Why is a futures bot more complex than a spot bot?
Futures trading on perpetual contracts with leverage is a qualitatively different level of complexity. We have encountered projects where clients lost their deposit in an hour due to incorrect liquidation price calculation. In our practice, there was a case: a spot bot worked for a year without a single drawdown, but on futures it lost 40% of capital in a week. The reason was ignoring the funding rate and lack of margin ratio monitoring.
Therefore, developing a futures bot requires not just code but a risk management system. We use isolated margin, dynamic position size calculation, and emergency closure at critical margin levels. Our bots are built on Foundry (testing) and viem (client). One recent project on Polygon with Chainlink integration for accurate prices reduced the risk of manipulation during a flash crash. Savings on funding rate reach $500 per month with a $100k turnover. Compared to bots without oracles, the risk of liquidation due to slippage is reduced by 2 times, saving up to $2000 on each large position.
How does the funding rate affect the bot's strategy?
The funding rate is a periodic payment between long and short positions. If the rate is positive, longs pay shorts. On volatile pairs, funding can reach 0.15% per 8 hours, eating up to 0.45% of margin per day. Our bots include a filter: when the funding rate is extremely positive (>0.1% per 8 hours), longs are blocked; when extremely negative, shorts are blocked. This prevents funding losses and preserves capital. The bot processes signals 3 times faster than standard implementations thanks to asynchronous monitoring with asyncio.
Architecture and tech stack
Calculation of key parameters — futures trading bot development
from decimal import Decimal
class FuturesPositionCalculator:
def calculate_position_size(
self,
capital: Decimal,
risk_pct: Decimal,
entry_price: Decimal,
stop_loss_price: Decimal,
leverage: int,
) -> dict:
risk_amount = capital * risk_pct
price_diff_pct = abs(entry_price - stop_loss_price) / entry_price
position_size_usd = risk_amount / price_diff_pct
required_margin = position_size_usd / Decimal(str(leverage))
if required_margin > capital * Decimal('0.3'):
position_size_usd = capital * Decimal('0.3') * Decimal(str(leverage))
required_margin = capital * Decimal('0.3')
quantity = position_size_usd / entry_price
return {
'position_size_usd': position_size_usd,
'quantity': quantity,
'required_margin': required_margin,
'leverage_used': leverage,
}
def calculate_liquidation_price(
self,
entry_price: Decimal,
leverage: int,
side: str,
maintenance_margin_rate: Decimal = Decimal('0.005'),
) -> Decimal:
if side == 'LONG':
liq_price = entry_price * (1 - 1/Decimal(str(leverage)) + maintenance_margin_rate)
else:
liq_price = entry_price * (1 + 1/Decimal(str(leverage)) - maintenance_margin_rate)
return liq_price
Funding rate awareness in strategy
class FundingAwareStrategy:
EXTREME_FUNDING_THRESHOLD = 0.001
async def get_adjusted_signal(self, base_signal: Signal, symbol: str) -> Signal:
funding = await self.exchange.fetch_funding_rate(symbol)
current_rate = float(funding['fundingRate'])
if current_rate > self.EXTREME_FUNDING_THRESHOLD and base_signal == Signal.LONG:
return Signal.HOLD
if current_rate < -self.EXTREME_FUNDING_THRESHOLD and base_signal == Signal.SHORT:
return Signal.HOLD
return base_signal
What risks need to be considered?
| Risk | Description | Our protection |
|---|---|---|
| Liquidation due to high leverage | A 10% price move at 10x leverage liquidates the position | We use 3-5x for automated trading, isolated margin |
| Stop-hunting | Price triggers the stop then reverses | Sliding buffer of 0.5-1% |
| Funding drain | Persistent positive funding eats profits | Filter: block longs when rate >0.1% per 8h |
| Flash crash | Sudden 20% drop in seconds | reduceOnly + closePosition orders |
Comparison of risk management approaches
| Approach | Liquidation risk | Additional costs |
|---|---|---|
| No margin monitoring | High (up to 100% at 5x) | None |
| Static stop-loss | Medium (30-50% drawdown) | Missed profit |
| Dynamic calculation (ours) | Low (less than 10%) | Rebalancing fees |
Development process
- Analytics — study volatility, liquidity, historical liquidations for the chosen pair. Use data from CCXT to unify exchanges.
- Design — select margin mode (isolated), leverage, custom stops. Design architecture with asynchronous monitoring.
- Implementation — code in Python with asyncio, integration via CCXT, testing on Foundry.
- Testing — backtest on historical data accounting for fees and funding rate. Mandatory fuzzing with Echidna for contracts (if on-chain components exist).
- Deployment — on VPS with monitoring via Telegram bot. Configure alerts on critical margin ratio.
Monitoring details
- Check margin ratio every 30 seconds.
- Emergency closure when ratio drops below 1.5x maintenance.
- Notifications in Telegram for critical events.
Common mistakes in futures bot development
- Ignoring the funding rate: even a small rate of 0.05% per 8 hours at 5x leverage yields 0.25% daily losses. We've seen projects where funding drain consumed 60% of profits.
- Lack of margin ratio monitoring: the bot may miss approaching liquidation. Our monitoring checks every 30 seconds and closes positions if the ratio falls below 1.5x maintenance.
- Incorrect position sizing: using the entire deposit without considering slippage. We limit margin to 30% of capital.
Timeline and cost
Timeline: from 2 to 6 weeks depending on complexity. Cost is calculated individually based on the scope of logic, tests, and integrations.
What's included
- Architecture documentation
- Source code with comments
- Monitoring and alert setup
- Operation manual
- 2 weeks of post-launch support
- Training session for your team
We have 5+ years of experience in crypto development and 10+ implemented trading bots. We guarantee no reentrant vulnerabilities and adherence to best practices.
Contact us for a consultation — we'll help design and implement your futures bot from scratch or modernize an existing one. Order futures bot development that accounts for all risks and operates stable.