A couple of thoughts:
The Kerry bounce/bump or lack thereof:
I think a couple of things are going on here. First, it depends on the poll you look at. Newsweek had it at about 8-10 points, ABC/WaPo a bit less than that, and CNN/USAToday/Gallup nothing. My guess is that Kerry probably helped himself a bit, both in picking up a few points and firming up some people who were leaning his way, but not solid.
Second, the historical comparisons with other candidates are problematic. The "averages" for bumps cited in the news reports include Clinton in 1992. That year he had a bump of anywhere from 16 to 30 points, largely because Perot's dropping from the race created a big pool of voters who needed to find a new home. When you take Clinton out of the mix, the average bump is much more modest, about 5-8 points, and that about what Kerry did if you look average the three polls mentioned above.
Third, Steve's right that given the lack of network coverage, the early announcement of the Edwards pick, and the small pool of undecided voters, the potential for a bump this year was lower than in past years.
Fourth and finally, so what? There's no relationship between the bump you get out of a convention and how you do in the election. Hell, Mondale got a pretty good boost after his convention, but that didn't help him much. On the other hand, some candidates have gotten about what Kerry got and still went on to win. Perhaps there's a potential political science law here, something like the amount of media attention is inversely proportional to actual electoral impact. So if the media chatter on ceaselessly about something, go ahead and safely ignore it.
That leads into Steve's point about knowing more about the ground game and less about the polls. That's probably right, but the polls do tell us some things about the ground game. Right now Kerry is doing pretty well in the battleground states and this gives him more choices about where to deploy his resources. Also, Steve's leaving out the candidate debates. They could potentially change the race dynamic. Here I'm thinking about 1980, when Reagan and Carter ran pretty even until they debated. When Reagan showed that he was a reasonably safe alternative to Carter, the election broke his way and he won pretty handily. I don't know if that will happen this year, but they are the one event on the horizon could shake things up.