Today Gallup has a story on convention bounces. Using the final Gallup poll before the convention as the starting point of a "bounce," and the first poll after the convention as the end point of a "bounce," they show the following convention bounces since 1964 (incumbents in bold):

Year	 Candidate	Bounce
2000	   Gore 		 +8
2000	   Bush 		 +4

1996	   Clinton	 +5
1996	   Dole 		 +3

1992	   Clinton	+16
1992	   Bush 		 +5

1988	   Dukakis	 +7
1988	   Bush 		 +6

1984	   Mondale	 +9
1984	   Reagan		 +5

1980	   Carter		+10
1980	   Reagan		 +8

1976	   Carter		 +9
1976	   Ford 		 +5

1972	   Nixon		 +7
1972	   McGovern	 +0

1968	   Nixon		 +5
1968	   Humphrey	 +2

1964	   Goldwater	 +5
1964	   Johnson	 +3

Apart from Bill Clinton being a major-league ass-kicker in 1992, the lessons to be drawn from this chart are murky. The challenger usually receives a larger bounce than the incumbent, but not always. Democrats usually have a larger bounce (seven times in a row, in fact) but not by much. The candidate with the larger bounce from the convention does not seem to win the election with any greater frequency than the candidate with the smaller bounce. However, since Kerry is ahead right now, Bush probably needs a larger bounce in order to win (or at least an equal bounce in order to stay alive until the debates).

It is going to be a frustrating week. Due to massively scaled back coverage, the number of viewers of the Democratic convention will not be large. Kerry's bounce will largely be determined by the way the SCLM reports on the event. While many intrepid bloggers will offer their own reporting and spin, this one will be largely out of our hands. Go to a Meetup, cross your fingers, and get ready to write a letter to the editor.