journal poll, a traditional poll in which calls were made by human operators, showed Northup winning handily, while Yarmuth has led in the so-called robopolls by SurveyUSA and RT Strategies/Constituent Dynamics, where questions are asked by recording and answers are given by pushing buttons. Since the robots outperformed the humans in the last cycle, I’ll call this one for Yarmuth (Dem. gain). If I’m wrong: If KY-3 and KY-4 both change hands, the Democratic wave that some have predicted may be coming. If neither changes hands, Team Dean may be in for yet another disappointing night. And if there’s an upset in IN-7, the Democratic takeover instantly becomes much less likely.

7:00 PM — Polls close in the rest of Indiana and Kentucky, the Eastern Time Zone portion of Florida (that’s everything but the panhandle), most of New Hampshire (it varies by precinct), and statewide in four other states, two of them with hot races: Georgia and Virginia.

In the Virginia Senate race, George Allen faces Jim Webb, and they’ve been close to tied in most polls. The latest SurveyUSA poll, shows Webb surging ahead, but it’s so far out of line from other polls that have shown the race a tie that I’m inclined to discount it as an outlier. With low confidence, I’ll predict an Allen victory (Rep. hold). If I’m wrong: A Webb win would indicate that the dysfunction of Allen’s lackluster campaign extends down to the get-out-the-vote (GOTV) level.

In the House, Georgia’s 8th and 12th districts may see a rarity in this cycle: Republican gains. I’ll predict that Republican challenger Mac Collins wins in GA-8 (Rep. gain) while Democratic incumbent John Barrow hangs on in GA-12 (Dem. hold), but I wouldn’t be surprised to see those reversed, or to see Republican victories in both districts.

Republican incumbents are in serious danger in Indiana’s 2nd, 8th, and 9th districts. I expect them all to be defeated (Dem. gain times 3). If I’m wrong: If Republican GOTV operation can pull out poll-defying victories in Indiana, this could be a sign of more to come as the night wears on.

In Florida, my hunch is that Republican GOTV can overcome the marginal Democratic leads in the 13th and 22nd districts (Rep. hold times 2). In Florida’s 16th district, Republicans won a victory when courts allowed them to post a sign at polling places that makes clear that a vote for Mark Foley is a vote for Joe Negron. Will it be enough? I’m going to go out on a limb and say that it will (Rep. hold). If I’m wrong: The sort of massive Democratic blowout that people like Stuart Rothenberg predict is only possible with victories in Florida (though still not probable).

7:30 PM — Polls close in West Virginia and Ohio, as well as in many precincts in North Carolina. No incumbents of either party are in real danger in West Virginia. The same can’t be said for Ohio, where a Republican party that has been plagued by scandal is headed into a buzzsaw.

In the Senate race, Mike DeWine is going to lose to Sherrod Brown (Dem. gain). Republicans will also face congressional losses in Ohio: certainly in the 18th district, where Rep. Bob Ney resigned following revelations of his involvement in the Abramoff scandal, and probably in Districts 1 and 15, as well (Dem. gain times 3), though Jean Schmidt can probably pull through in District 2 (Rep. hold). If I’m wrong: Upsets in Ohio would mean that Republicans have a serious shot at holding the House.

8:00 PM Polls close in the rest of Florida and New Hampshire, as well as in the Central Time Zone portion of Texas, which includes the 22nd District, where Tom DeLay’s late resignation left the ballot without a Republican; I don’t believe that Republicans rallying around Shelley Sekula-Gibbs can overcome the problem of the cumbersome write-in process (Dem. gain). (Polls also close now in the Eastern Time Zone portion of Michigan and the Central Time Zone portion of South Dakota.) Polls close statewide in over a dozen states, seven of them with hot Senate or House races: Connecticut, Maryland, New Jersey, Tennessee, Missouri, Pennsylvania, and Illinois.

In New Hampshire’s 2nd District, Paul Hodes is making a surprisingly strong challenge against incumbent Republican Charlie Bass. Bass beat Hodes by 20 points when they faced off two years ago. But the financial advantage that belonged to Bass then belongs to Hodes now, Hodes is leading in most of the polls, and Bass seems to have been caught off guard: The Evans-Novak Political Report has reported Republican grumblings about “lackadaisical staffers” and “extremely disorganized” GOTV. All that adds up to a Dem. gain.

Joe Lieberman is squashing Ned Lamont in Connecticut, which counts as a Dem. hold (I believe that Lieberman will keep his word and caucus with the Democrats). But some Nutmeg State observers say Lieberman’s independent run may have coattails for struggling Republicans. If this is true, then Republicans may defy the polls and hang in there in Connecticut’s 2nd, 4th, and 5th districts (Rep. hold times 3). If I’m wrong: As in Florida, Connecticut victories are necessary for a true Democratic blowout.

In Maryland, Michael Steele has run one of the best campaigns in the country, running even with Democrat Ben Cardin in a state where Democrats enjoy a two-to-one registration advantage. (It’s not hard to see why: Even in his worst nonpartisan poll in the past two months, a Washington Post poll that showed him down by 11 points, 59% of respondents thought that the statement “He has an appealing personality” applied more to Steele than to Cardin.) I wouldn’t be at all surprised by a Steele upset, but in the end it looks like this just isn’t the right year. Dem. hold. In New Jersey, the momentum that Tom Kean seemed to show earlier in the fall has stopped showing up in recent polls, and New Jersey isn’t a place where Republicans are known for pulling off upsets. Dem. hold. If I’m wrong: If there’s an upset in either of these races, it’s difficult to see how the Democrats can possibly take the Senate.

In the Tennessee Senate race, Democrat Harold Ford, Jr. hasn’t led in a non-partisan poll in almost four weeks, and I expect Bob Corker to win (Rep. hold). The Missouri Senate race is as close as they get, and a victory for Democrat Claire McCaskill would be no surprise. With low confidence, I predict that Jim Talent pulls out a squeaker (Rep. hold). If I’m wrong: If Tennessee and Missouri flip, a Democratic Senate is well within reach.

In Pennsylvania, Rick Santorum is almost certainly going to lose his Senate seat (Dem. gain). Probably going down with him: Incumbent Republicans in Districts 6, 7, and 10 (Dem. gain times 3). If I’m wrong: If any of Pennsylvania’s most vulnerable Republicans hold on, thank Santorum’s campaign– even if his GOTV effort can’t save him, it might save others. On the other hand, if you’re on the lookout for signs of a real GOP disaster, watch out for losses in PA-8 or PA-4 (both of which I predict as Rep. holds).

The most closely watched race in Illinois is in the 6th district, the open seat vacated by Henry Hyde, where the polls would suggest a toss-up. But DuPage County, which accounts for 4/5 of the district’s votes, is the nerve center of the Illinois Republican Party, and I’ll bet they know enough about turning out voters in their own back yard to hold this seat (Rep. hold). If I’m wrong: If the Illinois GOP falters in IL-6, watch out for them to falter next door in IL-10. Though the 10th has flown under most analysts’ radar, the one nonpartisan pollster following it (RT/Constituent Dynamics) shows a close race, and while Republican Mark Kirk has a two-to-one financial advantage, that’s quite a bit less than the advantage he’s had against his last two challengers (both of whom he beat with totals in the 60s, even as the district gave 53% to John Kerry).

8:30 PM — Polls close in Arkansas, which has no hot races at the federal level, and in the rest of North Carolina, which does: Charles Taylor, while not to be confused with the Liberian warlord awaiting trial in The Hague, is one of the most vulnerable Republican incumbents, has trailed in every poll, and will probably lose (Dem. gain).

9:00 PM — Polls close in the rest of Michigan (and Texas and South Dakota), and close statewide in 10 states, six of them with races worth mentioning: Arizona, New York, Rhode Island, Minnesota, Colorado, and New Mexico.

I’m not even sure why the Michigan Senate race is considered a contest in some quarters, but I’ll mention it anyway: Dem. hold. The same goes for Arizona’s Senate race: easy Rep. hold. The House races in Arizona are more interesting: the 8th is looking like a Dem. gain, but I bet J.D. Hayworth pulls it out in the 5th (Rep. hold). If I’m wrong: Republican losses in both Arizona districts will tip immigration politics to the left, even if Republicans somehow manage to hold the House anyway.

As Hillary Clinton cruises to victory in New York, every upstate Republican seems on the defensive. Look for GOP losses in Districts 20 and 24 (Dem. gain times 2).

In Rhode Island, though the Senate race is tightening a bit, Sheldon Whitehouse is still set to replace Lincoln Chafee with a liberal who isn’t afraid to go by the name “Democrat” (Dem. gain). If I’m wrong: If Chafee is re-elected, don’t be shocked if he switches parties at some point.

In Minnesota, Mark Kennedy is facing a loss in the Senate race (Dem. hold), but Republicans are holding on in their House races. If I’m wrong: I’m not predicting it, but it isn’t inconceivable that Minnesota Republicans could be so dispirited by the state of the Senate race that they stay home and hand a seat like MN-1, where the Republican is leading but only by a few points, over to the Democrats.

Colorado’s 7th district looks like it will be an easy Dem. gain. RT Strategies/Constituent Dynamics shows CO-4 as a toss-up, too — but they’re a new pollster using a fairly untested method, and Mason-Dixon, which has been around long enough to build an excellent record, shows Republican Marilyn Musgrave out in front. I trust Mason-Dixon (Rep. hold).

New Mexico’s 1st district is another probable Dem. gain.

10:00 PM — Polls close statewide in four states and in parts of three others, but the ones to watch are Montana and Iowa.

Democratic challenger Jon Tester has been ahead in the polls in Montana’s Senate race for months. The race has tightened, with Conrad Burns tied in the latest Mason-Dixon poll and within the margin of error in the latest Rasmussen poll. I’m cautiously betting that Tester still has the edge here (Dem. gain). If I’m wrong: This is another race that is all-but-critical to any Democratic Senate takeover. It’s hard to see them running the table everywhere else if they lose here.

Iowa’s first district is another likely Dem. gain.

11:00 PM — Polls close in every state left (except for Alaska, which has no seriously contested national races). The one to watch now is California’s 11th District. It’s very close, but I’m guessing it’ll be another Dem. gain.

So in the Senate, that makes 4 Democratic gains; in the House, 21 Democratic gains and 1 Republican gain. Counting Lieberman and Bernie Sanders (the socialist independent who is poised to win in Vermont) as Democrats, that make my final prediction a Republican Senate divided 51-49 and a Democratic House divided 223-212.