It seems like you could plausibly use sports betting as a proxy for measuring/estimating the marginal effect of dumb money. It also seems plausible that you could have some exogenous shocks to the interest of ‘dumb money’ in particular outcomes (e.g. random rescheduling pushes a match to a more or less desirable timeslot) that would let you measure their effect on accuracy
Isn't the market for Jesus Returning just a bet on interest rates? If you are certain the answer is No, but paid a year later, you should only be happy buying with a No price of 1 minus the interest rate. Buying Yes is equivalent to Selling No, so if you think that No is over-valued, ie that interest rates will rise soon, then you can profit by buying Yes today (and selling it when they do rise).
-I think it's fairly clear that more dumb money would attract more smart money... that is the way it works in every other gambling environment. It would strike me as bizarre if more smart money meant lower accuracy.
-I also think dividing traders into "dumb" and "smart" leaves a lot of nuance out... obviously some prediction market players are savvier than others, but I tend to think that everybody knows something. In many fields (but not all) there is a "wisdom of the crowd" effect where averaging the opinions of many people leads to more accurate predictions/better performance than you get from an average individual. I would guess that this is generally true in prediction markets, because "morons" are generally morons in idiosyncratic ways that average out, not in systematic ways that create bias.
I think it's also important that there are actual hedgers in these markets. Back when student loan forgiveness was being contemplated, I had $10k of exposure to student loans and paid off about $4k by betting on Kalshi against student loans being forgiven. I considered, but ultimately didn't bet against ACA subsidies being extended (I sell ACA health insurance, the insurance companies I work with have already cut commissions for 2026 so I also have exposure. I wonder how many people betting "no" receive subsidies and are hedging the likelihood of it not being extended?).
Regardless of if the traders on Kalshi and Polymarket are informed or not, these event contracts offer the traditional advantage that futures markets typically offer: for producers and consumers of commodities to hedge actual exposure. If I am informed I can better calculate expected value. But if I have some economic exposure maybe I am just happy to take some chips off the table for now, and market makers will take the other side because they are more diversified than me in economic exposure.
For this purpose, I think that the FOMO inducing hype can be possibly helpful in encouraging some potential hedgers to decide to effectuate their desire to hedge.
I think there may be some degenerate gamblers on here too. But I wouldn't be surprised if the prediction markets are off more because producers and consumers are hedging their actual economic exposure than because of degens.
Polymarket doesn't charge trading fees; they don't profit when people make trades. (I think Circle cuts Polymarket in on some of the interest on USDC held by users, but they give most (all?) of that back to users through Holding Rewards.) (Polymarket US (which is separate from Polymarket) charges a 0.01% fee to the taker on the total contract premium, but it's not available to the general public right now.)
> but are instead buying it from a liquidity provider
This isn't true for Polymarket anymore. PM used to use an automatic fixed-product market maker, but today when you buy YES/NO shares you're buying them from other users through an order book.
It seems like you could plausibly use sports betting as a proxy for measuring/estimating the marginal effect of dumb money. It also seems plausible that you could have some exogenous shocks to the interest of ‘dumb money’ in particular outcomes (e.g. random rescheduling pushes a match to a more or less desirable timeslot) that would let you measure their effect on accuracy
Very interesting. Hm. Will consider.
Isn't the market for Jesus Returning just a bet on interest rates? If you are certain the answer is No, but paid a year later, you should only be happy buying with a No price of 1 minus the interest rate. Buying Yes is equivalent to Selling No, so if you think that No is over-valued, ie that interest rates will rise soon, then you can profit by buying Yes today (and selling it when they do rise).
2 notes:
-I think it's fairly clear that more dumb money would attract more smart money... that is the way it works in every other gambling environment. It would strike me as bizarre if more smart money meant lower accuracy.
-I also think dividing traders into "dumb" and "smart" leaves a lot of nuance out... obviously some prediction market players are savvier than others, but I tend to think that everybody knows something. In many fields (but not all) there is a "wisdom of the crowd" effect where averaging the opinions of many people leads to more accurate predictions/better performance than you get from an average individual. I would guess that this is generally true in prediction markets, because "morons" are generally morons in idiosyncratic ways that average out, not in systematic ways that create bias.
I think it's also important that there are actual hedgers in these markets. Back when student loan forgiveness was being contemplated, I had $10k of exposure to student loans and paid off about $4k by betting on Kalshi against student loans being forgiven. I considered, but ultimately didn't bet against ACA subsidies being extended (I sell ACA health insurance, the insurance companies I work with have already cut commissions for 2026 so I also have exposure. I wonder how many people betting "no" receive subsidies and are hedging the likelihood of it not being extended?).
Regardless of if the traders on Kalshi and Polymarket are informed or not, these event contracts offer the traditional advantage that futures markets typically offer: for producers and consumers of commodities to hedge actual exposure. If I am informed I can better calculate expected value. But if I have some economic exposure maybe I am just happy to take some chips off the table for now, and market makers will take the other side because they are more diversified than me in economic exposure.
For this purpose, I think that the FOMO inducing hype can be possibly helpful in encouraging some potential hedgers to decide to effectuate their desire to hedge.
I think there may be some degenerate gamblers on here too. But I wouldn't be surprised if the prediction markets are off more because producers and consumers are hedging their actual economic exposure than because of degens.
> thus profit when people make trades
Polymarket doesn't charge trading fees; they don't profit when people make trades. (I think Circle cuts Polymarket in on some of the interest on USDC held by users, but they give most (all?) of that back to users through Holding Rewards.) (Polymarket US (which is separate from Polymarket) charges a 0.01% fee to the taker on the total contract premium, but it's not available to the general public right now.)
> but are instead buying it from a liquidity provider
This isn't true for Polymarket anymore. PM used to use an automatic fixed-product market maker, but today when you buy YES/NO shares you're buying them from other users through an order book.
Those interested should see:
https://open.substack.com/pub/aashishreddy/p/prediction-markets-objections
A Manipulator Can Aid Prediction Market Accuracy
https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1468-0335.2008.00734.x