Ready to go further
Next: practise on real prices
Paper trading uses real prices and no real money, so it shows more of the real market than a play-money copy.
What is included
- What it is, in plain words
- The formula or mechanism, with exact arithmetic
- How it hits grid, martingale and DCA bots
- A check you can run before you switch a bot on
Who it is for
- People who run or plan to run a trading bot
- Anyone reading a backtest or a bot sales page
- Readers who want the numbers, not the pitch
- Not advice, and not a signal service
Key facts
- A testnet is a separate practice version of an exchange interface that uses play funds. Orders sent there do not touch real money, and the funds there have no value.
- It answers one question well: does my program connect, sign requests, place orders and handle the replies without crashing. It answers another question badly: what would real trading cost.
- Practice environments have their own keys and their own addresses, and they can be reset or changed by the venue. Read the current documentation of the venue for how to reach one.
- In a made-up order book, a buy of size 1 fills at 100.00 and a buy of size 18 averages 100.54, 0.539 percent higher (calculated by us on 31 Oct 2026, exact). A practice book may not look like a real one.
- This page explains how something works and does not recommend it. We have not tested it for this page. Check the current terms.
What a testnet is
A testnet (test network) is a copy of a trading interface set up so that developers can try their programs without risking money. The word comes from blockchains, but several exchanges also run practice versions of their trading interface. You usually make a separate key for it, you get play funds, and you point your bot at a different address. The code of the bot stays the same, which is the point.
Whether a given venue offers one, where it lives, which markets it copies and how often it is reset are all matters for that venue's current documentation. We have not checked for this page, so we do not state any of them.
What it is good for
- Connecting. Your program can reach the interface, sign a request and read the reply.
- Order flow. You can place, change and cancel orders, and see how your code reacts to a rejection, a partial fill or a cancel.
- Error paths. You can cut the network on purpose and see whether the bot recovers without sending a double order.
- Logging. You can check that every decision and fill reaches your log in a form you can read later.
- Rounding. You can check that quantities and prices match the steps of the market. See Binance API.
What it cannot show
Play funds trade against whatever else is in the practice book. That book may be thin, may be filled by other testers and test scripts, and may not move like the real market. So three things in particular are not learned there.
- Real slippage. The price you get depends on the size of the real book. See slippage.
- Real costs. Fees and spreads on practice trades may not match the live ones.
- How a strategy performs. A good week of practice results is not evidence of an edge. It is a small sample on a market that is not the real one.
Worked example: why fills differ
A market buy takes the cheapest sell offers first, then the next, until the order is full. Here is a made-up book of sell offers: size 1 at 100.00, size 2 at 100.10, size 5 at 100.30 and size 10 at 100.80. The table shows the average price paid for four order sizes, and how far that is above the best offer.
| Order size | Average fill price | Above best offer (price) | Above best offer (percent) | Same order on a thin book |
|---|---|---|---|---|
| 1 | 100.00 | 0.00 | 0.000% | 100.05 |
| 3 | 100.07 | 0.07 | 0.067% | 100.18 |
| 8 | 100.21 | 0.21 | 0.213% | 100.51 |
| 18 | 100.54 | 0.54 | 0.539% | cannot fill |
Calculated by us on 31 Oct 2026, exact for the made-up book. A bigger order pays more per unit because it eats deeper into the book. The last column repeats the orders on a made-up thin book with half the size at every level: the size 18 order cannot fill at all, and the size 8 order pays a higher average. A practice book that is thinner or deeper than the real one gives you the wrong lesson in either direction.
How a bot built on a testnet breaks later
- Fills assumed. The bot expects every order to fill at the price it chose. Live, orders fill in pieces, at other prices, or not at all.
- Size not scaled. A size that works on play funds is much larger than what the real book can take cheaply.
- Different rules. Minimum sizes, steps and available markets may differ between the practice interface and the real one.
- Old copies. The practice interface can lag behind the real one after a change, so a bot tested there may meet a surprise live.
- Keys mixed up. A practice key and a real key look alike. Keep them in separate settings, and make the bot print which environment it is using at start.
How to use a testnet properly
- Use it to test the plumbing: connect, sign, place, cancel, restart.
- Write down what you expect each order to do, then compare with what happened.
- Do not read the profit and loss of the practice run as a result.
- Move to paper trading on real prices, then forward testing, then a very small live size.
- Compare each live fill with what the bot expected and with the book you saw.
What this page does not tell you
It does not say whether a given exchange has a testnet, where it is, or how it behaves, because those are facts we did not read for this page. The order book is made up. It is a method for thinking about fills, and it is education, not advice.
Keep Going
Related Tools and Guides
Keep exploring
FAQ
Frequently Asked Questions
What is the Binance testnet?
It is a practice version of the exchange interface that uses play funds, so a bot can be tested without real money. Check the venue's current documentation for how to reach it.
Are testnet results a guide to live results?
No. The practice book can be thin, crowded with other tests and unlike the real market, so profit and loss there say very little about live trading.
Do I need a different key for a testnet?
Usually yes, practice environments use their own keys and addresses. Check the documentation, and keep practice and real keys in separate settings.
What should I test on a testnet?
The plumbing: connecting, signing, placing, changing and cancelling orders, handling rejections and partial fills, and restarting without sending double orders.
What comes after a testnet?
Paper trading on real prices, forward testing on new data, and only then a very small live size, comparing every fill with what the bot expected.
Are the numbers on this page real?
Example numbers are made up to show how the maths works, and the page says so. A figure only counts as exact when the page says it was calculated on a stated date. Nothing you enter in a calculator leaves your browser.
Can I use this to decide whether to run a bot?
It can show whether a bot's numbers hold together and how much an account can lose. It cannot say what the market will do, and it is not advice. Use it to ask better questions, then read the terms of your own broker or exchange.
Is this financial advice?
No. It explains what a practice environment is for, using made-up numbers. It does not recommend any venue.
Know the Risk Before You Run It
Free calculators and guides, added every week. Leave your details and we will email you when new ones go up.
Trading carries a risk of loss