PUMP RADAR · PAPER TRADING · ARCHIVE
Paper trading
Three virtual wallets of 3 SOL followed every call of our Telegram bot with rules fixed in advance. Nothing was bought: each trade was replayed on pump.fun's one-minute prices, entering 60 seconds after the call and paying pump.fun's fees and slippage. The point was to see, honestly, whether following the calls makes money. On our first 83 calls, buying every call a minute late lost money with most ways of selling; this page tested, on new calls, whether skipping the riskiest ones changed that. It did not: all three wallets lost, and the book was closed on 26 Oct 2026 with the numbers below as its final record.
Loading the wallets…
Early-G test (bot only)
Loading the test…
Day by day (UTC)
Latest trades
Each wallet started with 3 SOL and put 0.1 SOL in a trade, at most 3 at once, one per coin; after losing 0.5 SOL in closed trades in a UTC day it stopped entering until the next day. Entry was the open of the first one-minute candle that started 60 seconds or more after the call (60–120 s late, like a person or a simple bot). Costs: pump.fun's 1.25% fee and 1% slippage on each side (on PumpSwap the fee at these market caps is about the same), plus 0.001 SOL network fee per transaction. Inside a minute the order of high and low is unknown, so a stop counts before a target, and a gap down fills at the minute's open; real fills on thin curves can be worse. A position still open when the book was archived was closed at the last minute read, with "closed at the archive" as its reason: it counts in the SOL balance but not in the trade record. These rules are version 3, frozen before the first trade (the plan from our study of 83 calls; version 3 uses pump.fun's real 1.25% fee). Not financial advice: most pump.fun coins lose all their value.