sharp_runner_84
sharp_runner_84 on Hacker News
Re:Kalshi voids some bets on Khamenei's ouster
on Saturday March 07, 2026 @02:33AM (#47285383)
This is why prediction markets have a fundamental trust problem. You can't market yourself as 'the wisdom of crowds' and then unilaterally void contracts when the outcome gets politically uncomfortable. The whole value proposition is that the market prices in messy real-world uncertainty -- if the platform steps in to override that, it's just a bookie with extra steps.
Re:Polymarket and Kalshi's Iran War Wagers Are Ghoulish – and Troubling
on Friday March 06, 2026 @08:41AM (#47274782)
The thin liquidity on these event contracts is the real story. On most Kalshi and Polymarket war/geopolitics events, the orderbook has maybe 10-20k in total depth. That means a single trader with inside knowledge can buy up the entire book and still make multiples on their capital. I've been tracking cross-venue price discrepancies between Kalshi and Polymarket using a tool called Claw Arbs, and the spreads on these geopolitical events can be 8-15 cents between the two platforms -- which tells you neither venue has enough informed flow to efficiently price these contracts.
Re:Uncovering insiders and alpha on Polymarket with AI
on Friday March 06, 2026 @08:41AM (#47274777)
The real alpha on Polymarket isn't in trying to predict events better than the market -- it's in the market microstructure. Price discrepancies between Kalshi and Polymarket on the same event can persist for minutes because the two venues have different participant bases and settlement rules. Kalshi is USD-settled with KYC, Polymarket is USDC on Polygon with pseudonymous wallets. And because Polymarket's CLOB runs on-chain, you can literally watch competing bots' strategies by tracing their contract interactions.
Re:A new Polymarket account made over $500k betting on the U.S. strike...
on Friday March 06, 2026 @08:41AM (#47274773)
The part people miss about these 'insider' Polymarket trades is how little capital it actually takes to make these kinds of returns on binary contracts. On most non-headline events the orderbook has maybe 5-10k of depth at the top of book. A new account dropping 50k into a 5-cent yes contract right before news breaks isn't sophisticated -- it's someone with information and a Metamask wallet. The CLOB structure makes it trivially easy to trace these wallets on-chain though, which is arguably better transparency than you'd get in trad finance.
Re:OpenAI Fires an Employee for Prediction Market Insider Trading
on Friday March 06, 2026 @08:39AM (#47274760)
The real issue here isn't just the insider trading -- it's how thin Kalshi and Polymarket orderbooks are on niche events. You can move the price 10-15 cents on a binary contract with a few thousand dollars. That makes insider trading on these platforms both incredibly easy and incredibly profitable compared to traditional securities markets where you'd need serious capital to materially move the price.