On the prediction market insider trading debate, I was talking with a couple of new finance friends yesterday. One was for, and one was against. We had a good talk about why, intuitively, it seems like insider trading is good for truth seeking but it's actually bad.
Prediction markets have two primary utilities. One is accurate pricing of future events, and the other is hedging. These markets only work if there are smart people incentivized to trade on them and make them efficient.
Currently, and this is likely to change, there is not much hedging going on. So the main incentive for people to address inefficiencies is having counterparties that price things poorly and making money off of that bad pricing.
In order for sharp forecasters to trade on these markets, they have to be confident that:
1. The trading platform is secure
2. Rules are fair and understandable
3. Their counterparties don't have an insurmountable information edge
1. and 2. are mostly there, but there are still some hiccups. With 3, however, if traders think their counterparties are insiders, then they don't want to trade against them. Who can beat an insider? No one! So they don't trade in those markets. If insider trading is allowed in all markets, and it's unknown which are insidered or not, then the risk profile for traders increases, and some really sharp people may not trade any markets! (Except for markets that are insider-proof, of course.)
So now the insiders are the only ones setting prices Which is ?maybe? theoretically fine if there is an insider for all 20,000 markets constantly updating prices, but that's not currently the case and unlikely to ever be the case. The model in which prices are set by a combination of insiders+sharp forecasters is tenuous at best. And now with only insiders setting odds the pricing data is useless because it's illiquid and you don't know if the insider is manipulating it or not!
Obviously, if you're a hedger, you don't want to hedge against a manipulated price because it wouldn't be fair or reflective of real probability. So actual optimal price setting and fulfilling the promise of these markets only comes when all participants are confident there are no insiders, which in my opinion means consistently banning insider trading.
"In the context of prediction markets, there's a live debate on whether or not insider trading is even good or bad.
The reason, basically, is because the value of these markets is the idea that they can create accurate odds for everyone to see.
Of course, if there are people who are participating in the markets who have good information that they can trade on, that is actually going to create more accurate pricing in the markets.
That can be very valuable if people want to know the real odds of, let's say, an election."
— @pjchougule
Agreed with this take! I think given the wider range of underliers there's just a huge expansion in the types of fraud or manipulation that can be done (maybe?), so it becomes a more complex problem. Can solve it good enough tho
Context: a derivative becomes a useful thing for society to trade
Problem: trading the derivative incentivizes manipulating the underlier
Solution 1: ban trading the derivative
Solution 2: ban manipulating the underlier
Solution 3: accept that underlier manipulation is just part of the game, and should be priced in accordingly
tradfi has accepted #2, defi has accepted #3 because they can't enforce #2, but people seem heavily inclined to jump to #1 for prediction markets? Why is that?
I personally feel that 2 > 3 > 1 and would much rather live in a world that doesn't place blanket bans on incentivizing truthseeking when there are other options on the table
@giaset As someone with a lot of CLV on love island markets discovering you are married to a former contestant blew my mind. Could be fun to talk the finals together
I'm currently #1 for mentions volume a week into the month. I feel like I unlocked a new sizing gear. Also going big again this month in culture and some politics. Last month @CarstenPredicts and I were 2nd and 3rd in culture. This month I want to repeat and get top 3 mentions
Got down around $260k in predictions on the USA v Belgium mention last night. Around 200k of that was pre. Bummer to see our lads go (and the last host nation), but stoked to win large on my biggest mention to date. Unfortunately with USA/Mexico out I suspect volume to be sapped
June was my best month ever, eking out a narrow win over February. Still working on sizing bigger on the best positions, but my capital use was generally really high this month, which was a great improvement. Still got mogged by @CarstenPredicts
Incredible write-up on today’s Colorado races. Wish I’d seen this earlier.
Even expressed skepticism about Bennet, avoiding the upset.
One of the few politics sharps consistently putting out detailed, thoughtful election analysis.
May Prediction Markets Recap:
Over the past 30 days I have used my math and coding skills from MIT and my finance skills from Goldman Sachs to lose $2,900 on @Kalshi Mentions.
Mentions Sharps who rinsed me include:
- high school children
- youtubers*
- retired casino and food service employees**
Had a winning month in sports/politics by picking off:
- SIG
- @kylekuzma
AI cost via Claude:
- $200 via Max Subscription + $100 API fast mode
- $2,000 via market linking error when i fired orders for 5,000 contracts at 99 cents on the wrong event (skill issue)
As JFK once said: "we choose to trade Prediction Markets in this decade, not because they are easy, but because they are hard."
*including @PredMTrader and @locksy
**including @Foster and @catboyautist
I took a position on Lee Zeldin No after watching the same. Trump had some praise for Blanche at the Warsh ceremony today too. We'll see how it develops!
Who will be the next Attorney-General? I took a position on Lee Zeldin Yes today after seeing the EPA chief show off for Trump at an Oval Office event. Afterwards, the two had a closed door lunch together. Job interview?👀
In the last 3 days I’ve made -$53,569.73 by picking up money off of the ground
That works out to -$6,517,650.48/year. Why aren’t more people doing this?
@GrousARK@datadashboards IMO open interest is the most important metric for tracking activity. Volume means a lot to Kalshi's fee revenue, but isn't actual market depth/size most meaningful in the real world application of these markets?