Why Ghostbag
Every memecoin paper trading tool can show you a price and a fake balance. The question is whether the number it shows you would have survived contact with a real block. Here is exactly where Ghostbag differs, and how to judge any simulator you try.
| What decides whether a trade made money | Typical simulator | Ghostbag |
|---|---|---|
| Live price from public market data | Yes | Yes |
| Platform fee charged by your terminal | Rarely | Per site, real rates |
| LP / AMM fee on both sides | Rarely | Per venue |
| Network fee, priority fee and tip | No | Charged from balance |
| Token account rent on a first buy | No | Once per mint |
| Price impact against real pool depth | No — amount ÷ price | Constant product |
| A flat round trip correctly shows a loss | No — shows break-even | Asserted in tests |
| Transactions that revert, and still cost fees | No | Optional simulator |
| Slippage tolerance actually rejects a fill | No | Checked after latency |
| Solana | Usually | Yes |
| BNB Chain, including transfer-tax tokens | Sometimes | Tax modelled |
| Robinhood Chain | No | Yes |
| Analytics, telemetry or an account | Usually yes | None at all |
Judge any tool with one test: buy and sell immediately, without the price moving. If it hands your money back, its fee model is not real — and every result it has ever shown you is inflated.
The five differences that matter
Network fee, priority fee, Jito tip, token-account rent, the terminal's platform fee, the pool's LP fee and BNB transfer tax all leave your paper balance. This is the difference between learning that trading is free and learning what frequency actually costs you.
Constant product against the pool's own reserves, so one SOL into a $3,000 pool hurts exactly as much as it would on chain — and you see the number before you commit, not after.
Solana through Jupiter and DexScreener with a direct pump.fun feed. BNB Chain with live gas and transfer taxes. Robinhood Chain through WETH-quoted Uniswap pools, where depth is reported directly so impact is exact rather than estimated.
No account, no email, no analytics, no telemetry, no remote config, no external scripts. Wallets, positions and full history live in your own browser's database. There is no server to send them to, because there is no server.
Ghostbag never replaces the chart you are reading. It appends a single container, remembers its position per site, collapses into a pill when you want the space back, and reads only explicitly labelled market values from the page.
Every value is an integer in base units — lamports, wei, raw token units — with floats only at the render boundary. A full round trip reconciles to the last raw unit, and the test suite fails if it does not.
Privacy
Most free tools are free because you are the product. Ghostbag has no account system, no analytics endpoint and no server that stores anything about you — so there is no database with your name in it, and never will be.
Paper wallets, open positions and your complete trade history are written to your own browser's local database. Settings sync only between your own signed-in Chrome profiles, through Chrome, not through us.
Only read-only market lookups — the token address on screen, and nothing attached to it. No identifier, no wallet address, no trading activity. It never touches window.solana or window.ethereum.