The Capacity Problem: Why 10-Lots Win and 100-Lots Lose
Your EA returned 47% on your $50k account over six months. You're ready to scale. You fund a $500k account, apply the same strategy, and immediately it dies.
The strategy isn't broken. Your position size is.
Capacity is the maximum position size you can trade without moving the market against you. Every trader has a capacity limit. Most traders never find it until they hit it in live trading—and by then it costs them money.
Here's the thing: backtests don't account for market impact. Your backtest assumes you can buy 100 contracts of ES at the ask without moving the price. In reality, a 100-contract order moves the market. You get filled 2-3 ticks worse. That's 20-30 points of slippage on a single entry.
Scale that across 100 trades a year, and market impact alone wipes out your edge.
Market Impact: The Hidden Tax on Large Orders
Market impact is the price movement your order causes just by existing.
A 10-contract order into ES? Barely moves the needle. Market makers absorb it. A 100-contract order? You're now the biggest player in the order flow. The market moves against you before your order is fully filled.
The math is brutal:
- Small order (10 contracts): You hit the ask at 5,100.50. Filled instantly. Slippage: 1-2 ticks.
- Medium order (50 contracts): You absorb the bid-ask spread and partial fills at 5,100.75, 5,101.00. Slippage: 5-8 ticks.
- Large order (100+ contracts): You exhaust the bid-ask spread and move into the next level of orders. Slippage: 20-50 ticks. On ES, that's $1,000-$2,500 per contract.
That's not a bug in your strategy. That's the cost of capacity.
According to CFA research, large institutional orders incur 3-5x more slippage than small retail orders. Retail traders don't account for it at all.
Liquidity Constraints by Asset Class
Your capacity limit depends entirely on what you trade.
Equities (ES, large-cap stocks): High liquidity. You can scale to 100+ contracts before hitting meaningful slippage. Your capacity limit is in the millions.
Forex (EURUSD, GBPUSD): Extremely high liquidity. Your capacity is in the tens of millions. But micro pairs (EURGBP, NZDJPY)? Much lower. 10-lot orders move these pairs noticeably.
Crypto (Bitcoin, Ethereum): Lower liquidity than equities. A $1M order on Binance moves the price. A $100k order on a smaller exchange causes 2-3% slippage. Your capacity depends heavily on the exchange and the pair.
Bonds, commodities, small-cap stocks: Even lower liquidity. A $100k position in a small-cap stock can move it 5-10%. Your capacity might be $10k-$50k total.
Most traders backtest on one contract size, then apply it to a different asset class with a fraction of the capacity. The result: real-world slippage destroys the backtest.
Position Sizing for Real Accounts
Here's the formula that actually works:
- Find your true capacity. This is the max position size you can move without 5%+ slippage. For ES, it's 100+ contracts. For a small-cap stock, it's 2,000-5,000 shares.
- Size your positions at 25-50% of capacity. If you can safely trade 100 contracts, your working position is 25-50 contracts. This gives you room to scale without hitting liquidity walls.
- Use Kelly Criterion adjusted for slippage. Your position size should be f* = (bp - q) / b, where b is your odds, p is your win rate, and q is your loss rate. Reduce this by 30-50% to account for slippage you didn't forecast.
- Run Monte Carlo sims on real market data. Backtesting on historical data is one thing. Running 10,000 random walks through your data with realistic slippage and spread assumptions shows you where you actually break.
The result: a position size that survives real market conditions, not just backtest conditions.
Testing Capacity Before Going Live
Don't discover your capacity limit in live trading.
Before you scale, run a capacity stress test. This means:
- Paper trade at 2x your target size for 2-4 weeks and track actual slippage vs. backtest assumptions.
- Track how many contracts you can move without expanding the bid-ask spread on your broker.
- Test your exit strategy at scale. A 50-contract position that exits cleanly becomes a 500-contract position that bleeds on exit.
- Run Monte Carlo simulation on your backtest data with 30-50% larger slippage assumptions to stress-test your model.
This takes time. But it's infinitely cheaper than discovering your capacity limit after you've scaled to live trading and blown through slippage.
The Cost of Ignoring Capacity Planning
A trader makes $50k on a $50k account. He's thrilled. He scales to $500k and applies the same strategy.
What happens:
- His average entry slippage goes from 1 tick to 20 ticks.
- His average exit slippage goes from 2 ticks to 15 ticks.
- His win rate stays the same. His risk/reward deteriorates by 50%.
- His strategy goes from +15% annually to -8% annually.
- He loses $40k in the first four months.
The strategy didn't break. The capacity math did.
This is why most traders who scale blow up. It's not because their trading logic failed. It's because they didn't build their position sizing around real market conditions.
How Institutions Handle Capacity at Scale
Institutional traders don't discover capacity. They architect for it from day one.
When an institution deploys $100M in a single strategy, they don't multiply the position size by 2,000. They:
- Run capacity analysis on every instrument they plan to trade.
- Reduce position sizes to stay below the liquidity threshold for each asset.
- Diversify across multiple asset classes to avoid capacity walls in any single market.
- Use smart order routing to split orders across multiple venues and reduce market impact.
- Build custom execution algorithms that execute gradually, not in one hit.
You don't have a $100M prop desk. But you can think like one.
The Real Solution: Automated Capacity Planning
Here's the problem: manual capacity planning is a full-time job. You'd need to backtest every position size, stress-test slippage, and monitor live trading slippage against forecasts. Most traders can't do this consistently.
This is exactly why custom EAs built with capacity planning in mind exist. A properly built trading bot accounts for slippage in advance, adjusts position size based on real liquidity, and runs full Monte Carlo simulations before going live.
When we build a custom Expert Advisor, we include full backtest reports with slippage models baked in. Every entry, every exit—accounted for. No surprises when you go live.
Our AI trading bots go further. They monitor real-time slippage and adjust position sizing dynamically. If market conditions change and liquidity drops, the bot scales down. If liquidity improves, it scales back up. All automatic.
The difference: strategies that scale from $50k to $500k without breaking. Starting from $100 for a simple EA. Demo delivered in 45 minutes.
Key Takeaways
- Backtests lie about slippage. Your $50k account backtest assumes frictionless execution. The $500k live account doesn't.
- Capacity is real. Every asset class has a max position size. Scale past it and your edge becomes negative.
- Position sizing fixes it. Size at 25-50% of true capacity and use Kelly Criterion adjusted for real slippage.
- Test before scaling. Paper trade at 2x your target size and monitor slippage for 2-4 weeks. Run Monte Carlo simulations.
- Automate the math. Custom EAs with proper capacity modeling scale without breaking. Tell us what you trade and we'll show you the EA.