Perhaps we could get an exact percentage of what you mean by significant, as well as what polling data will be used to resolve the question?

@Mana @eliot

@mattyb Let me do some research and find a number - I’m happy to take suggestions though!

Okay, I spent some time looking into this – it seems like 10% would be a good benchmark. I'm also happy to consider if it majorly influences the results of the election – i.e. if a number of Republican lawmakers are elected BECAUSE of this migration then that seems to meet the criteria for significant.

@eliot What standard will determine whether it was a migration of Jewish voters that caused a Republican to be elected? Also what does "a number" mean here, and do they have to be lawmakers or would (for example) governors count?

For the 10% standard, does that mean "there is a shift of at least 10 percentage points in Jewish votes for president from the Democratic candidate to the Republican candidate, relative to the 2020 presidential election" or something else? And what exit poll or combination of exit polls will you use to determine that percentage?

@sesquipedalianThaumaturge

1. If Jewish voters are the deciding factor in an election, i.e. a Rebpublican who would've lost without a shift in Jewish voters (compared to '22) wins in '24.
2. I think it makes the most sense to use '20 as a reference point for the presidential race and '22 for everything else. Exit polls are whatever is available, with an obvious preference for Gallup / Pew / NYT / 538 / etc.

There is some subjectivity in the resolution of all markets, especially ones like this – I don't anticipate the resolution criteria being a problem! If there is any controversy, that will obviously be discussed before market resolution.

@eliot are we counting exclusively Congress members, or do e.g. state legislators count

@SemioticRivalry I don't think it applies exclusively to Congress members, but they are certainly a better way to see a nationwide trend. I'm happy to take state legislators into account, my only worry is finding the data.

@eliot Wait, so what vote counts does the 10% threshold apply to, if not just the presidential totals? If there's a 10% swing in one Senate race, does that count? Or does it have to be the total vote for Congress across all seats? Can state level races count for the 10%?

Also, how will you resolve if some of the pollsters you mentioned show less than 10% and some show more?

I realize it's impossible to make completely objective criteria but at the moment this market is too ambiguous for me to bet on. I can easily imagine there being a lot of disagreement over how it should resolve with the current criteria.

@sesquipedalianThaumaturge I think the main issue is: how do you quantify a national shift with the limited voting data that we have available. I’m using 10% as a benchmark, just like one might say that “x% of southerners migrated from the Democratic Party in the 1960s” but I agree with the core of your message. I’m open to better ideas for resolution criteria as well, but the status quo is the best way that I can think of to attempt to quantify a market of this nature.

@eliot If I were you I would pick a very specific resolution criterion that you think is a good proxy for the question you're interested in, like "according to [particular exit polling average], there will be a 10 percentage point shift in Jewish votes in the presidential election from the Democratic candidate to the Republican one, relative to 2020". Because there are big differences between that and other interpretations you haven't ruled out, like "in at least one federal or state election anywhere in the US, at least one exit poll shows a swing of at least 10 percentage points among Jewish voters from Democratic to Republican". I feel like the probabilities of those two things could be quite different, but currently I don't know which one this market will end up being about. Ultimately it's up to you though, you can pick the criteria you want and then people will decide if they want to bet.

Putting a word in italics doesn't make the definition any more precise