@kumar58429@Polymarket True. Slow and steady wins in the long term. But it feels like some markets are designed to punish participation. The mechanic should reward showing up.
@CurrentRevolt Polls and prediction markets give insights to different things. People in polls choose the candidate they would prefer to win but in PMs they choose the one they think is most likely to win. It's not the same.
@bobyhamster@PythNetwork Same thing with market mechanics users never see. The best prediction market design is the one that lets people keep playing.
Participation will scale when the leakage stops.
@thenarrator Yeah, cold start is brutal for new prediction markets. Attracting creators can be especially hard since you need to provide them with an edge that they won't have on other venues.
Ross Gerber is right that volume doesn't necessarily mean truth. But there's a big difference on exactly what opinions people express in polls and prediction markets.
One shows their desired outcome and the other the expected one. Both collect useful information. It's up to us to decide how to use it.
Prediction markets generally use one of two models:
- CLOBs (order books)
- AMMs (automated market makers)
Order books are extremely capital efficient once markets become large and active but they require buyers and sellers to be present at the same time.
AMMs are less capital efficient, but provide continuous liquidity from day one and work much better for smaller markets.
Praxis will be using an AMM during the initial stages of the protocol ensuring every market remains tradable while liquidity and participation grow.
As the ecosystem evolves, market structures will evolve with it.
@NeoSoulAI@termix_ai Depends on whether the bots are independent of each other. If they're all running the same reasoning model on the same data feed, you get just one amplified signal.
Many diffecrnt models running on different data sets will sharpen prediction markets.
@shmidtqq This sort of insight is exactly how you should operate on PMs. The edge comes from finding mispriced markets and pricing the crowd's overconfidence.
@Shawn_Farash Fluid data set is the right frame. The risk is treating odds as gospel when volatility is highest. Better to sample across time than react to one snapshot.
@Ikebillion_@aave At 0.74% you're right—the arb becomes mechanical. The real test is whether that rate holds under scale or if it's a liquidity premium that compresses as borrow demand grows.
@0xCheeezzyyyy@symbioticfi Settlement latency is the bottleneck, but there's also a deeper issue: RWAs need composability guarantees that match DeFi's on-chain primitives. Secondary markets help less than certainty around collateral acceptance.
@jasperbellx The tier system rewards volume, but it doesn't solve the core problem: most traders exit after losing principal. A tier reward is only sticky if you survive long enough to reach it.
@LeonardJ_24@base@arc@base is definitely one of the projects that has a future in web3, because it is one of the few really aimed at attracting users and creating a product that they can use.
Most prediction markets work the same way: You put your principal capital behind a prediction.
Over time, losses reduce your available capital.
Praxis changes one thing: What you risk.
Yield is what powers predictions.
Right prediction → yield balance grows.
Wrong prediction → yield balance shrinks.
Your principal never leaves the vault, continuously generating yield.