Leo Labs Big Group 4/26 Group Member Discussion Highlights (Recent 380 Messages)

━━━━━━━━━━━━
1️⃣ Kelly Criterion was collectively rejected in the group — unanimously switching back to fixed positions
Group discussion on position management for over 30 minutes:
• "Kelly needs to know the true probability, retail traders simply can't calculate it"
• "Kelly is an optimization method under stable profit conditions, it doesn't determine profit and loss itself; claiming it as an essential system component is nonsense"
• Practical consensus: fixed position size + increase principal after profit + don't change parameters casually
• A famous quote: "If you are God, then you can use the Kelly formula"
💡 Drawdown and wanting to change parameters ≠ position size issue, that’s a problem with the strategy itself.
━━━━━━━━━━━━
2️⃣ Extreme price strategy at 0.99 — statistical trap in verification window
Group members asked, "Can you survive buying at 0.99," leading to the best statistical discussion:
• 99% reversal probability = 1/1000
• 98% reversal probability = 5/1000
• To buy at 0.99, you need a true win rate >99%, but retail traders can't verify this in the short term
• An intuitive analogy: "I have 2 coins, one with 99% heads vs 99.5% heads, you need to flip hundreds of times to tell which is which"
💡 Your strategy might not be "impossible," but you won't last until you can verify whether it works.
━━━━━━━━━━━━
3️⃣ Real numbers for simulation vs live trading gap
• Simulated ROI 2.5% → Loss in live trading
• Simulated ROI 5% → Still a loss in live trading
• Simulated 80% win rate → Live trading execution rate less than 40%, only bad orders are filled
• Live trading threshold: "Place orders with 50% failure + fees + 2c adverse slippage and still make a profit"
💡 Simulation assumes you click and it immediately executes at the best ask; in live trading, both conditions are not simultaneously true.
━━━━━━━━━━━━
4️⃣ EOA wallet gas fees surge + rebate mechanism
• Previously, gasLimit 15,000 was enough; these days, 50,000 is needed — due to many users switching to EOA
• PM rebate = taker fee 20% (dynamic, originally 100% → now 20%)
• EOA referral rebate entry point is hard to find for most group members
• PM often doesn't follow rules (one member owed $500)
💡 Redemption frequency limit (~100 times per day / once every 15 min) + merge not counting as relayer transactions = outdated early documentation.
━━━━━━━━━━━━
5️⃣ V2 cutover on 4/28 — most people haven't tested before going live
• "I feel like my code will have a bunch of issues during the 28th migration"
• "V2 strategy is completely rebuilt, finally no longer relies on flipping payoff"
• "V2 Rust SDK is ready, why can't I switch yet"
• Most choose "Upgrade on Tuesday, wait and see"
💡 V2 matching mechanism involves significant changes; market maker/taker strategies will likely need review.
━━━━━━━━━━━━
📌 Three noteworthy points today
1. Kelly formula ≠ usable by retail traders — can't calculate true probability = formula fails
2. Verifying 0.99 strategy requires 1000+ trades — short-term failure to verify doesn't mean it won't work
3. EOA gas from 15,000 → 50,000 — old scripts need review
---
Every day, I compile the highlights of Leo Labs group member discussions into a daily report and post it in the channel.
💬 Big Group
📡 Daily Highlights Channel
🔗 For those interested in data research and analysis addresses →