Mitt Romney has spoken, and the balloons have dropped: The Republican National Convention has officially come to a close. So what should we expect the last three days to mean for Romney's popularity in the polls?

University of Wisconsin-Milwaukee political scientist Tom Holbrook has done a lot of work on this; on his Web site, he describes convention bumps as one of his "favorite topics." He measures a convention bump as any change in polling in the week after the event, compared to polls taken the week before or two weeks after.

In 2012, Holbrook does not expect to see much change in the candidates' polling numbers at all. For a variety of reason, these conventions aren't shaping up to be ones that hugely sway public opinion.

Candidate bumps can be all over the place. Clinton saw, by Holbrook's measure, a 13.6 percent bump after his 1992 address. For Obama in 2008, that number stood at a paltry 1.8 percent.

Even within all that variation, Holbrook did find two metrics that can best predict the size of a candidate's poll bump. The first is how the candidate is doing in the polls already, and whether that aligns with the expected outcome of the election:

Candidates who are running ahead of where they "should" be (based on the expected election outcome) tend to get smaller bumps, and those running behind their expected level of support get larger bumps. In this way, the conventions help bring the public closer to the expected outcome and help to make elections more predictable. A perfect example of this phenomenon is the 1964 conventions. Goldwater got a huge bump, in part because he was running 16 points behind his expected vote share, and Johnson got no bump, in part because he was running 6 points above his expected vote share prior to the Democratic convention.

The second is timing, as the candidates who speak at the first convention tend to see more of a spike in polling numbers. That was true, at least, a few decades ago. But as the conventions have inched closer and closer together, timing has become less of an important factor.

When taking these changes in timing into account, it turns out that the number of days a convention is held before or after1  the other party's convention is a stronger predictor of convention bump (r=.37) than simply going first (r=.22).  Of course, the "days between conventions" measure captures both how early the first party convenes and how close together the conventions are.

There has also been a trend in recent cycles, Holbrook notes, toward smaller bumps. No one has shown a double-digit polling increase, by Holbrook's metrics, since Clinton in 1992.

What does this all mean for 2012? Holbrook has used these variables to create a convention poll forecasting model. Here's how it stacks up in predicting previous convention bumps:

As for what that means for 2012, Holbrook does not expect any major swings. Both candidates are running pretty closely in the polls to where they should be. The back-to-back conventions gives less of an advantage to the Republicans, as it mutes some of the effects of going first.

All told, Holbrook expects to see a 3.8 percent point bump for Romney and 1.8 percent point increase for Obama.

"Right now, the two candidates are in a tight race, with Obama holding a slight advantage in the polls," he writes.  "Based on my bump predictions, I expect that the race will continue to be tight after the conventions but that Romney will hold a slight lead."

There is one new variable, introduced in 2012, that Holbrook was not able to account for: Clint Eastwood delivering a monologue to an empty chair. All bets are off on what that will mean in ensuing poll numbers.