ELECTION NIGHT GUIDE: This guide will take you hour by hour through the key things to watch on election night.

KEY: For each state, I have given three numbers. The first (RCP) is the average of recent polls on realclearpolitics.com, a simple mean of all reputable polls from different sources in the last week. The second (538) is from Nate Silver’s fivethirtyeight.com, a weighted, trend-adjusted average of polls based on date and reliability of pollster. The final is the % chance of an Obama victory, which comes from Nate Silver’s quantitative model. All polling data will be cited as a spread, with positive numbers (+5) indicating a 5% Obama lead and negative numbers (-5) representing a 5% McCain lead. These numbers are as of Sunday night before the election.

Make no mistake, Obama is definitely the favorite. To win, he needs to hold PA/NH/NM, and then win one of either 1) Ohio, 2) Virginia, 3) Florida, or 4) Colorado . Any of the other swing states (i.e. NC, IN, MO) would be insurance or icing on the cake.

6:00pm – Indiana the first tea leaf to read (IN, KY). Should Obama win Indiana , it would signal a landslide. A lopsided McCain victory could foretell weakness in Ohio .

In the first hour of the night, we get an early indication of Obama’s strength in a red state he is not expected to win, but where the action is supposed to be competitive (RCP -0.5, 538 -1.8, 27%). The risk here is making too much of the early returns without seeing where the votes are coming from. Indiana is an early voting state with about 20-25% voting early (slightly under the national average), but I cannot find any data on the party affiliation of these voters. Because Obama is well organized here, he may jump out to an early lead based on early voting.

Kentucky (RCP -13.5, 538 -12.5, 0%) is not in play, however major deviations from this would be an early read-through for Missouri, Virginia and North Carolina .

Bottom line: The 6:00 hour is probably a non-event, with McCain likely carrying a narrow IN lead into the next set of poll closings. If it’s going to be an Obama rout, however, you may see it early on in Indiana .

7:00pm – The battle for the old Confederacy (FL, VA, GA, SC, VT). Florida and Virginia are the first of the critical battlegrounds; if Obama wins either one it nearly guarantees a victory; African American turnout is key.

The night will whip into high drama shortly after 7:00 as we get a set of returns dominated by southern states, where the strength of black turnout will be tested.

Virginia (RCP +3.8, 538 +4.5, 92%) is perhaps the most intriguing contest of the night, and the lead bellwether of the entire election. From here, we will see whether Obama’s organization, ad spending, and strength in the polls actually translates into the voting booth.  If Obama is not slightly ahead after the first 10% counted, then McCain is probably outperforming his poll numbers.

Most of Florida is in the Eastern Time zone and will close at 7:00. The panhandle closes at 8:00, but we are likely to see results before then. Florida (RCP +4.2, 538 0.5, 50%) is a big early voting state, with as much as 45% of the state casting their ballot before election day. If the initial returns include early voting, where Obama is outperforming his party affiliation data, we will probably see Obama out in front by 3% to 5% initially. Biden has helped Obama here, and Palin has hurt McCain. Remember that a chunk of the Hispanic vote (17%) is Cuban and traditionally Republican.  Regardless of what the polls say now, this is a red state that McCain cannot win without.

Georgia (RCP -3.0, 538 -5.4, 7%) should be an interesting first take when the polls close. Obama is not expected to win here, but has closed the gap significantly. Don’t be totally shocked if the initial results are close to a tie or even putting Obama ahead. If Obama wins, he needs to have black turnout higher than those numbers and an in-line turnout among young voters. If McCain opens up a lead close to double-digits, it might be a sign that NC is not in play.

Finally, watch SC (RCP -10.0, 538 -12.2, 0%). If networks do not call it immediately for McCain, and the early returns are reasonably close, that’s a bad sign for Republicans and a positive read-through for Obama’s chances in NC. Vermont (RCP +25.0, 538 +21.6, 100%) is not in play; outperformance here will have limited read-through to NH, but this is not a particularly interesting state.

Bottom Line: By about 7:30, we’ll have a decent indication of how the night will shape up. If Obama looks to be winning either FL or VA, then PA in the 8:00 hour could be the clincher and we could be seeing the early signs of a decisive victory. If McCain is close in VA and holding serve in FL, however, then his upset hopes are alive.

7:30 Ohio joins the fray (OH, WV). We will barely get to digest the early numbers from FL and VA when OH closes its polls; an absolute must-win for McCain.

The quintessential swing state, Ohio will close its polls at 7:30. Although Obama is favored here (RCP +4.2%, 538 +3.0, 76%), any path to victory for McCain requires this state. In other words, while it’s a must-win for McCain it’s probably about third on Obama’s priority list of “knockout” states that would put him over 270 electoral votes (behind VA and CO). A nasty loss in OH, however, would not be a good sign for Obama in PA.

West Virginia (RCP -8.5, 538 -8.7, 0%) is not expected to be in question, but if Obama keeps it close it is another indication that his ground game is working. Also a read-through for central PA.

Bottom Line: Depending on how quickly the results are released following poll closings, we could actually know a fair amount by 8:00 ET. Obama gets several early shots at knockout punches in VA, FL, and OH. All three are likely to be close, and winning any one of them would put Obama in the driver’s seat for victory.

8:00 PM – Lots of action, but Pennsylvania looms the largest as Obama defends his flanks (PA, NH, MO, MI, CT, MD, DC, NJ, SD, TN, MA, AL, and some TX). All eyes will be on Pennsylvania, as it is McCain’s best hope for an upset stunner; Obama also plays defense in NH and MI; Missouri a red state in Obama’s sights.

Though from a poll perspective Pennsylvania is not the closest state (RCP +7.3, 538 +7.0, 96%), it is pivotal to a McCain upset bid and thus perhaps the most important tipping point contest. McCain has spent a lot of time recently here, hoping in part to sway registered Democrats (old Hillary voters) to cross over. Winning PA could for McCain buffer the loss of another swing state like CO or VA (but not both).

In reality, if Obama is winning Ohio early, then his chances in Pennsylvania look pretty good. We have no early voting data (I’m not sure if PA even offers it).

Missouri is an odd state. Despite being a classic bellwether and a virtual toss-up in the polls (RCP -0.7, 538 -0.7, 34%), it is unlikely to be a tipping point contest. An Obama win here would probably come as part of a landslide, and a McCain victory would simply be holding serve in a state he should win. Virginia and Ohio are likely to telegraph the results here. Like the Super Tuesday primary, this one could swing wildly over the course of the night, and it could be well past 10pm before we have a decent grasp on the results.

NH (RCP +10.7, 538 +8.3, 97%) should go comfortably for Obama, but this is a state that likes McCain and he has a very outside chance of swinging this rather unpredictable batch of voters. Michigan (RCP +13.0, 538 +11.3, 100%) is a former swing state that McCain gave up on, so this will go blue. NJ (RCP +16.2, 538 +11.7, 100%) might teeter for a minute, but that is a democratic state also. The remaining 8:00 states will be called within 20 minutes of the polls closing.

Bottom line: By the midpoint of the 8:00 hour, we’ll be in one of two places: 1) Obama could be looking to sew up victory, piecing together PA with one of VA/FL/OH; or, 2) McCain could be consistently outperforming the polls and potentially shutting out Obama in VA/FL/OH, forcing the election to a tiebreaker in the west (namely, CO, NM and NV). A PA shocker by McCain would mean Obama needs to sweep VA, CO, and NV (or simply win FL).

8:30pm – North Carolina is intriguing, but will it matter? (NC, AR) NC is another potential knockout punch for Obama, but we will likely have a good grasp on the outcome here based on early VA/ GA/SC results. Unlikely to be a tipping point state, but if Obama wins here, it’s game over.

North Carolina (RCP +0.3, 538 +0.6, 57%) is a deeply red state on the verge of turning blue, a fascinating study in demographic evolution this member of “the solid south”. On election night, however, it seems doubtful that the presidency will hinge on this state. NC has a huge amount of early voting, nearly two-thirds of the state, and we’ll see most of those results right away (expect a slight Obama lead early, based on relatively high number of democrats voting early, mostly African Americans). If Obama sweeps VA/NC, then he will have essentially won the election before polls even close on the west coast.

By 9:00, Democrats would also hope to have a Michigan called for Obama by the networks, and would want to at least be nursing leads in PA and NH.

Bottom Line: The half-hour between 8:30 and 9:00 is probably the most critical, as results from the key eastern swing states begin to stabilize. NC is unlikely to be the most crucial among them.

9:00pm – Go west, old man! McCain hopes to force a Game 7 (CO, AZ, NM, NY, RI, KS, LA, MN, NE, WI, AZ, SD, TX, WY). If the outcome is still in question by now, it means McCain has outperformed on the Eastern seaboard and is holding Ohio . If so, we play best of 3 for all the marbles: Colorado, New Mexico, and later on Nevada .

In the 9:00 hour, there will be a lot of balls still up in the air. It’s possible Obama may have already clinched the election with victories on the east coast – in which case the closing of polls in NY, NJ, and WI could put the networks on the verge of calling the whole election for Obama. More likely, however, we will still be watching the deciding results of a very close contest (VA/FL/OH/PA) that will determine whether or not McCain is still alive. If and when the focus turns out west, Colorado is the focal point.

Colorado is the Virginia of the west, a traditionally red state where Obama has built a lead in the polls (RCP +5.5, 538 +4.8, 91%), though early voting makes this state somewhat easier to defend.  A McCain upset win here would hinge on three things: strong evangelical turnout, weak younger voter turnout in Boulder , and strong performance among independent, white swing voters.

New Mexico is solidly leaning for Obama (RCP +7.3, 538 +9.3, 99%), but this is a state that has potential to swing wildly if Hispanic groups (42% of the electorate) and veterans unexpecedtly pull hard for McCain.

In the event that Obama is already well ahead, Arizona could be the ultimate “insult to injury” state. Obama put up an ad buy here as polls closed to single digits (RCP -3.5, 538 -6.0 , 2%

Bottom line: Pushing the election to the west is the best case realistic scenario for McCain. Colorado is probably Obama’s firewall and his insurance against an 0-fer on the east coast, but if it too topples then he is in real trouble, probably relegated to a coin flip outcome in Nevada.

10:00pm – Nevada hopes to recreate its West Wing fame (NV, IA, ND, ID, more KS, MT, OR, UT) In the fictional TV series, the Presidential election between Arnold Vinick and Matt Santos comes down to a nail-biter in Nevada . This is technically possible if McCain grabs a lead in Colorado . Should the election reach this point, however, a lot of things would have had to go wrong for Obama.

Nevada is a topsy-turvy political arena, driven by the politics of unions, Hispanics, and nuclear waste (no malicious word association intended). In other words, welcome to Las Vegas , because anything could happen.

Should the election still be close, you’ll also want to watch Iowa , which closes at 10ET. This one has already been put in Obama’s column, and by slamming ethanol in the debates McCain has both earned a bit of my admiration and effectively conceded the state.

In the case of an Obama blowout, see if he can run up the score in MT (RCP -3.8, 538 -5.0, 6%) and ND (RCP -1.0, 538 -2.2, 19%), which would be historic events if either went blue. Or, see if John King can entertain you with those really cool maps that he moves with his fingers.

Bottom Line: An election night that extends into the 10:00 hour means something has gone fundamentally wrong for Obama.

11:00pm – Senate races? Prop. 8, anyone? (CA, WA, OR, ID, HI, AK) The final wave of poll closings on the west coast bring little in the way of important data. OR was once thought to be in play, but will go handily for Obama. At this point in the night, we may be watching a nail biter in one or two crucial swing states. Or, we are pouring another drink and tallying up the Senate results (there is another big one in OR). The Proposition 8 measure on same-sex marriage in California will be a closely watched outcome nationally. Also, this may be the hour in which we see a concession speech from one of the candidates.

Happy viewing, and enjoy what should be a historic election night, regardless of the outcome. I may have to work late, but I will be up and running on gmail eventually if you want to give me a shout.

Cheers,

Drew