How much can a crypto trading bot make? There is no typical daily or monthly figure. What a bot makes depends on the market, the strategy, the position size, the fees and the risk you take, and the result can be negative: the same settings can gain in one month and lose in the next.
What you can calculate is your own net result. Below, one hypothetical trade goes from gross profit to net profit after exchange fees and Freya's performance fee, then through a month with the $40 fee cap. The numbers are arithmetic, not a forecast. Whether bots are profitable at all is answered in are crypto trading bots profitable.
Key Takeaways
- There is no typical amount a trading bot makes per day or per month; it depends on the market, strategy, position size, fees and risk, and can be negative.
- Net profit is gross profit minus exchange fees, platform fees and other costs, across winning trades, losing trades and open positions.
- Freya's performance fee is 20% of the gross profit of each closed profitable trade, before exchange fees, capped at $40 a month.
- Hypothetical example: 1.5000 USDT of gross profit, minus 0.2015 USDT of exchange fees and a 0.3000 USDT Freya fee, leaves 0.9985 USDT.
- Losing trades pay no Freya fee but do not reduce the fee on winning trades, so a month can end negative after fees were paid.
How much do trading bots make? It depends on five factors
A bot carries out your entry and exit rules; it does not make them profitable. These five factors decide the result:
| Factor | What it changes | What to check |
|---|---|---|
| Market conditions | Whether price moves as the strategy needs | Behavior in rising, falling and sideways periods |
| Strategy and settings | How often it trades and how long it holds | Entry conditions, take profit, safety orders, stop-loss |
| Position size | How large each gain or loss is in USDT | The full amount a deal can use, reserve included |
| Fees and execution | How much of each gross profit you keep | Exchange fees, slippage, futures funding, platform fees |
| Risk limits | How bad a losing deal can get | Stop-loss distance, leverage, bots on the same coin |
Market conditions matter most and are the part you control least: a long DCA bot needs a recovery to close in profit, and in a long decline it can hold a growing position at a loss. Also ask what a percentage return is divided by: the first order, the whole budget or the account. How take profit, safety orders and stop-loss work together in a DCA bot is shown in DCA bot settings explained.
How to calculate a trading bot's net profit
For one closed trade:
Net profit = gross profit - exchange fees - platform fees - other costs
- Gross profit: the sale value minus the cost of the position, before any fees.
- Exchange fees: charged on every fill, from the first entry to the exit.
- Platform fees: on Freya, 20% of the gross profit of each closed profitable trade, paid from your Freya platform balance in USDT. Your trading capital stays in your exchange account.
- Other costs: slippage and, on futures, funding.
Hypothetical example: one 100 USDT DCA trade
A spot DCA bot has bought a position for 100 USDT in total and sells all of it at a take profit 1.5% above the average entry price: one completed trade, not a month. Assume an exchange fee of 0.1% per fill, paid in USDT, and no slippage.
| Step | Calculation | USDT |
|---|---|---|
| Cost of all entries | assumed | 100.0000 |
| Value at exit | 100 × 1.015 | 101.5000 |
| Gross profit | 101.5000 - 100.0000 | 1.5000 |
| Exchange fees | 0.1000 on entries + 0.1015 on the exit | 0.2015 |
| Result after exchange fees | 1.5000 - 0.2015 | 1.2985 |
| Freya performance fee | 20% × 1.5000 | 0.3000 |
| Net profit | 1.5000 - 0.2015 - 0.3000 | 0.9985 |
- The Freya fee is based on gross profit: 20% of 1.5000 USDT, not of the 1.2985 USDT left after exchange fees and not of the 100 USDT position.
- The costs sit in two places: your exchange balance grows by 1.2985 USDT, and the 0.3000 USDT fee comes from your platform balance.
- Costs took about a third of the gross profit (0.5015 of 1.5000 USDT). A trade with a tiny price gain still pays the Freya fee, even if exchange fees make its net result negative.
How to count a whole month and the $40 fee cap
A month's result is the sum of all closed trades, winners and losers, after every cost, plus the change in value of open positions. Over a month, Freya's fee is:
- charged on each closed profitable trade separately, when it closes;
- not charged on losing trades, which do not reduce it: wins and losses are not netted;
- capped at $40 a month for your whole account, not per bot, on every plan, including bots bought on the strategy marketplace. At 20%, the cap is reached at 200 USDT of gross profit; later profitable trades that month pay no fee.
A Plus ($15) or Pro ($30) subscription renews every 30 days whatever your bots do, so count it too. The Free plan has no subscription but the same performance fee. What a crypto trading bot costs per month adds these costs up for a profitable and a losing month.
Continue the example: in the same month, the bot closes three such trades and one losing one, where the same 100 USDT position closes 10% below its average entry price, for 90 USDT.
| Trades in the month (USDT) | Gross result | Exchange fees | Freya fee | Net result |
|---|---|---|---|---|
| 3 profitable trades as above | +4.5000 | 0.6045 | 0.9000 | +2.9955 |
| 1 losing trade | -10.0000 | 0.1900 | 0 | -10.1900 |
| Month total | -5.5000 | 0.7945 | 0.9000 | -7.1945 |
The month ends 7.1945 USDT down, yet its winning trades paid 0.9000 USDT of Freya fees, well under the $40 cap: a negative month does not mean no fees. The full rules are in Freya Finance fees explained, and plan prices are on the pricing page.
Why past results and backtests don't guarantee income
A backtest runs your settings on a chosen stretch of history. Freya's backtests model exchange fees and slippage, but live trading meets different prices and liquidity, and the market may not repeat the test period; why backtest and live results differ explains the gaps. Any net figure from a test must also count the performance fee and any subscription.
Overfitting is the bigger trap: pick the best of many settings and you may have found a lucky fit to past prices. Testing other periods helps, and the Pro plan's Monte Carlo and walk-forward analysis are built for this check, but no test turns a past result into a promise.
In live statistics, check the closed deals, the open positions and their unrealized result, and the largest drawdown. Never multiply one trade's profit by the days in a month: a bot can wait days for an entry, hold a position for weeks or close the next trade at a loss. A take profit is an exit setting, not a monthly return.
How to compare bot profit with simply holding the coin
A useful test for any bot is whether it beats buying the coin at the start of the period and holding it to the end. Use the same coin, dates, starting capital and costs for both; comparing a backtest with buy and hold shows how to read this benchmark.
Count the bot's whole budget, including the safety-order reserve, and value open positions at the end price. Closed trades alone can hide a loss: 5 USDT of closed profit with an open position 8 USDT down is a result of 5 - 8 = -3 USDT so far. Compare drawdowns, not just profit.
How to limit potential losses
- Set the budget before the bot: an amount you can afford to lose, split between the first entry, the safety-order reserve and costs. How much capital you need for bot trading walks through the sizing.
- Size each deal so one loss is survivable. In the example, one losing trade cost more than ten winning ones.
- Use a stop-loss unless you have a reason not to. Freya watches the price and closes with a market order at the stop-loss level, so a fast market can fill worse than the level.
- Keep leverage low, or avoid it, while you learn. On futures, a position can be liquidated.
- Watch what your bots share. Bots on coins that move together can lose in the same drop.
How losses happen on spot and on futures is covered in realistic expectations: can you lose money on Freya?
Bot trading can lose part or all of the capital you allocate, and leverage makes losses larger and faster. No backtest, past result or take profit setting tells you what a bot will earn. The numbers here are hypothetical arithmetic, not a forecast. Only trade with money you can afford to lose.
Frequently Asked Questions
How much can a trading bot make per day?
There is no meaningful daily figure. On many days a bot closes nothing, and a day of closed profits can hide an open position that is down by more. Daily numbers also invite compounding claims, which assume that every day wins and every profit is reinvested. Read your own backtest or live statistics over many trades instead: net result after fees, open positions, worst drawdown and capital tied up.
How much can a trading bot make per month?
There is no typical monthly figure either; the same settings can gain one month and lose the next. Add up every closed trade after all fees and any subscription, plus the change in open positions, and compare several months and market phases in your own statistics or backtests. Check the drawdown and the capital used, not just the total.
Do trading bots make money?
They can, and they can also lose it: a bot carries out your rules, but it cannot make a losing strategy profitable. Whether bots are profitable in general is answered in are crypto trading bots profitable.
Does the take profit show the return on my whole deposit?
No. It is the exit target of one position, measured from its average entry price. The return on your deposit also depends on unused reserve, other trades, open positions and every cost.
Does Freya charge a fee if the bot ends the month at a loss?
Not on the losing trades. But each profitable trade that month was charged 20% of its own gross profit when it closed, and losses do not reduce those fees, so a month can end negative after fees were paid.
If a backtest is profitable, will the live bot make money too?
Not necessarily. A backtest covers one stretch of history, while live trading meets a different market. Settings tuned to the past often do worse on new data, so test several periods, start small and compare live results with the backtest before adding capital.



