Sometimes the shifts of elite Washington opinion are not based on underlying reality. So it is with Democratic prospects for taking over the House.

The consensus punditry on this question has shifted wildly from “maybe” to “yes” to “no,” while little has really changed.

Contrast this with the remarkably constant state of our knowledge of the Senate election. In March, the analysis went something like this: Republicans had one incumbent in deep trouble, Rick Santorum (Pa.), and four — Conrad Burns (Mont.), Lincoln Chafee (R.I.), Mike DeWine (Ohio) and Jim Talent (Mo.) — in dogfights that could go either way. To get to the magic number of six, Democrats needed to win all of these plus another, such as GOP Majority Leader Bill Frist’s open seat in Tennessee. And they had to prevent Republicans from winning in Mark Dayton’s open seat in Minnesota or their somewhat longer shots at beating incumbents Maria Cantwell (Wash.), Robert Menendez (N.J.), Ben Nelson (Neb.) or Debbie Stabenow (Mich.).

Not much has changed since March. We knew then and we know now that Democrats should gain seats, and counting to 48 or 49 Democrats in the Senate was not so hard, but that it is difficult to count to 51.

Now think about the House. March analysis was much more general: Conditions are bad for Republicans. Midterm elections usually go against the president’s party. The president and Congress have a low approval rating. The right-track, wrong-track number is very negative. And Republicans control all of the levers of power and are affected more by Washington scandals, making this election a national referendum on Republican performance.

Balancing the bad news for Republicans were structural factors. The vast majority of seats are either safely Republican or Democratic, incumbency reelection rates are high and there are few open seats in competitive districts.

The consensus view was that Republicans were in trouble but that the wave in the Democratic direction might not be enough to overcome Republican advantages.

Then the common wisdom about the House changed. In April, we began to receive dribs and drabs about individual races from private polling and analysis and from ad buys by candidates and outside groups.

Call this the Thelma Drake period, named after the Republican congresswoman from Virginia. Drake is in her first term but had not been on anyone’s short list of vulnerable candidates, having won 55-45 over her Democratic opponent in a district President Bush won 58-42 over John Kerry. But since she and others like her were targeted by MoveOn.org and given special attention by the GOP, speculation began that if Drake is in trouble then maybe other fairly safe Republicans could be washed over by the Democratic tide.

These tidbits liberated the pundit class, and many declared that the House was in play, or even likely to switch hands.

The past three weeks have seen another swing of opinion. Bush has slightly bounced back, with his White House reshuffle and the killing of terrorist leader Abu Musab al-Zarqawi in Iraq. And the victory of former Rep. Brian Bilbray (R-Calif.) in a California special congressional election caused commentators to minimize the strength of the Democratic advantage in the fall. So the latest chic prediction is for a smallish Democratic gain.

Despite the fluidity of opinion, the fundamentals have not significantly changed. The nature of House races is that we will not have the kind of specific, frequent, race-by-race polling that we see in Senate races, at least until early fall. Until then, Democrats can still talk about the strength of the national tide, and Republicans about how fortified their incumbents are against the storm.

Fortier is a research fellow at the American Enterprise Institute.