Regarding good economic news, public opinions about the economy are very much a lagging indicator... and tend to lag especially when jobs do. In late 1983 the economy was in a huge expansion but Mondale managed to tie Reagan in the polls due to the lagging job market. Same thing in 1992, although the expansion was not so huge then, it had nevertheless run a year and a half between March of '91 and November '92. And most people polled in 1995 would probably still have said the economy was bad.
The most typical trend is to look at the economic facts from the 1st and 2nd quarter and project that onto attitudes in Nov. In 1992 the best numbers began to show up in the 3rd quarter and were too late to register with public opinion. In 1984 the economy (though still in worse shape than today) had been clearly picking up from late 1983 into early 1984 and by Nov. registered with the voters to give Reagan a landslide.