Anyway, there is a huge difference between prediction market and traditional investing, which is capital vs wagering. If the bet is lost the entire wager amount is gone like 100% pf the capital to relate with investment but it is unlikely to happen in traditional investment, it can go down 20%, 50% or even 99% but 100% doesn't happen everyday.
Yep, I've covered this topic in
this post. You could compare a lost bet also to a bankruptcy.
I've re-thought a bit about calling it "investment" and now I think it's indeed closer to a complex trading strategy. If you wanted to use it as a long term investment, you would need to cut losses and re-balance all the time.
Originally I opened this thread because I considered prediction markets an easy way to bet on scenarios which aren't covered by a traditional investment. If the amount of bets related to your scenario (e.g. AI progress, political shifts, even certain develpment in sports) is high enough you can diversify your "investment" into several bets placing small amounts on sub-items with some of them having a higher probability to succeed and others less.
I don’t want to mix things up, but think about poker for a second. In tournaments, you start with a stack of chips, and your goal is to keep growing that stack because, over time, the “investments” (your bets) increase in value. The dynamic forces you to take risks, because players who don’t (those playing a tight style, only premium hands) eventually get pushed out.
I think this can definitely also be applied to prediction markets. If we assume the bettors are relatively good in pricing the probability, the profit expectation you have is usually only better than the fees if you also take some risks and bet against the mainstream.
Now I'll apply this to the concept on "creating an investment strategy using prediction markets". As I wrote in my answer to Findingnemo above in this post, the idea is to bet on a (economic, political, etc.) scenarios that could materialize in less than a year. So you'll have to rebalance the bets regularly. You can however bet on related events / developments. For example, if your idea is to bet on "AI progress", you can bet on different companies' announcements related to AI progress.
Of course if you have to always take into account that you have to take more risks than the "average bettor" to really be able to profit, you should know your matter either very well (which is borderline insider trading) or bet against the mainstream in several cases. On average this makes this kind of "investment strategy" quite risky and you will face several total losses. The idea is then to find bets that could complement each other a bit, for example in the AI scenario bet both on OpenAI, Google and Anthropic announcements if you feel that if one of these companies has an edge it could lead to less success of the other ones' announcements.
But on the other hand, if you want to bet on such general scenarios like AI progress, prediction markets are one of the simplest ways to do that. The alternative would be to search for stocks related to your scenario, but this is less "precise" because companies can switch their business model or focus on different products if their own "bets" don't succeed (think of examples like Bitcoin mining companies getting into AI, or Strategy with its software-to-Bitcoin evolution).
An investment is usually a long-term process, shaped by multiple factors; many of them implicit (and that word matters here), not just pure, direct probabilities like in a bet. [...] I guess the simplest way to put it is: one is about owning something, and the other is about pricing a probability.
The interesting thing is just that these multiple factors could be condensed into different bets. The "implicit" factors can also be present in the probabilities assigned to bets, and at a first glance I don't think it's more complex to "decipher" then in prediction markets than in the stock market. If you think differently here let me know an example
