PAUSED03:25:03 UTC
Ratt Damon Paused · rolling stop
Wallet $0.62
Since start +$7.47of −$6.85 put in
Ledger realized −$44.12
Today realized $0.000 closed
Last SELL USPR $10.936d ago
SOL $83.16
$RATLABnot launched
RatLab

Ratt Damon's trading desk

Ratt Damon trades a real Solana wallet in public, $10 a bet since 13 Oct (earlier $6⁠–⁠$40), with a simulated rat brain (55,583 neurons, no living animal). The rules decide which coins are allowed; the simulated striatum's score picks among them; measured on 3 Nov, the pick is no better than a coin flip. So far he is up: the wallet holds $0.62 against the −$6.85 put in, +$7.47 (−109.1%). Bot ledger: −$44.12 realized over 485 closed trades, 50.5% won. Paused by its stop rule right now: no new buys since 5 Nov 04:12 UTC. Every trade links to its transaction.

Paused· rolling stop

Ratt Damon
Paused by its stop rule. He only sells until a person reviews him.no buys since 5 Nov 04:12 UTC

Wallet

Real money
$0.62
$0.62 cash (0.0075 SOL) · no coins held
5ahZfd…E97mu3
Put in−$6.85by the owner
Since start+$7.47−109.1%
Realized−$44.12bot ledger
Unrealized$0.000 open
Won50.5%245 of 485

In SOL: +⁠0.3946 SOL (−101.9%). SOL was about $17.69 when the money went in and is $83.16 now; the dollar result includes that move.

Bot ledger · realized, cumulative

Total since 17 Sep 2026: −$44.12 over 485 closed trades.

Bot ledger · realized (−$44.12)Real wallet minus money put in: occasional book backups, plus the latest mark (filled ring); the dashed line only joins them, nothing is measured in betweenNo new buys until a stop rule was lifted (the last one is still in force)Largest one-trade loss in view

The ledger leaves out buy-side fees and token-account rent, so it does not match the wallet. The wallet is the number you can check on Solscan.

Trade tape

Reallast 6d ago
Real transactions, newest first
UTCSideCoinResultTransaction
5 NovSellUSPR+$0.93
5 NovSellALON+$0.03
5 NovSellDEM+$0.04
5 NovBuyUSPR$10.00 in
5 NovSellTROLL+$0.03
5 NovBuyALON$10.00 in
5 NovBuyDEM$10.00 in
5 NovSell$WIF−$0.14
5 NovBuyTROLL$10.00 in
4 NovSellJEANPHIL−$0.87
4 NovBuy$WIF$10.00 in
4 NovSellGATO+$0.15
4 NovSellFLOAT−$1.57
4 NovBuyGATO$10.00 in
4 NovSellGATO+$0.13
4 NovBuyFLOAT$10.00 in
4 NovBuyJEANPHIL$10.00 in
4 NovSellCATE−$0.59
4 NovSellTROLL+$0.03
4 NovBuyGATO$10.00 in
4 NovSellFartcoin−$0.30
4 NovBuyCATE$10.00 in
4 NovSellJEANPHIL+$2.80
4 NovBuyTROLL$10.00 in
4 NovBuyFartcoin$10.00 in
4 NovBuyJEANPHIL$10.00 in
4 NovSellAODF−$9.49
4 NovSellGATO+$0.28
4 NovBuyGATO$10.00 in
4 NovSellPercolator−$0.56
All 970 transactionseach one opens on Solscan

Token

Not launched
$RATLAB
Contract addressAppears here, and only here, the moment it exists.
What holding gives
No share of Damon's wallet, no payout.
Creator fees
Their use is published here, with transactions, once there are any.
Get a rat
Not open yet.
Follow @RatLabAI

Open positions

Real0 open
Nothing open. Paused by its stop rule.

He buys nothing new until the stop is cleared. No new buys since 5 Nov 04:12 UTC.

rolling stop: the real wallet lost $9.70 over its last 14 days (20 trades) — only sells until a person clears "stopped"

What we learned · what grew

Mixedday 57 · two months on 17 Nov

From real records since the first real buy on 17 Sep. The brain is a simulation; the trades and the money are real.

Grew

real money: Real trades48550.5% won · −⁠$44.12 net (bot ledger)
real money: Safety rules added or fixed95 new rules, 4 fixes · 18 Sep → 3 Nov · listed on his Timeline
paper: Tokens logged47,18016,934 looked up · 1,482 traded · 344,488 lookups
paper: Ideas tested on paper138 rejected · 5 waiting · 0 proven

Learned, and changed

  • real money: 3 Oct: one rug (GOIF) took $38.66, the biggest single loss. Since 6 Oct a holder check runs before every buy; the worst trade since is −⁠100.0% (184 trades). Too rare to prove: one such loss in 301 trades before. 2 more since, so the check did not stop them.
  • real money, simulated: 30 Sep: the brain's high-conviction picks had won 78% on past trades, so their bets were doubled. On the trades since, they won 37.5% (88 trades), −⁠2.21% a trade, −⁠1.08% without GOIF. Bets are flat again: $10 a bet since 13 Oct (earlier $6–$40).
  • real money: 30 Sep: a rolling stop pauses buying when recent real trades are net negative. It fired 6 times, pausing buying 11.6 days in all. The breaker fired 3 times (23.0 days; 19.4 in one stretch until a person lifted it on 2 Nov). In all, no buying 62% of the time.
  • real money, hindsight: Coins born on pump.fun: −⁠$70.44 over 261 trades (−⁠$38.66 of it GOIF alone); the others: +⁠$26.32 over 224. Found after the fact, so no filter yet. Since 3 Nov: 31 and 21 trades, far too few to tell.

Did not grow

  • real money: Money: −⁠$6.85 put in, $0.62 now (−⁠109.1%). The same SOL simply held would be worth −⁠$32.20, so SOL's fall cost $25.35 and the trading added $32.82 (+⁠0.395 SOL).
  • real money: Since the 6 Oct changes: 184 trades, −⁠$36.36 in the bot ledger, 46% won, −⁠1.79% a trade (before, without GOIF: +⁠0.65%). The worst losses got smaller; results did not get better.
  • real money, simulated: The brain's conviction does not predict results: correlation +0.01 over 485 real trades (1 would be perfect, 0 is none).
  • simulated, measured 3 nov: Its memory does not keep: a lesson's trace shrinks by about two-thirds every 8 brain-minutes and is nearly gone within about 40, and its plastic wiring is within ±0.3% of how it was built.
  • simulated: Brain: 55,583 neurons since 14 Sep, never grown; one run, never reset; 45.6 simulated days in 56.7 calendar days.
  • paper: Proven edges: 0. Swarm 0 of 13 ideas, exit rules 0 of 8 (a random exit still leads on pooled paper and real trades), tournament none.
Brain internals (simulated)The two-month recordeach row: real money, paper or simulated

Research

Paper

Research rats test trading ideas on paper. A skeptical judge decides; nothing reaches the real wallet without proof.

Swarm614 paper trades

13 ideas0 proven0 promising8 rejected5 still collecting

Exit lab03:24:48 UTC

Other ways out, replayed on 98 of his real trades. Best so far: stop −10%; at 45 min out unless ≥ +5%; then trail 20% under peak; 24… (+$0.09 a trade against his actual exits), not proven. 3 of 8 rules discarded.

Slow lab03:05:27 UTC

Funding carry is the only candidate of 3; its paper forward test began 3 Nov.

Open the labpaper trades and replays, no real money

His browser

Real screen
gmgn.ai
Reading GMGN.AI on gmgn.ai.

How it works

  1. Sniff the board

    Liquid Solana tokens: momentum, order flow, volume and on-chain risk become smells.

  2. Read about it

    His own Chromium searches the name, X, the holders and who deployed it.

  3. Pick one

    Safety rules decide which coins are allowed. Among them, his simulated striatum's score picks one once it clears a bar; measured on 3 Nov, that pick is no better than a coin flip.

  4. Small, capped bets

    $10 a bet since 13 Oct (earlier $6⁠–⁠$40), a cap on what is open at once, and stop rules that freeze buying when the wallet slides.

  5. Sell by rule

    A time limit, a trailing stop, a stop loss, a take-profit or the callers leaving sells, never the brain. Each real close is sent to the brain; the trace is nearly gone within the hour.

  6. In public

    Every buy and sell is a Solana transaction anyone can open.