Democratic strategists see a path to retaining their House majority.

To do so, they are aiming to pick off four seats held by Republicans, two open and two held by incumbents; then hold onto at least eight of their most endangered 16 open seats; and keep their incumbent losses down to less than 35 -- two-thirds of the Democratic incumbents in competitive districts, or just over 40 percent of the number if you include those who are potentially endangered.

In a normal election year, these would be very modest ambitions; today, it's a tall order. But it is plausible.

According to "Vital Statistics on Congress," in the last half century, a party has lost 35 or more incumbents only four times.

The first occurred in 1964, when 39 Republican incumbents were swept out in President Lyndon Johnson's landslide victory over Sen. Barry Goldwater, R-Ariz. In the very next election, Democrats lost 38 seats in a "six-year itch" -- a second-term, midterm election. Republicans lost 36 in 1974, another second-term, midterm election, just three months after President Gerald Ford's watch started in the wake of Richard Nixon's resignation.

Twenty-two Democratic incumbents are in the Toss Up category, the political equivalent of an intensive care unit.

The most recent was 1994, a first-term, midterm election like this one, under President Bill Clinton, when 35 Democratic incumbents lost.

Incumbent losses of this magnitude are thus not unheard of. They happen in wave elections, but it is a high bar.

In the Democrats' math, the first step is to pick off two open seats held by Republicans -- Rep. Michael Castle's Delaware seat and Rep. Mark Kirk's in Illinois. Both are running for the Senate and both districts have gone Democratic in recent presidential elections. Then Democrats have to beat two incumbents in Democratic districts whom they see as flukes -- Republican Reps. Anh (Joseph) Cao in Louisiana and Charles Djou in Hawaii.

These districts routinely vote even more heavily Democratic in presidential races, so these targets are hardly delusional.

There are a couple of other Republican incumbents where Democratic goals are more ambitious, like beating Rep. Dan Lungren in California and Charles Dent in Pennsylvania.

If Democrats were to win four, Republicans would have to grab 43 seats held by Democrats to score a net gain of 39, the number they need for a 218-217 majority.

Next, Democrats are looking at their 16 most endangered open seats and thinking that, based on their strong performances in special elections over the last 18 months, their worst-case scenario would be to lose no more than eight of the districts held by the following: Reps. Charlie Melancon in Louisiana; Bart Gordon and John Tanner in Tennessee; Vic Snyder and Marion Berry in Arkansas; Brad Ellsworth in Indiana; Dennis Moore in Kansas; Paul Hodes in New Hampshire; Bart Stupak in Michigan; Joe Sestak in Pennsylvania; Alan Mollohan in West Virginia; Brian Baird in Washington; David Obey in Wisconsin; Bill Delahunt in Massachusetts; Patrick Kennedy in Rhode Island, and former Rep. Eric Massa in New York.

In the Democratic calculus, if they picked up the four currently Republican seats, then lost just eight of their 16 most vulnerable open seats, that would put them down four seats, going into their own incumbent seats.

That means that 35 Democratic incumbents would have to lose for the House to flip. Even going with a pretty large number of theoretically vulnerable Democratic incumbents, say 70, that means that at least half would have to lose before the House turns over, if all of the other numbers are right.

The Cook Political Report currently rates 22 Democratic incumbents in the Toss Up category, the political equivalent of a hospital's intensive care unit.

Many of these will lose, though even under the worst case scenario for parties, there is often a random survivor or two.

Thirty-one more Democratic incumbents find themselves in races rated Lean Democratic, meaning they are in competitive races but retain some advantage.

If history is a guide, Democrats will suffer some losses in this category; it could be a quarter, maybe even a half on a really bad night for them.

Thirty-four other Democrats are in the Likely Democratic column, indicating they have races that are potentially competitive, but today are not so.

In a wave year, there may be a few losses in this column, though more likely, a few of these will find themselves in increasingly serious contests and their ratings will migrate to Lean Democrat or even Toss Up.

Today, a race-by-race analysis of the House suggests Democrats will lose roughly 34 seats, five short of losing their majority.

If one believes and factors in a substantial Republican wave, driven by higher proportional turnout among GOP than Democratic voters and independents breaking toward Republicans, it clearly goes over the 39 seats that would tip control to Republicans.

While I fall into that school of thought, when Democrats begin to point to their substantial financial advantage, it's intellectually honest to acknowledge they could be right.

It basically comes down to whether one believes there is a wave and, if there is one, just how big it is and how much force it carries.