Risk Management is Your Only Guarantee
No trading strategy wins 100% of the time. The only guarantee you have in trading is that losses will occur. Risk management doesn't prevent losing trades — it ensures that losing trades can never threaten your ability to continue trading.
Professional algorithmic traders follow a principle: protect capital first, grow capital second. This guide walks you through the complete framework.
The Four Pillars of Bot Risk Management
- Position Sizing — how much capital to risk per trade
- Stop-Loss Placement — where to exit losing trades
- Drawdown Limits — circuit breakers that pause the bot
- Portfolio Diversification — spreading risk across assets and strategies
1. Position Sizing
Position sizing is the most impactful risk lever you have. The mathematical foundation is the fixed fractional method:
The 1% Rule
Risk no more than 1% of your total portfolio on any single trade. If your portfolio is $10,000, your maximum risk per trade is $100.
Calculating Position Size
Position Size = (Portfolio × Risk%) ÷ (Entry Price − Stop Price)
Example: Portfolio $10,000, risk 1%, entry at $40,000 BTC, stop at $38,500:
- Dollar risk = $100
- Price distance = $1,500
- Position size = $100 ÷ $1,500 = 0.067 BTC
This formula ensures that if the stop is hit, you lose exactly $100 regardless of how far the stop is placed.
Scaling Risk by Confidence
More experienced traders vary risk by signal quality: 0.5% for lower-confidence signals (single indicator), 1% for standard setups, up to 2% for highest-confidence setups (multiple timeframe confluence). Never exceed 2% per trade.
2. Stop-Loss Placement
A stop loss is non-negotiable. Every trade must have a pre-defined exit point. The three most effective stop placement methods for bots are:
Structure-Based Stops
Place the stop below the most recent swing low (for longs) or above the most recent swing high (for shorts). This respects market structure and gives the trade room to breathe.
ATR-Based Stops
Use the Average True Range (ATR) as a volatility-adjusted stop distance. A common setting is 2×ATR below entry. During high-volatility periods your stop widens automatically; during quiet markets it tightens. This prevents both premature stop-outs and excessive losses.
Percentage Stops
Simplest approach: set a fixed percentage below entry (e.g., 3% for swing trading, 5% for position trading). Less precise but easy to implement and backtest.
What NOT to do
- Never move a stop loss further away once hit
- Never use a mental stop ("I'll close manually if it drops 5%") — bots must use hard stops
- Never set stops at round numbers ($40,000, $50,000) — they are magnets for liquidity grabs
3. Drawdown Limits
Drawdown limits are circuit breakers that halt your bot when losses exceed predefined thresholds. Every production bot should have two levels:
Daily Drawdown Limit
If the bot loses more than X% of its equity in a single day, it stops and waits until the next day. A typical value is 3–5%. This prevents a bad day from becoming a catastrophic week during unusual market conditions or API failures.
Maximum Drawdown Limit
If the bot's equity falls more than X% from its all-time high, it shuts down completely and requires manual review. A typical value is 15–20%.
A 20% drawdown requires a 25% return to recover. A 50% drawdown requires a 100% return. Protecting against large drawdowns is more mathematically powerful than increasing win rate.
4. Portfolio Diversification
Running a single bot on a single pair concentrates all risk in one place. Professionals spread risk across multiple dimensions:
Asset Diversification
Run strategies on 3–6 uncorrelated pairs. BTC/USDT and ETH/USDT are heavily correlated (r ≈ 0.85). Better to diversify across market-cap tiers and sectors: BTC + a mid-cap + an altcoin.
Strategy Diversification
Combine strategies that perform differently in different market conditions. A trend-following strategy excels in trending markets; a mean-reversion strategy excels in ranging markets. Together they smooth out the equity curve.
Timeframe Diversification
Running both a swing strategy (4H timeframe) and a shorter-term strategy (1H) on the same pair provides signal diversity that reduces drawdown during choppy trend transitions.
The Kelly Criterion: Advanced Position Sizing
For optimised position sizing once you have sufficient backtest data, use the Kelly Criterion:
Kelly % = Win Rate − ((1 − Win Rate) ÷ Win/Loss Ratio)
Example: 55% win rate, 1.5:1 average win/loss ratio:
Kelly % = 0.55 − (0.45 ÷ 1.5) = 0.55 − 0.30 = 0.25 (25%)
In practice, use half-Kelly (12.5% here) to account for estimation error. Full-Kelly is theoretically optimal but practically leads to extreme volatility.
Risk Checklist Before Going Live
- ✅ Maximum risk per trade defined (≤2%)
- ✅ Stop loss configured on every entry signal
- ✅ Daily drawdown limit set in bot configuration
- ✅ Maximum drawdown limit set with auto-pause
- ✅ Maximum concurrent open positions defined
- ✅ Portfolio exposure per sector capped at 30%
- ✅ Position sizing formula uses dynamic calculation (not fixed USDT amount)