How will Manifold address share rounding making some bet sizes strictly worse than others?
2
280Ṁ3resolved Jun 5
ResolvedN/A
1H
6H
1D
1W
1M
ALL
94%Other
1.9%
This won't be fixed
1.9%
Automatically floor bet sizes to smallest with same profit
1.9%
Support for fractional shares & fractional M$
Because share amounts are rounded, some bet sizes are inefficient. E.g. for a yes/no market currently at 90%, $9 buys 9 YES shares, $10 buys 11 YES shares, $11 buys 12 YES shares. In this case, $10 is the lowest reasonable bet amount, and you should never bet $11, since it risks strictly more than betting $10, but you still only profit $1 if correct.
The current UX for this seems to be "manually try a bunch of integer bet sizes and find the lowest one with a given profit".
I see two obvious solutions: either automatically support this "rounding bet size down" behavior (so I don't have to do it manually), or supporting fractional shares (and, correspondingly, fractional $) - but perhaps there are others!
This question is managed and resolved by Manifold.
Get
1,000 to start trading!
Sort by:
The presupposition of this question is false. Manifold displays the floor of share and M$ numbers to users, but doesn't actually round anything. Full floating point precision is there under the hood!
I guess we could do a better job communicating this to users. Perhaps showing more digits in a tooltip?
@SG Oh, awesome! Purely for discoverability I think displaying some decimal points in a tooltip would be good, yeah. Particularly when floor(shares purchased) == M$ bet amount, but always displaying one decimal point would at least cue the user that full-precision floats are being used, if there was also a tooltip?
People are also trading
Related questions
How big will the largest individual bet on manifold be in 2024? (Ignoring exploits, bugs, and wash trades)
When will Manifold resolve this bet?
Will Manifold add the ability to edit the amount of your bet within seconds of making it by the end of 2025?
10% chance
Should Manifold betting be zero sum?
POLL
If Manifold changes loans to stop using cost basis, will this change retroactively affect earlier bets?
65% chance
Will Manifold stop having obviously wrong probabilities on markets with >=50 traders by the end of 2024?
7% chance
Are the fees on Manifold bets too high?
POLL
What is your Manifold betting strategy?
POLL
What types of Manifold markets should users prefer to bet on?
POLL