I’ve been working on a prediction market platform for quite a bit of time now (built with clojure of course). There’s still a long way to go but since everyone is watching the World Cup, it’s time to take some baby steps to do a test launch with a couple of topics that people can start playing around with.
So anyways… the launch taught me a couple of things:
The main thing is to not have it in my head for people to be rushing in to sign up, and the second thing is that I have to be in constant communication with people that may potentially want to use it and then build more features for them.
Ultimately, it’s a beginning.
I’m reaching out to see if there is any interest to build some toy trading bots for these prediction markets. Basically, it’ll probably involve scraping a live data source, doing some machine learning and then placing an order in the system. I spoke to @daslu a couple of months back but it was still early. It’s still early but at least I can start initiating conversation with the community to gauge interest.
I’m reaching out to see if there is any interest to build some toy trading bots for these prediction markets. Basically, it’ll probably involve scraping a live data source, doing some machine learning and then placing an order in the system.
@zcaudate would you like to tell a little bit about what you are hoping to do in terms of machine learning?
Just to make things clear. I’m not in a position to pay anyone right now as Statstrade is not (yet) VC backed - but I’ve got decent expectations for the platform and I’ll be really grateful for those that help make it succeed.
Prediction markets to me are really interesting because you don’t really a middleman to bet any more. Any topic of interest can be created and then users on the platform are just betting with other users on whether an event will succeed or not succeed. I built statstrade for the possibility for anyone to create a topic and to run their own prediction market. The hardest problem (and it’s something that I’m working on) is to be able to bring on users that 1. want to use the platform and 2. are interested enough in the topics on there to engage/place a bet.
In regards to the toy trading bots… I’m not a trader, nor a market-maker nor a bookie. So I have no clue about trading algorithms. I was hoping that if there where people interested, I can maybe go over the API. As a group, we can create a topic that might peak the clojureverse’s interest:
Will there be a new clojure release in the next 3 months?
Will there be more than 100 posts on clojureverse in the next month?
Will there be more than 20 swear words used on clojureverse in the next month?
Once a topic is created, the price kinda acts as a ticker. It could be absolutely anything and it’ll be really interesting to see what people can come up with. The bot could be something super simple heuristic or super complex Deep Learning model.
I’m not sure if we talked together in any of the jointprob sessions but Hi! I find prediction markets very interesting too @zcaudate !
How is the project progressing? I am getting a 404 from the links above.
Have you considered signing up to the Y-combinator startup school course which seems a good resource for people who have ambitions in this kind of direction.
It seems to me that the biggest challenge for many online business ideas is to capture the initial interest of people and get a seed group of users.
Lots of work still needs to be done. You’re right in terms of users - or rather, who pays and why would someone want to pay. It is a problem and a completely different problem altogether from prediction markets.
I’d love to get in touch with the YC people but will need a bit more prep. I think now, especially in the current economic climate, people are going to be more cautious - and that just means that pathway to profitability has to be made quite clear - which is not a bad thing at all.
