2026-11-04
Trench lessons — 2026-11-04
Key takeaways
- The "2-second trap": the wallets that keep winning are launch bundles that sell ~2 s in; copying 2 s late lost money
- Blizz "Pump Safe Migration": fall back to PumpSwap when the curve fills before the copy lands
- Smaller items
The "2-second trap": the wallets that keep winning are launch bundles that sell ~2 s in; copying 2 s late lost moneyevidence: C: one researcher's 7-day, ~150K-launch backtest
- What: repeat winners were bundle wallets buying 2–3 SOL in the create block, sniper bots piling in, and the bundle selling about 2 s later. 98% of first sells landed within 10 s. Bundle wallets: ~1.3x median sell multiple, losing on ~1 in 8 coins, a few making tens of SOL in both halves of the week. Only ~1.7% of those coins graduated. Dev accounts were disposable; the bundle wallets kept returning. Entering 2 s after them lost money in the backtest. Caveats from the author: one week, collector misses trades, links inferred.
Blizz "Pump Safe Migration": fall back to PumpSwap when the curve fills before the copy landsevidence: C: promo thread
- What: Blizz now routes a buy to PumpSwap if the bonding curve completes before the order executes (including dev full-curve buyouts at launch), instead of failing. Costs extra compute units.
Smaller itemsevidence: no action
- Token-2022 fee check (C): a reminder that an advertised transfer-fee rate isn't the result; check rate, max cap and fee-update authority (a mint can raise its fee later).
Sources
- https://x.com/ustadimiz/status/2117772839956721673
- https://pumpdata.fun/report/2026-11-03.md
- https://pumpdata.fun/stats/2026-11-03.json
- https://x.com/hoanhphuc01/status/2117928090739605824
- https://x.com/VectoraAI/status/2117936091701682183
- https://x.com/XPOSED_sol/status/2117779568002039876
Evidence grades: A = large dataset, B = official docs, C = vendor/thread/small sample. Research notes for education only. Not financial advice.