This op-ed in The New Republic by two political scientists effectively demolishes the notion that this election was either decisive or gave Bush or the Republicans a mandate. Interestingly they say that mathematical models created by political scientists using economic variables on average predicted Bush would win with 54% of the vote.
There is more good analysis at Talkingpointsmemo here and here.
I also found this post-mortem on Ohio from a journalist in Athens, a small univeristy-centred Kerry-voting town in southern Ohio, interesting. I don't think he reconciles his perception of Ohio as Republican-leaning with the very close outcome in Ohio overall - a switch of 70,000 votes changes the outcome - but his observations are worth reading.
And we all have to keep our eye on Barack Obama. Here is an interesting account of his very successful Senate run.