Currently, options in a free response question will split off from "Other" at an equal probability (or approximately equal) to the "Other" option itself, while also increasing the overall probability of "Other" + new option when you add them. For example, the first answer added always starts out at 50%, and if you add a second option before trading on it, then the probability of "Other" + second option will be about 2/3, increasing from the 50% that "Other" was at before you added it. This leads to a problem where users are not incentivized to add new options to a free response market, especially early on. The old free response markets were very flawed, but the one benefit they had was that new options started out at a small probability, so users were encouraged to submit options that they actually thought might pay out. Now, in most free response markets, it is a terrible idea to submit an option in order to bet YES on it (esecially early on), since there is likely no option that's more than half as likely as all "Other" possibilities combined to pay out. Of course, users could submit options in order to bet NO on them, but that creates a perverse incentive where users are incentivized to submit options that they think are extremely unlikely to be chosen (or even just nonsensical options), rather than options that they actually think are plausible. Users might also submit options with no intention of betting on them, simply because they think they're plausible, but this costs Ṁ25 to do, and the user isn't compensated for it in any way, so there is an incentive against this.
One way to fix this problem would be to have new options added split of at a lower probability than "Other". This would make it more likely that the person submitting the answer actually believes their answer to be more likely than the probability it split off at, and bet YES on it. For example, if the new options only split off at 10% of the initial "Other" probability, then the first choice submitted would start at 10%, a much more plausible probability than 50%, and later option could split off at very low percentages, giving the users submitting them an opportunity to make a substantial profit if they turn out to be right.
This market will resolve YES if, at any point before the end of October, options submitted to a free response question split off from "Other" in a way that gives "Other" a noticably higher probability than the new option submitted, or if it is possible for the user submitting the option to decide for themselves what probability they want it to split off at.
There is a Discord thread about this and other ways to potentially improve incentives for free response markets here: https://discord.com/channels/915138780216823849/1142225031422820412
For other possibilities about ways to improve incentives, see:
/JosephNoonan/will-manifold-give-users-who-submit
1,000🏅 Top traders
| # | Name | Total profit |
|---|---|---|
| 1 | Ṁ138 | |
| 2 | Ṁ59 | |
| 3 | Ṁ6 | |
| 4 | Ṁ1 |
People are also trading
This market will resolve YES if, at any point before the end of October, options submitted to a free response question split off from "Other" in a way that gives "Other" a noticably higher probability than the new option submitted
I added a new answer and it split off at a much lower probability than Other. See screenshots. The new answer got 15% and Other was left at 27%.
@Eliza yeah, you are right. Not sure why it didn't work in my test markets. Maybe there are some circumstances that must hold for it to happen. Which means that it still hasn't really solved the problem, but it is enough for a YES here.
@JosephNoonan If you read the comment in the code link to GitHub, you will see it depends on the ratio of yes and no shares in the Other answer and how they are distributed. In the case of your test market, it behaved exactly like 'before' because there were enough No shares (5b in that code comment), but in the Main Event market it did (5a).
@Eliza I actually noticed this about a day earlier and kind of went "huh" and then forgot about it and this market until now...seems like a good change to me.