Ready to go further
Weigh the small wins against the rare loss
Enter your win size, loss size and how often each happens to see the average per trade.
What is included
- How the bot works, step by step
- The arithmetic of the ladder or grid, with tables
- How it fails and how to test it honestly
- The tools that check the numbers
Who it is for
- People thinking of using this kind of bot
- People already running one and wanting a stress test
- Readers comparing bot types before buying
- Not advice, and not a recommendation to use one
Key facts
- The honest answer to whether grid bots are profitable is that it depends on the path price takes, the costs paid and the settings used. A single answer for all grids does not exist.
- On a made-up range-bound path the example grid completes 16 grids and realises +220.00. On a made-up trending path, with the same grid and start, it realises +100.00 but is left holding 50 units with an unrealised result of -910.00 (calculated by us on 27 Oct 2026, exact, fees ignored).
- In a trend a grid behaves like a seller of volatility: its gains are capped by the spacing and its losses grow as price moves against it.
- Evidence that would count: tests on many assets and periods, all costs included, a period the settings were not tuned on, and a forward test. A single screenshot of profit proves nothing.
- Published results are filtered by survivorship: the accounts and bots that blew up are seldom shown.
Why the question has no single answer
People ask whether grid bots are profitable the way they would ask whether a car is fast. A grid is a rule for placing orders. Whether it makes or loses money depends on three things that the rule does not control: the path price takes, the fees and other costs paid on every fill, and the settings chosen. Change any one and the answer changes. This page shows how, with arithmetic, and does not claim that grids do or do not make money.
One grid, two made-up paths
Use the same made-up grid as the grid trading page: range 90 to 110, ten grids two apart, 5 units per grid, starting at 100 with 25 units bought at the start price to cover the sell orders above. Total money needed is 4,850. Fees are ignored here so the arithmetic is easy to follow, and the next section adds them.
Path one is range-bound: it swings between 92 and 108 and comes back to 100. Path two is a trend: it rises to 108 first, then falls steadily to 80 and stays there.
| Made-up path | Grids completed | Realised profit | Units held at end | Average cost of held units | Unrealised result | Total | Holding the first 25 units instead |
|---|---|---|---|---|---|---|---|
| Range-bound: swings 92 to 108, ends at 100 | 16 | +220.00 | 25 | 102.40 | -60.00 | +160.00 | 0.00 |
| Trending: up to 108, then down to 80 | 4 | +100.00 | 50 | 98.20 | -910.00 | -810.00 | -500.00 |
On the range path the bot completed 16 grids and realised +220.00. It also still holds 25 units bought at an average of 102.40, which carry an unrealised result of -60.00 at the end, so the total is +160.00. On the trend path it completed 4 grids early and realised +100.00, then bought all the way down. It holds 50 units at an average of 98.20 with price at 80, an unrealised result of -910.00, and a total of -810.00. Calculated by us on 27 Oct 2026, exact. The last column shows what simply holding the first 25 units would have given: 0.00 on the range path and -500.00 on the trend path. So the grid did better than holding on the range path and worse on the trend path. That is only what these made-up paths do, not a rule.
The grid is short volatility in a trend
Look at the shape of the pay-off. Each grid pays a small, fixed amount when price swings across it. In a strong move down, each additional line crossed adds a unit at a lower price, and the loss on the whole pile grows with every step. The gains are capped by the spacing and the losses are not capped by anything but the bottom of the price. Traders call this being short volatility: you are paid a little when price stays calm and you pay a lot when it moves in one direction. It is the same shape as martingale trading, with equal order sizes in place of growing ones.
Note this is a description of the pay-off, not a prediction. If a market spends most of its time swinging in a range, the pay-off suits it. If it trends, the pay-off hurts. Which one it will do next is exactly what nobody knows in advance.
Costs move the answer too
The table ignored fees. Every completed grid pays a buy fee and a sell fee, and funding if the bot trades futures. On the range path the bot completed 16 grids. At a made-up 0.2 percent round trip on roughly 100 per unit and 5 units, the fee per grid is about 1.00, so 16 grids cost about 16.00 of the 220.00 realised. That is an approximation. Narrower grids pay a bigger share, and a grid that is too narrow loses on each cycle, as grid bot fees and costs shows.
What evidence would count
- Many assets and many periods. One coin in one month is one path. A fair test includes calm periods, trends up and trends down, and crashes.
- All costs. Fees on every fill, spread, funding and slippage when the bot starts or stops.
- The whole position, not just the realised profit. A report that shows only completed grids hides what the bot is still holding. Ask for equity including unrealised results and for the worst drawdown.
- A period the settings were not tuned on. Otherwise the test shows how well the settings fit the past.
- A forward test on live prices before real money. See forward testing.
Why a screenshot of profit proves nothing
A screenshot shows one bot, one period and usually one number. It does not show the unrealised loss, the capital needed, the fees, or how many similar bots were started and stopped. A grid screenshot taken in a sideways month looks the same as one taken before a trend arrives. Published results are also filtered: bots that did badly get switched off and are seldom posted. That is survivorship bias, and it makes the average shown look better than the average earned.
Putting a number on it
If you have tested a setup over many paths, take the average profit of a completed cycle, the average loss when price leaves the range, and how often each happened. The expected value calculator combines them into an average per trade. A small average gain with a large rare loss is the pattern to look for. Tested results of our own will appear here when they are ready, with data, dates and costs. None are published yet, so no figure on this page is a result from a bot.
Keep Going
Related Tools and Guides
Keep exploring
FAQ
Frequently Asked Questions
Are grid bots profitable?
It depends on the path of price, the costs and the settings. A grid tends to do better when price swings inside its range and worse when it trends out of it. This page shows the arithmetic and does not claim either result as fact.
Why does a grid look profitable and still lose?
Realised profit counts completed grids only. The units the bot still holds can carry an unrealised loss that is larger than all the realised profit, especially after a trend down.
What does short volatility mean for a grid?
The bot collects small fixed gains while price swings and takes growing losses when price moves one way. The pay-off is capped on the upside and open-ended on the downside until the range ends.
How many tests are enough to trust a grid setup?
No number settles it. Use many assets and periods, include all costs and a period the settings never saw, then run it forward on live prices without real money.
Why do screenshots of grid profits mislead?
They show one bot in one period and usually leave out unrealised losses, capital needed, fees and all the bots that were switched off. That is survivorship bias.
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 is general education about how a type of bot behaves. The prices here are made up and it does not take your situation into account.
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