So, how’s tomorrow going to turn out?

There are a huge number of uncertainties, of course. Perhaps Hillary’s supporters will remember that their candidate is a corrupt criminal, that she has no agenda other than serving her own ambition, and that returning her to power enshrines elite politics at the top level. Perhaps they don’t show up in the numbers Hillary requires.

Perhaps Trump’s supporters have been hiding from pollsters. Perhaps the polls don’t really represent Americans’ feelings on this election. Perhaps a Brexit-style shocker is in order.

But now it’s time for some real-world predictions. And so we’ll break this one down, state-by-state, before we get to the popular vote numbers.

In order for Donald Trump to win this election, he must win North Carolina, Florida, Ohio, Iowa, and Nevada, as well as one more state: New Hampshire, Colorado, Virginia, Wisconsin, Michigan, or Pennsylvania (Minnesota isn’t happening). His ground game is shabby. His enthusiasm, however, is high. So, let’s survey.

North Carolina: Clinton. Thus far, 3 million ballots have been cast in North Carolina, more than the 2.8 million cast in 2012. Hispanic turnout is up, black turnout is down. Nate Cohn of The New York Times estimates as of October 31, “Already, about 2,892,000 people have voted in North Carolina, out of about 4,642,000 we think will eventually vote. Based on the voting history and demographic characteristics of those people, we think Hillary Clinton leads in North Carolina by about 6 percentage points.” By the same token, FiveThirtyEight gives Hillary just a 50.6 percent to 49.4 percent lead over Trump in probability.

Florida: Clinton. In Florida, early voting doesn’t tell us much. According to Politico, “Democrats increased their lead over Republicans in casting pre-Election Day ballots to nearly 33,000 as of Sunday morning.” Apparently, some 6.1 million ballots have already been cast, with Democrats leading Republicans by just 0.5 percent. But as Politico notes, Republicans vote more on Election Day – although independents are showing up in massive numbers. Hispanics are voting disproportionately in Florida, which bodes ill for Trump. Florida’s a true toss-up. The question will be the turnout game – and it’s hard to imagine that Clinton trails Trump there.

Ohio: Trump. CNN says that Hillary trails Obama’s 2012 numbers in Ohio. Early voting doesn’t look good for her. FiveThirtyEight estimates a 65.2 percent chance Trump wins the state. That’s largely because Ohio has a heavy population of non-college educated white voters, where Trump has a heavy electoral advantage. His Rust Belt appeal is real, and Ohio’s his state.

Iowa: Trump. Trump has a serious lead in Iowa. FiveThirtyEight has a 72 percent chance Trump wins the state. He’s been leading in the state for months. The latest Des Moines Register poll has Trump up seven points. Here are the last ten polls: Trump +7, Trump +3, Clinton +1, Trump +1, Trump +4, Tie, Trump +6, Trump +8, Trump +5, Tie.

Nevada: Clinton. Trump’s being slaughtered in Nevada in the early vote. According to Jon Ralston at KTNV.com, “About 770,000 votes have been cast, likely two-thirds of the vote. Let’s suppose that there is an Election Day turnout of 450,000 voters. Trump would probably need to win Tuesday by about 10 points to win. This is almost impossible, unless the Democrats decide not to turn out voters on Election Day.” FiveThirtyEight calls Nevada “relatively safe” for Clinton.

New Hampshire: Clinton. New Hampshire is incredibly close. Clinton has jumped back into a tie or the lead in the state after one week of Trump leading. Before that, Clinton had a massive lead in New Hampshire for months. Could Trump pull out the state? Sure. Is it unlikely? You bet. According to FiveThirtyEight, Trump’s got a 66 percent chance of losing the state.

Colorado: Clinton. Clinton is matching Obama’s numbers in early voting. On November 4, Republicans trailed Democrats in the early vote by 0.3 percent, with unaffiliated voters constituting 28 percent of the vote. But in early voting in 2012, Republicans led by 2 percent, and Romney lost the state by nearly five percent.

Virginia: Clinton. Hillary’s got an 82.7 percent chance of winning Virginia according to FiveThirtyEight. This state used to be reliably Republican, but Republicans haven’t won it since 2004 – before that, Republicans won it every election since 1984. That’s a massive loss for Republicans, who lose 13 electoral votes every cycle. Much of the shift is due to the influence of Washington D.C.

Wisconsin: Clinton. Wisconsin has seen heavy early voting, but the polling out of the state looks devastating for Trump. The RealClearPolitics poll average has her up a whopping 6.5 points.

Michigan: Clinton. Clinton has a 4.7 percent lead in the state according to the RCP average, and a 78 percent chance of winning the state according to FiveThirtyEight. It would be a stunning turn of events for Trump to pull out the state.

Pennsylvania: Clinton. Pennsylvania was the state upon which Trump staked his early hopes, but those hopes are unlikely to materialize. Not much early voting happens in Pennsylvania, but FiveThirtyEight gives Trump just a 23 percent likelihood of a pickup here.

Trump will likely win the second district in Maine. That means he finishes up with 217 electoral votes, slightly better than Romney – he loses North Carolina but wins Ohio and Iowa. If Trump begins to collapse, the states he’s likeliest to lose are Ohio, Arizona and Georgia.

Now for the popular vote.

National polling tightened considerably last week. Now a gap has appeared again. The latest polls show Clinton with a widened gap again – anywhere from 3 to six points. RealClearPolitics’ poll average gives her a 2.9 percent lead in a four-way race; FiveThirtyEight puts her at a 3.3 percent lead. My hunch, based on Trump’s lack of a ground game and the fact that Trump will drive higher-than-expected voter turnout in opposition to him in leftist states, is that the gap will be larger than that – that it will be more like five points, greater than the Romney loss but less than the McCain loss. Based on pure guesswork – and hey, I got it wrong in 2012, so I could be dead wrong here, too -- I’d estimate the final result as something like 49 percent for Hillary, 44 percent for Trump, 4 percent for Gary Johnson, 3 percent for everyone else.