A random idea popped into my head the other day, so naturally I discussed it with my "hey can we code this" guy 
Prediction markets are making a big buzz these days, a lot of volume happening on polymarket where people can basically "bet" on anything. I'm not entirely too sure how everything works, nor would this be similar to it as we're not really prepared nor do I think we want to be in terms of taking people's coins and opening a casino, but maybe down the line someone could/would be able to when we get smarter contracts on hive.
We could however start out with a "soft" prediction market, albeit a bit controversial to a degree hence I'm writing this post to see what people would think about it. Either way, here's a rough sketch of how it could work, save your "OMFG HOW DID U THINK THIS WAS A GOOD IDEA?!" comments for after you read it, please. :D
Prediction markets through voting.
I guess this should already give you an idea that this isn't meant to be some kind of big league high roller project, but more of a fun and new thing to use your voting power on and potentially even power up some hp or winnings for the long run!
It could either use polls or comments to cast your votes.
- If you vote on the polls it doesn't cost any voting power, but you'd then still need to vote on the post/comment to participate and "bet" on the poll choice. Thus the bot would need to check which option in the poll you chose, your % of the rewards if you picked the correct choice compared to the total rewards of the post/comment along with the rewards from users that voted on the wrong option(s).
It would require a bot to track the poll/votes cast.
It would require a service/person to confirm the correct outcome.
It would require a payout bot.
This would basically be a "you have nothing to lose" outcome, if you're wrong, you still get curation rewards just like voting on any other post (as long as they or this doesn't get downvoted) which is what makes this controversial. Although times are a bit different now where we see stakeholders vote on a variety of things that aren't always content, maybe there's room for more things?
Another slightly more complicated version of this could exist where there would be something to lose, this one would require the votes not be in the poll option however but on comments.
This version would require outcomes to be determined before comments pay out (within 7 days), thus long-term predictions like "when will Trump nuke the planet?" may not be feasible for it. This version would simply have a larger stakeholder(s) use their downvote mana to downvote the wrong predictions. Thus, those who bet their vote on the wrong outcomes would not receive curation rewards.
This version would however give winners only their own bet 2x since the bets of the losers would be downvoted. To make up for this, additional rewards could come from the prediction post itself if they were upvoted or required to be upvoted as well by participants/others.
There may be more options/gamification to this. Post rewards of many predictions could be pooled for weekly/monthly prizes distributed towards accounts based on their history of predicting outcomes.
There could also be restrictions in place where we'd only validate one vote per day as valid by an account as to not take too much voting power from other things such as content/burn posts/hbdstabilizer posts, etc.
Either way, I'm just opening up discussions to the idea with this post, if people think it'd be fun and there's not much bad actor activity in it, it could be fun to try. The service itself could take a small fee mainly to take care of liquidity to provide liquid hive to winners it receives as stake but also to reward those who coded and maintain it along with helpers down the line if it becomes very active on a daily basis.
The fun thing with hive is that prediction markets would be tied to usernames and their history more compared to other blockchain's that hide behind anonymous addresses and you can never really be sure who's who and who's voting for what. This could sway certain users one or the other direction and see if "wisdom of the crowd" works with the social layer on top. We could even have people engage in the prediction posts about the predictions and later the outcomes which isn't something you could see in other places as transparently and openly.
These reasons were some that made me think maybe this wouldn't be such a bad idea, and as mentioned, down the line someone could potentially evolve the idea to take bets that aren't just "free upvotes".
Let me know what you think!
