By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Just for the funzies, let's look at how elections turned out since the last great realignment, compared to where they stood about now:

Presidential national average lead, 17 days before E-day:

1992: Clinton +11.1*
1996: Clinton +13.8*
2000: Bush +3.1*
2004: Bush +3.4
2008: Obama +5.6
2012: Obama +0.1
2016: Clinton +6.4
2020: Biden: +10.6

* for the month of October as a whole.

Final stats, popular vote:

1992: Clinton +5.6 (5.5 toward Rep)
1996: Clinton +8.5 (5.3 toward Rep)
2000: Gore +0.5 (3.6 toward Dem)
2004: Bush: +2.5 (0.9 toward Dem)
2008: Obama: +7.3 (1.7 toward Dem)
2012: Obama: +3.9 (3.8 toward Dem)
2016: Clinton: +2.1 (4.3 toward Rep)

Average shift: +0.85 toward Rep
Predicted Biden margin: +5.1 to +14.4 (median +9.75)
Predicted EVs, assuming the same partisan lean as 2016: 308-230 to 422-125

Conclusions: October polls are a strong indicator of the end result and either a) the Clintons always underperform their polls badly; or b) stronger third-party voting correlates with Republicans making up some ground.