https://circle.tufts.edu/2024-election-youth-poll

Tufts' newest analysis:

57% Extremely Likely to Vote

57% of youth, ages 18-34, say they’re “extremely likely” to vote in 2024, and another 15% say they’re “fairly likely” to cast a ballot in the election.

https://www.npr.org/2023/09/22/1201183160/taylor-swift-instagram-voter-registration

Somewhat relevant news.

A Taylor Swift Instagram post helped drive a surge in voter registration

More than 35,000 new voters registered at Vote.org after the singer posted a link on her Instagram Stories. "I've heard you raise your voices, and I know how powerful they are," Swift wrote.

To traders: In case the turnout rate falls precisely on 50%, how do you think it should be resolved?

A. Resolve as YES (inclusive of 50%)
B. Resolve as N/A (boundary case)
C. Resolve as NO (technically not 'above 50%')
D. Resolve to Prob%

@SarkanyVar Please vote below with likes to the comments.

A. Resolve as YES (inclusive of 50%)

B. Resolve as N/A (boundary case)

C. Resolve as NO (technically not 'above 50%')

(reserved)

@SarkanyVar What about resolve to Prob%, to the percentage of the estimated range from CIRCLE above 50%?

@adele That is a good suggestion too. I suppose I can do that too in the boundary case.

I would still lean towards resolving to YES or NO otherwise because participation by at least half of the demographics is an important barometer, and so I would want the resolving of the market to reflect so.

If other traders agree with resolving to Prob% in case of 50%, please feel free to like @adele's comment.

@SarkanyVar I'd honestly be so surprised at the boundary case happening I'd want it to be N/A out of respect. It gives "flipped a coin and landed on its edge" energy.

@AndrewQuinn Curiously, that's exactly where the latest report for the 2020 election, released almost half a year after the election, landed on.

https://circle.tufts.edu/latest-research/half-youth-voted-2020-11-point-increase-2016

Half of Youth Voted in 2020, An 11-Point Increase from 2016

We estimate that 50% of young people, ages 18-29, voted in the 2020 presidential election, a remarkable 11-point increase from 2016 (39%) and likely one of the highest rates of youth electoral participation since the voting age was lowered to 18. Our new estimate is based on newly available voter fi…

Given rounding, it is not exactly implausible. Still, I plan to use the two-week estimate instead of the figure in a final report (if any) for resolving the market because the latter's release date seems to be less regular. Food for thought though!

@SarkanyVar Given rounding! Bah, I should have considered that! Fair enough!

what has it historically been?

@RahulShah You may see below for the historical estimates at that time point from CIRCLE at Tufts. https://circle.tufts.edu/latest-research/election-week-2020

So by the criterion set out above, the 2016 youth voter turnout estimate would be 43% (the mean of 42% and 44%) and the 2020 youth voter turnout estimate would be 53.5%.

Note that the estimates had sometimes seen adjustments months after the point, when CIRCLE published their formal reports later. But for the sake of timeliness, I would stick with the two-week estimates as a reasonably accurate indicator for turnout.

For data before 2016, perhaps you may refer to this graph below:
https://www.statista.com/statistics/984745/youth-voter-turnout-presidential-elections-us/

Note that it seems to be taking the CIRCLE data and using the upper estimates, at least for the 2016 and 2020 data. I think using the mid-points would lead to fairer estimates. But the graph in general can serve as a reference to historical trends.