One hears a lot of speculation about the Republicans’ chances of taking over one or both houses of Congress in November. If you have too much time on your hands you could follow daily (and more than daily) updates on the 37 Senate races that will be on the ballot. Even a political junkie like me has to tune it out a bit. But with almost all of the primaries done, perhaps it’s OK to take an overview of the landscape of Senate races (House races another day perhaps, but that can really make you nuts —there are 435 of them).

There is little disagreement among handicappers that the Repubs will make big gains this year. It’s routine and expected for the party in the White House to lose seats in the midterms, but this seems likely to be a better than average for the current out party — the Repubs.

Still, from where things stand now, the odds appear high that the Dems will still control the Senate during the second half of the President Obama’s term. The biggest reason, which isn’t often remarked upon, is almost pure luck. There are 37 Senate seats on the ballot this year. The Repubs will probably win a solid majority of them, maybe even two-thirds, maybe close to three-fourths! But because of the big Dem successes over the past two cycles (2006 and 2008), a large number of Dems — including a lot of freshmen (bear in mind that incumbents are often most vulnerable on their first reelection bid) — aren’t on the ballot this year.

Of the 63 seats that are not up this year, Dems hold 40 and Repubs hold 23. This is big. It means that of the 37 that are up, the Dems need to hold only 10 to maintain control. (Another aside: A 50-50 Senate still equals Dem control because the vice president, who holds the tie-breaking vote, is Joe Biden.)

(And, an aside to that aside: When Biden became veep, it seemed unlikely that Dems would have to worry about his seat. Delaware is a solidly Democratic state with a Dem governor who appointed a Dem caretaker to serve out Biden’s term. Biden’s son Beau was considered the likely best in any race to fill the seat. But when Mike Castle — a former governor, sitting congressman and by far the biggest Repub name in Delaware politics — decided to seek the Senate seat (despite the fact that he will be a 71-year-old freshman), he became the instant frontrunner. Beau Biden decided not to risk his political career on a Senate bid. Castle is now heavily favored to turn that seat from blue to red.)

But to nail down the point I was making before the double aside: In order to gain control of the Senate, the Repubs would need to win 28 of the 37 races. Not sayin’ that’s impossible, but that kind of sweep is a very tall order.

So let’s take a quick overview four news organizations that maintain an ever-changing assessment of  every single Senate race: Congressional Quarterly, the Cook Political Report, the New York Times and the Rothenberg Political Report. Each of them has been doing this for a long-time; each has a solid reputation as race trackers.

Each of them uses slightly different terminology and methodology. The NY Times has just five categories for the seats that are up this year: Solid for the Dems, likely to be won by the Dems, a toss-up, likely Repub and Solid Repub. CQ and Cook have seven categories (they divide the close but not quite toss-up races into likely and leaner categories). Stu Rothenberg gives himself the most options (nine) that range on each side of the divide from solid (for one party or the other) to a toss-up that Rothenberg thinks “tilts” ever-so-slightly Dem or Repub.

For comparison purposes, I’ll ignore those fine gradations and group everyone’s leaners, likely and tilters together so that every race is listed as either not on ballot (and held by Dem or Repub), solid for one party or the other (meaning not considered in play), and leaning one way or the other (whether the lean is listed as a lean, a likelihood or a tilt).

Category 1: Dem-held seats not up this year.
Category 2: Safe for the Dem.
Category 3: Likely, leaning or tilting to the Dem.
Category 4: Pure Toss-up.
Category 5: Likely, leaning or tilting to the Repub.
Category 6: Safe for the Repub.
Category 7: Repub-held seat not up this year.

So here’s one way to look at the overall takeaway, remembering that this is just four assessments two months before Election Day: If you assume that each party will get the seats that are currently leaning its way, and if you assume that the toss-ups will be roughly split, but if you assume that the Repubs get the “extra” toss-up when there is an odd number, it comes out:

NY Times: Dem majority by 55-45

CQ: Dem majority by 54-46

Cook Political Report: Dem majority by 54-46.

Rothenberg Report: Dem majority by 52-48.

Political handicappers often point out that the toss-ups usually don’t split down the middle. One side or the other usually gets most of them, based on late momentum. But a complete sweep of the toss-ups is unusual. In order to construct a Repub majority from this point, you have to surmise that the Repubs get a very big majority of the toss-ups. On the NY Times or Rothenberg charts, Repubs could sweep the toss-up and still be in the minority. On the CQ or Cook lines, the Repubs would need 12 out of 13 of the toss-up to get to 51.

Again, all of these numbers are just the educated guesses of political savvy prognosticators looking at a lot of polls and other factors. (In fact, the numbers above were current as of Friday but may have changed by the time you are reading this.) And take into account a considerable enthusiasm gap that keeps showing up between likely Repub and Dem voters. Still, pouring all that salt over the possibilities, the chart convinces me that to take control of the Senate, the Republicans will have to draw to an inside straight.

Addendum: Obviously, the various raters are not unanimous about which are the toss-up races. Rothenberg has the small number because he is willing to assign a race to the blue or red side based on a lean so small that he calls it only a tilt. Cook and CQ have the biggest toss-up lists and they agree on 12 out of the 13 that they put in that category.

But if you want to know which races all four agree are in the Toss-up category, they are the Senate races in Colorado, Florida, Nevada and Pennsylvania.

One last addendum: There’s another possibility almost too funny to mention (although this situation has recently come to pass in the Senate, in 2001, when liberal Repub Sen. Jim Jeffords left the Repubs and switched to an independent caucusing with the Dems, giving the Dems majority control).

This year, there is the Florida Senate race, where current Gov. Charlie Crist — a lifelong Repub who launched an independent candidacy only after it became clear that he would lose the Republican primary for Senate — has a serious chance to win as an independent. In fact, he has led in five of the last 10 polls published on pollster.com. If Crist wins that race and we end up with a Senate balanced on the head of a pin, the moderate Republican turned independent Crist could hold the balance of power.

Crist has refused to say which party he would caucus with.

If you like to follow on your own the Senate-watching ratings of the publications in the table above, here’s Rothenberg, here’s CQ, here’s Cook and here’s the NY Times.