Is it now about beating August 27, where (49.2 SB+47.1 538)/2 = 48,15? There hasn't been any day before the DNC with a higher average, remains to be seen if there'll be a higher post.

Edit: Oh sorry wait you're doing difference, not vote share. But the way to average it is the straightforward way described above, just picking the right date during DNC when the diff was the highest?

Or are you calculating the average some other way @SaviorofPlant ?

@HenriThunberg That's correct. I'll go ahead and do the math now:

August 19: 2.9 538, 2.3 SB = 2.6
August 20: 2.8 538, 2.5 SB = 2.65

August 21: 3.3 538, 2.3 SB = 2.8
August 22: 3.6 538, 2.5 SB =3.05
August 23: 3.7 538, 4.3 SB = 4.0
August 24: 3.6 538, 4.0 SB = 3.8
August 25: 3.6 538, 4.1 SB = 3.85
August 26: 3.4 538, 4.1 SB = 3.75
August 27: 3.4 538, 3.8 SB = 3.6

August 28: 3.6 538, 3.5 SB = 3.55

August 29: 3.5 538, 3.8 SB = 3.65

August 30: 3.4 538, 3.3 SB = 3.35

This market resolves NO instantly if there's any date on which the average exceeds the highest value here (which looks like August 23), and resolves YES if there's no point before the election where that's true. Someone please double check these numbers, I wouldn't want to misresolve.

I suggest moving the close date to Election Day. That's the latest date her polling averages can be higher than whatever they'll be at the convention.

Or maybe the idea is to leave it closed between the convention and the election, so that people won't just trade already knowing the result. That's also fair.

Upgraded!

Idea of this market: capture Manifold's probability that the convention bump will fade