Gandalf wrote:
We tend to re-elect war-time presidents because people are reluctant to change leadership when they are uncertain about the future. It is not a huge numerical bias, but it is big enough to swing an election.
Is this common, or one of those odd American things?
Well...
Madison: re-elected in 1812
Lincoln: re-elected (but we know how that went, and Johnson didn't run for re-election IIRC)
McKinley: re-elected AFTER the Spanish American War was concluded
Wilson: unusual case as we weren't actually involved in WWI at the time (we didn't go in until 1917) but he was re-elected in 1916.
FDR: Re-elected twice (1940, 1944).
Lyndon Johnson and Nixon both rode the Vietnam War in 1964 and 1972. Unusually enough, Hubert Humphrey wasn't elected in 1968 despite being Johnson's VP. Of course, Nixon being Nixon, there's a decent chance the election was crooked, and the Democrats were unpopular thanks to the burgeoning neocon isolationist movement and the hippies protesting at the 1968 convention. Johnson could've run, but decided not to.
Finally, W was re-elected in 2004.
Exceptions: Polk declined to run again after the Mexican War; Truman decided not to run again in 1952; as noted above, Johnson in 1968; and H.W. Bush lost a re-election bid despite winning in the Persian Gulf.
So that's 8 to 4, and Truman and Johnson were promoted VP's who had the option of running for a second full term but declined so in theory they could've run again. Polk is far enough back that nobody cares. HW Bush is somewhat of the odd one out, but he got the double whammy of an economy in the toilet (coupled with people remembering the "no new taxes" fiasco) and Ross Perot sucking away a bunch of his votes. McKinley did win an re-election bid... but he was assassinated very shortly afterwards, and Theodore Roosevelt covered most of his term.
It's a strange world. Let's keep it that way.