We are now officially down to the wire for the 2014 midterm elections. How well will the Republicans do this year? Will the Democrats hold the line in the Senate? Tune in tomorrow night....

Until then, however, I'm posting my own election picks. I've been doing this since 2006, because I have always felt that political pundits -- who, after all, make their living predicting what is going to happen in the political world -- should not only make solid predictions before elections, but also show their own personal record. I've been saying this all along, in fact, but I have been remiss in adding up my own past record for midterm elections so far this year.

I don't know why I didn't do this before now (actually, I do: I'm lazy, and it took some data digging to figure out), because my results are pretty impressive, if I do say so myself. Here's a quick rundown of my picks for the last two midterm election cycles (click on the links at the top to see the column run on Election Eve of that year). To simplify things, I'm counting "Independents who caucus with the Democrats" as just "Democrats," I should add.

2006
This was the first "Election Eve" column I ever wrote, so it was a little bare-bones. But my predictions were pretty accurate.

House elections:
229 Democrats, 206 Republicans -- My picks
233 Democrats, 202 Republicans -- Actual results
Plus-or-minus (range of my error) -- 4 seats

Senate elections:
My picks: 51 Democrats, 49 Republicans
Democrats win battlegrounds: MD, MI, MN, MO, MT, NJ, OH, PA, RI, VA, WA
Republicans win battlegrounds: AZ, NV, TN
Actual results: 51 Democrats, 49 Republicans
States called wrong: None

2010
Again, I got pretty close, even in a bad year for Democrats.

House elections:
199 Democrats, 236 Republicans -- My picks
193 Democrats, 242 Republicans -- Actual results
Plus-or-minus (range of my error) -- 6 seats

Senate elections:
My picks: 53 Democrats, 47 Republicans
Democrats win leaning states: CA, CT, WV
Democrats win battlegrounds: CO, NV, WA
Republicans win leaning states: AK, KY, MO, WI
Republicans win battlegrounds: IL, PA
Actual results: 53 Democrats, 47 Republicans
States called wrong: None

Now, that record looks pretty good, but there are a few caveats. The first is I only had time to dig the data out for my midterm election columns, so this doesn't include my predictions for Senate races in 2008 or 2012 (where I'm positive I probably got at least a few guesses wrong). Second, past performance is no guarantee of future success, so I certainly don't think I'm going to get every Senate race right this year (there are a lot of extremely close races).

Finally, even when guessing the numbers right, I got a few details wrong. In 2010, I predicted that Harry Reid would face a recount and a court case, and while he would eventually triumph, the distraction would force him to step down as Majority Leader (I predicted Dick Durbin would replace him). So it's not like everything I prognosticate comes to pass, or anything (although it certainly does bring up the question of whether Reid will continue in the Senate leadership after tomorrow night if Democrats lose control, doesn't it?).

In any case, enough bragging over past results, let's get on with predicting tomorrow's outcomes.

House races

I'm starting my final picks for the 2014 midterm election with the House races, which I haven't individually addressed in previous columns (for the simple reason that 435 races means I don't have time to examine even just the close ones in any detail). But since it is Election Eve, I'll take a stab at predicting the partisan makeup of the House of Representatives in the 114th Congress.

The current makeup of the House is 199 Democrats, 233 Republicans, and three vacancies. RealClearPolitics shows (as of this writing) the breakdown of the House as: 181 at least leaning Democratic, 228 at least leaning Republican, and 26 up for grabs. I'm going to give Republicans an edge (but not an overwhelming one), and say they pick up 15 of the close races. This means my prediction for the incoming House is:

192 -- Democrats
243 - Republicans

Got your own House prediction? Let me know in the comments. Picking House races is (to me, at least) an exercise in tossing a dart at the wall, so I'm not very confident in my choices, I have to admit.

Safe Republican

Moving on to the individual Senate races, we have no change in the Safe Republican seats from last week's column.

The full list of Safe Republican races: Alabama, Idaho, Maine, Mississippi, Montana, Nebraska, Oklahoma (both seats), South Carolina (both seats), South Dakota, Tennessee, Texas, West Virginia, and Wyoming.

With these fifteen wins, Republicans will pick up three seats previously held by Democrats: Montana, South Dakota, and West Virginia. They will need a net total of six such pickups to win control of the chamber from Democrats.

Safe Democratic

Some would argue that polls have weakened for Democrats in Virginia, but I am not one of them. The list of Safe Democratic seats also did not change from last week's list: Delaware, Hawai'i, Illinois, Oregon, Massachusetts, Michigan, Minnesota, New Jersey, New Mexico, Rhode Island, and Virginia.

None of these eleven races represent a pickup for Democrats, though.

Leaning Republican

I've moved Louisiana down to the tossup category, because there will likely be a runoff election in December. More on this in a moment.

Early in the year, Democrat Mark Pryor of Arkansas looked like he might survive and keep his Senate seat. But from the beginning of the summer onward, he has trailed in almost every poll. In the past month, the lead Republican Tom Cotton has opened up has steadily been getting wider. What was once seen as a possible "hold the line" state for Democrats now appears to have been won for Republicans.

Democrats also had some high hopes for snatching Kentucky away from Mitch McConnell, but Alison Lundergan Grimes hasn't delivered on these hopes since, again, about the beginning of the summer. Democrats would have been a lot happier losing control of the Senate if they had managed to take down the man who would be Majority Leader, but alas, this does not look possible now. McConnell has widened his lead in the polls in October, and looks like a lock tomorrow night.

Winning both these contests means Republicans would pick up one more state (Arkansas) from Democrats -- which leads to a cumulative total of four pickups for the GOP, so far.

Leaning Democratic

Two states, Alaska and Georgia, moved down to Too Close To Call this week. This leaves two states which are currently Leaning Democratic.

New Hampshire was seen early on as a possible pickup for Republicans, but they nominated carpetbagger Scott Brown, who had earlier failed to win a Senate seat in neighboring Massachusetts. The race has tightened in the last month, leading to a resurgence of Republican hopes. But sitting Senator Jeanne Shaheen has been up (or at least tied) in every poll, and my guess is she holds on to her edge. Especially since Brown was just publicly caught in confusion over one of New Hampshire's counties (there just aren't that many> to keep track of, Scott), which plays right into the whole "carpetbagger" theme.

The North Carolina race has a similar profile in the polling as New Hampshire. In both races, a Democratic incumbent woman is fighting off a Republican challenger in a purple state. North Carolina may be one of the nation's purplest states right now, as a matter of fact, which is why the race has been so close. Kay Hagen is fighting off a Republican who led the state's legislature to pass some awfully extreme right-wing bills. She's maintained a slim margin in just about every poll for the past two months, but that margin is pretty thin indeed. But my guess is that the polling's right and Hagen returns to the Senate.

Both of these states, it bears mentioning, will be bellwether states tomorrow night. Because they are on the East Coast (and therefore close their polling places earlier than other time zones), these races may indicate the direction of the whole Senate election. If either (or both) New Hampshire or North Carolina fall to Republicans, it may portend a wave across many other battleground states. So the pundits will be watching these two very closely tomorrow night, that's for sure. However, even winning both states for Democrats would not represent any pickup. Republicans would still be up by four states overall.

Too Close To Call

We've got six states in the Too Close To Call category, but fear not! After providing a rundown of the races, I am adding a section below where I make my own picks for each of them (since it is Election Eve, after all, and thus time to put up or shut up).

I optimistically put Alaska up in Leaning Democratic last week, but subsequent polling has not completely justified my enthusiasm, so it must return to Too Close To Call status. But the good news for incumbent Democrat Mark Begich is that the recent polling has been a lot better than most of September and October. This late movement in Begich's favor may portend an upset, but if it comes it will come so late in the evening that most East Coast politics-watchers won't even notice it until Wednesday.

In Colorado, the big question is whether the all-mail voting will change the size and makeup of the actual electorate, and if so, how? Incumbent Mark Udall has been counting on his ground game to make up for a very slight deficit in the polling. Whether he can pull this off or not is anyone's guess.

Georgia is extremely close in the polling, although Republican Dave Perdue may have some last-minute momentum. However, the question here is whether either candidate can post the "50 percent plus one vote" majority needed to avoid a runoff election. If neither does, then this race won't be decided until the first week in January. My guess is that neither candidate does win enough tomorrow night, meaning the outcome will be moot (whichever candidate "wins" will have to start campaigning all over again, and may ultimately lose in the runoff). If Michelle Nunn does win here, it will represent a significant victory for Democrats, continuing a trend of turning some of the South from deep red to purple (see: Virginia and North Carolina).

Iowa was a disappointment for many Democrats, who expected Bruce Braley to coast to an easy victory. While the polling remains close, Joni Ernst has had the (slight) advantage for about two months now, and late movement seems to be in her direction. Losing Iowa would be a blow to Democrats, but at this point it looks more likely than not.

Kansas is a real wild card, because Independent Greg Orman still refuses to say which party he'll caucus with (for the important Majority Leader vote). While electing any non-Republican here might be seen as a huge Democratic victory (it hasn't happened in three-quarters of a century), if Orman winds up voting for Mitch McConnell, it will be a fairly empty one for Democrats. The polling is very close, but Orman has held a razor-thin edge all month long.

I'm also predicting Louisiana will head for a runoff election. Which means we should move right along to the part where I actually predict the outcomes of all the close races. My predictions for the remaining races may be a bit of a wild ride, so fasten your seatbelts.

Final Picks

Before picking the final six state races in the Too Close To Call category, Republicans are already up four seats. Overall, this puts the total rather symmetrically at 47 Democrats and 47 Republicans. To gain control, Democrats need only three wins while Republicans need four (because Joe Biden would break a 50-50 tie in Democrats' favor).

The first pick I'll make is Iowa, where I think Republican Joni Ernst will win. I've personally met Bruce Braley and think he'd make a wonderful senator, but at this point I think he's going to fall short. This puts the tally at 47 Democrats and 48 Republicans.

I'm going to go out on a limb and predict that Mark Begich pulls off an upset up in Alaska. How stunning this will be to other pundits depends on how closely they have been paying attention to the race. Polling is notoriously difficult in Alaska, and Begich's vaunted ground game will be the deciding factor. If I'm correct in calling this upset, it will leave the count at 48 Democrats and 48 Republicans.

I think that Greg Orman will win in Kansas, and furthermore Governor Sam Brownback will also lose his race here to a Democrat. The backlash against Tea Party fantasy budgeting math will be brutal for Republicans in the Sunflower State. However, even if I'm right and Orman wins, we still don't know which side he'll join. Total has to be stated as 48 Democrats, 48 Republicans, and Orman (with a question mark next to Orman's name).

Not unlike Iowa, I'd really like to believe that Mark Udall is going to pull it out in Colorado. But I think that even with the mail-in voting factor, he's going to come up just a tiny bit short. A number of these races are in my own personal "I hope I'm wrong about this" category, but I have to overrule my unbridled optimism here and call Colorado for the Republican. The tally if I'm right so far would be 48 Democrats, 49 Republicans, and Orman (?).

This is the point where I play Punxsutawney Phil, and predict four-to-eight more weeks in the 2014 campaign season. I think that Mary Landrieu will win tomorrow's "jungle primary," but that she will fail to achieve the 50 percent plus one vote threshold to avoid a runoff. I am also predicting that neither candidate will clear this bar in Georgia tomorrow night either. Leaving us all with two more months of "who will control the Senate?" fun and games.

During this period, both Harry Reid and Mitch McConnell will be furiously trying to ever-so-legally bribe Greg Orman into caucusing for their leadership. Both sides will be fully aware that with Orman's vote, they'll only need to pick up one of the runoffs, but absent Orman they'll need to sweep both races for control. Orman will likely stay cagey for as long as he can manage.

In December, I think Mary Landrieu will lose her runoff. She's won runoff elections before, and both parties will sink an enormous amount of money into the race, but I think eventually she'll lose. In head-to-head polling, she trails the Republican currently, and while that could change, I don't think it's going to. In early December, the calculus will be 48 Democrats, 50 Republicans, and Orman (?).

In January, however, I am going out on a limb to predict a surprise win for Michelle Nunn in Georgia. This will leave control of the Senate on Orman's back, split 49 Democrats to 50 Republicans. In the end, I think Orman will choose to caucus with the Democrats, though, leaving them with the smallest majority possible: 50 plus Joe Biden.

Think this is all too wildly optimistic for Democrats? Perhaps. Think I'm being blinded by my own biases? Could be, could be. If I call even one of these races wrong, it could indeed mean control of the Senate flips, I fully realize. Whatever happens tomorrow (and beyond), though, I've tossed my markers down on the boards. Anyone who has a different opinion, feel free to let me know about it in the comments. Let me know your picks, and I'll see you here in two years for my next election rundown column series!

-- Chris Weigant

Cross-posted at The Huffington Post

Follow Chris on Twitter: @ChrisWeigant