Poll: Should we keep the quick-bet arrows?
Basic
44
Ṁ1548resolved Jun 5
Resolved
YES1D
1W
1M
ALL
Write "YES" or "NO" in the comments to cast your vote on whether you think Manifold should keep its current one-click betting interface using up- and down- arrows. (If you like the current implementation or think we're only a few tweaks away from this being a good experience, vote YES, otherwise vote NO.)
This market will resolve YES if a majority of the votes are YES; NO if a majority are NO within a week of the creation of this question. (In the case of a tie, I will resolve PROB at 50%.)
This question is managed and resolved by Manifold.
Get
1,000
and3.00
Sort by:
NO. I'm open to being convinced but I just don't see the point, I would never use them, I always want to look more closely at what people are saying in any comments, in the market description, look carefully at how its been traded so far and put a specific amount in rather than be encouraged to put in 10 one way or the other without giving it sufficient thought. I would think that we wouldn't want to encourage people to act in a more uniform and less thoughtful manner. It is maybe a tad more convenient than not having it, but really, not very much is it? I seem to be in the minority here though
@MartinRandall i suppose this is why PROB resolution is a bit better, it wouldn't settle at absurdly high percentage (though market manipulation would persist)
@JoyVoid I think PROB is better mostly because we don't just care about who has the majority, we care about how strong the consensus is in either direction.
Yeah, I think betting on the majority is more interesting if the poll is close, because it amplifies the result. IMO the key thing is zooming in on the range of likely outcomes (as I discussed in https://manifold.markets/jack/how-should-manifold-encourage-tradi). If the poll is somewhere in the 60-80% range, then you can make it more attractive to bet on by mapping 60 to NO and 80 to YES. You could also ask the binary form of this as "Will the poll result be > 70%" but I think that's less informative.
But that requires some prior knowledge of the range of likely poll outcomes. If you don't really know, which is the more common case, then there's a pretty high chance that the outcome will be strongly tilted towards YES or towards NO, and then a numeric/PROB market is much better than a majority-wins market.
Reducing accidental clicks. Display how much has already been bet. More prominent visual confirmation of bet, with an interval during which you can undo or edit the bet amount
And maybe editing uses a slider UI instead of a text box, which shows both amount and adjusted probablity since some people bet amount centric and some prob centric.