Board Thread:Questions and Answers/@comment-10975360-20131014124545/@comment-32656-20131026001019

Exactly.

Yeah, that UK election is a classic case of bad polling. Really not a good reason for that one happening, especially given the scale of error. Way I figure it, the "Shy" factor generally blamed would be a factor, but it much more seems to be a general problem with polling during the UK in that decade. Got me at the reason, but I suspect it would have something to do with a combination of press coverage on voting day, incumbent support, and an over-concentration of polling in large cities.

The over-concentration would then have continued through the decade before getting fixed.

Yes, had Johnson managed that Humphrey would probably have won.

Agreed with regards to Perot. The same factor that gives incumbents extra support would have led to his support tailing off by election day. However, had he not "left" the race temporarily, it is likely that he would have won some EV's, and thrown things into Congress... giving Clinton and Gore the election anyway, since the Dems had solid control of both chambers.