Then, yesterday, he told CNN that he backs a constitutional amendment banning gay marriage, even though he is “OK” with two dudes living together, or whatever. A few hours later, the Crist campaign “clarified” that the governor just meant an amendment to the Florida constitution. That is, apparently, the “moderate” compromise on the issue.

Crist’s refusal to say whom he’ll caucus with is actually making it impossible for voters to make an informed decision, but if elected it is almost certain that he will be either the sort of senator who will occasionally randomly decide to withhold his support from something until it is made worse, or he will be the sort of senator who will occasionally dangle the carrot of his support for something and then never follow through.

So if voters like the gridlock that makes them hate Congress, go nuts and vote Charlie Crist.

(To be fair, some people do like gridlock. There are people who do things like vote for Obama and then decide, one year later, to sabotage his major domestic policy proposals by voting for Scott Brown — but there are not many of them. Even though all of them seem to work in political journalism or punditry.)

But North wasn’t any generic Republican: The taint of Iran-Contra and his embrace of his party’s far-right fringe enabled Robb to turn the race into a referendum on his opponent — and to win. The silver lining for Republicans, of course, was that North was an aberration: Just about everywhere else in ’94, vulnerable Democrats fell, allowing Republicans to take back the House and Senate and a host of governorships.

On the surface, the parallels to ’94 are striking: A first-term Democratic president with middling poll numbers, big Democratic majorities in the House and Senate, and a reinvigorated GOP base. The basic recipe is in place for a similarly seismic GOP victory. But as Tuesday night’s results reminded us (yet again), there’s a potentially big difference: This year, the GOP has nominated a whole bunch of Oliver Norths — Tea Party-backed candidates who are seriously jeopardizing the party’s prospects in races that should be gimmes.

Take Rick Scott, the controversial healthcare executive who knocked off the GOP establishment’s favorite, Bill McCollum, in Florida’s gubernatorial primary. The difference in each GOP candidate’s fall prospects is jarring. In a recent poll, McCollum was running neck-and-neck with the Democratic nominee, Alex Sink — but Scott trailed by 16 points. The GOP electorate’s embrace of Scott, who was forced out of Columbia/HCA when the company was ensnared in a massive fraud investigation in the 1990s, is reminiscent of North’s nomination in Virginia 16 years ago. Then as now, the GOP base could have tabbed an electable, (relatively) mainstream candidate, but hubris got in the way.

Nor is the governorship the only high-profile office in Florida the Tea Party might muck up for the GOP. Marco Rubio’s victory in Tuesday’s Senate primary was a foregone conclusion; his original opponent, Charlie Crist, dropped out to run as an independent months ago. But the basic dynamic is the same: Rubio, is beloved by the GOP’s base but polarizing to swing voters. With Crist, the GOP would have won the Senate race with ease. With Rubio, their odds have decreased and their fate now depends on whether enough of the Democratic establishment rallied behind Crist. If that happens and Kendrick Meek, who won the Democratic nod on Tuesday, is marginalized, Crist can defeat Rubio.

The hand of the Tea Party was apparent elsewhere on Tuesday, though the implications are slightly different. In Arizona. John McCain moved decisively to the right to stave off a primary challenge from ex-Rep. J.D. Hayworth. There’s no reason to believe McCain will lose in November, but the steps he’s taken to survive politically have come at a cost to his reputation. (Although, as I’ve noted before, McCain’s posturing can also be seen as part of a pattern of sore loser-dom.)

The Tea Party seems to have scored a more decisive triumph in Alaska, where little-known Joe Miller leads Sen. Lisa Murkowski with more than 85% of precincts reporting. If Miller does hang on, he should be safe in November, given what a deeply red state Alaska is. But his success is a reflection of the influence the Tea Party is exerting in GOP primaries across the country — and the results aren’t always as benign: Sharron Angle, Rand Paul, Ken Buck, and  Pat Toomey, among others, are all products of the Tea Party movement and could all lose races that Republicans should win with ease this year.

In 1994, nominating Oliver North made for one disappointing result for the GOP on an otherwise perfect night. But in 2010, they’ve nominated many Oliver Norths — and it could be the difference between winning and losing the House, Senate, and even a key governorship.

But Crist’s surprising endurance in his independent Senate candidacy in Florida — an effort that was initially dismissed by most Beltway observers as a desperately lost cause — has complicated the Senate outlook, increasing the chances that Election Day will not yield a clear majority party for the chamber.

Here’s how it could happen. Democrats now control 47 Senate seats that are either not on the ballot this fall or that are on the ballot but are considered safe for them. Republicans have 36. This week’s news that the GOP’s top prospect in West Virginia’s special election, Shelley Moore Capito, is not running is probably enough to move that state into the safe category for Democrats — increasing the Democrats’ edge to 48-36. (Keep in mind, this counts Lieberman as a Democrat.)

Now, let’s look at what is in play this fall. Republicans are defending seats in Kentucky, Missouri, Ohio, New Hampshire and Florida. For the sake of argument, let’s say they win all of those contests – except Florida, where Crist wins. That would leave the Democrats with a 48-40-1 advantage. (Again, this is counting – for now – Lieberman as a Democrat.) Add in the four Democratic-held seats that the GOP is highly likely to claim — North Dakota, Indiana, Arkansas and Delaware — and the margin is reduced to 48-44-1.

That would bring us to the seven remaining races: in Pennsylvania, Illinois, Colorado, Nevada, California, Wisconsin and Washington. The first four of these essentially rate as toss-ups right now (although the identity of the GOP nominee in Colorado won’t be known until next month), while Democrats have a slight edge in the latter three. But for our purposes, let’s lump all seven of them together and consider the implications of a two different scenarios:

* If the GOP wins five of them, the Senate balance would be 50-49 for the Democrats, with Crist as an independent. Under this scenario, the GOP would be able to secure control only by convincing both Crist and Lieberman to caucus with them. If they did, the GOP would gain a 51-49 majority. But if only one of them — Crist or Lieberman — joined with the Republicans, Vice President Joe Biden would be poised to cast the tie-breaking vote for the Democrats. Or, if Lieberman were to stick with the Democrats and Crist were to join him, the Democratic majority would be 51-49.

* If the GOP wins six of those seven races, the Senate balance would be 50-49 for the GOP, with Crist as an independent. Under this scenario, Democrats would need to convince Lieberman to stay put and woo Crist in order to create a 50-50 tie for Biden to break. In other words, either Crist or Lieberman would be able to deliver Republicans control of the chamber by siding with them.

So, if the GOP wins fewer than five of those seven toss-up/Democratic-leaning races, Crist and Lieberman won’t be crucial to determining Senate control. Nor will they be if the GOP wins all seven of those races, since Republicans would then have 51 seats without them (unless, of course, the GOP slips up somewhere else, like, say, Kentucky). But if Republicans wins five or six of those seven, then Crist and Lieberman will loom large.

Who would they side with? Certainly, the signs now point to Crist joining the Democrats. He’s been sounding many of the party’s themes in his campaign, and if he does win, he’ll owe it to Democratic defections. There’s also news of a recent phone conversation between Crist and Harry Reid, who may end up surviving in Nevada, and as the Democrats’ Senate leader.

But his thinking could change. Given his strategic imperatives in Florida right now, it’s in his interests to play footsie with Reid. But after the election, he might have more room to maneuver. And if the GOP is desperate for control, who knows what they might offer him? Plus, it’s still possible that Reid will lose his reelection bid in Nevada, and thus his Senate leadership perch, which could invalidate any informal arrangements that he and Crist might now have.

As for Lieberman, it’s hard to imagine him bolting for the GOP, if only because he’s held back despite having so many opportunities to do so for the last four years. Part of this, of course, is because top Democrats (like Barack Obama) have bent over backwards to accommodate him. If control of the chamber is on the line, they’d presumably make him yet another attractive offer.

But Lieberman will also have to consider his own reelection in Connecticut in 2012. Overall, his standing in the state is weak: a 40 percent job approval number, something that was enough to force Chris Dodd to abandon his own reelection effort earlier this year. What’s worse for Lieberman, though, is that his approval rating is significantly lower among Democrats (a negative 33-57 percent score) than among Republicans (a positive 51-42 percent score).

Realistically, to win reelection in ’12, Lieberman will have to either run as a Republican or form the kind of informal alliance with the GOP that he enjoyed in his ’06 campaign. In that context, swinging the Senate to the GOP this November would be a fine way to jump-start either effort.

That Lieberman could still haunt them like this four years after losing a primary in Connecticut surely irks many Democrats. There’s a chance — not a very good one, but not a very remote one, either — that Republicans will soon know the feeling.