Most traders running 10+ EAs believe they're diversified. After all, they have breakouts, mean-reversions, scalpers — different strategies on different pairs. The backtest correlations looked fine. But if you look at what these EAs actually do on any given day, a different picture emerges.

The illusion of diversification

Diversification in a trading portfolio is usually described in terms of strategy type or instrument. One breakout EA, one mean-reversion, one scalper — different approaches, different pairs. That logic works in theory, but it misses what matters most in practice: what direction are all these strategies going at the same time?

A breakout EA on XAUUSD buys when gold breaks above resistance. A trend EA on XAUUSD buys when price is in an uptrend. A volatility EA on XAUUSD buys when ATR expands. Three different strategies, three different signal methods — and in a strong gold uptrend, all three trigger simultaneously in the same direction. You haven't placed three trades. You've placed one bet, three times.

The concentration problem

When EAs share underlying assets or correlated pairs (EURUSD/GBPUSD, XAUUSD/XAGUSD, NAS100/US30), their positions cluster during trending markets. Trending markets are also when individual position sizes become most consequential.

How to actually measure concentration

The standard approach is to run a Pearson correlation on daily P&L across strategies. If two EAs have a correlation below 0.3, they're considered uncorrelated. This works as a rough filter, but it obscures something critical: correlation is a backward-looking average. It tells you how similar their past returns were, not whether they're pointing the same direction right now.

A more direct measure is directional overlap: on any given day, what fraction of simultaneously-open trades share the same symbol and direction? If EA-1 has 3 long XAUUSD trades and EA-5 has 2 long XAUUSD trades open at the same moment, your effective XAUUSD long exposure isn't 5 trades — it's one concentrated position expressed through five entries.

68%of multi-EA portfolios show >40% overlap on at least one symbol in trending conditions
2.3×average drawdown amplification when 3+ EAs open same-direction positions simultaneously
12 minaverage time window where overlap trades cluster during news-driven moves

A concrete example: seven EAs, one trade

Consider a portfolio with seven EAs running across EURUSD, GBPUSD, USDJPY, XAUUSD, and NAS100. On a day when the US dollar strengthens sharply — say, after a hot CPI print — here's what happens:

  • EA-1 (trend, EURUSD): goes short EURUSD on dollar strength
  • EA-2 (breakout, GBPUSD): goes short GBPUSD on resistance break
  • EA-3 (trend, USDJPY): goes long USDJPY on dollar strength
  • EA-4 (correlation, XAUUSD): goes short gold on dollar inverse relationship
  • EA-5 (macro, NAS100): goes short NAS100 on rate expectations
  • EA-6 (scalp, EURUSD): opens short EURUSD on micro momentum
  • EA-7 (swing, GBPUSD): goes short GBPUSD on daily structure

On paper, you have seven independent EAs on five different instruments. In practice, every single one of them is expressing the same macro trade: long USD. If the CPI data gets revised, or the dollar reverses, all seven positions move against you simultaneously. Your maximum drawdown isn't the worst-case loss of any individual EA — it's the sum of all of them.

What to do about it

There are three practical approaches, and they work best in combination.

1. Measure actual simultaneous exposure

Before you can fix concentration, you need to see it. Track your open trades in real time, grouped by underlying symbol. EURUSD and GBPUSD should be treated as partial proxies for each other; XAUUSD and XAGUSD are highly correlated. The question isn't 'how many EAs do I have?' — it's 'if USD drops 1% right now, what's my total P&L impact?'

2. Set exposure limits by correlation group

Define maximum simultaneous exposure per symbol or correlation group. For example: no more than 3% of account equity in net long/short XAUUSD exposure across all EAs combined at any moment. This forces portfolio-level risk management instead of per-EA sizing.

3. Monitor directional overlap, not just correlation

Check the percentage of your open positions that are directionally aligned. A healthy portfolio should rarely have more than 50-60% of its open positions pointing the same direction on the same underlying exposure. When this number spikes — especially into trending market conditions — that's your early warning signal.

The key insight

Diversification in an EA portfolio isn't about having different strategy types. It's about having different directional exposures at the moment risk events occur. Strategy type diversification is easy. True directional diversification is hard — and invisible without the right tools.

How AlgoLens surfaces this

The Overlap Analysis dashboard in AlgoLens maps your open and historical positions to their underlying exposure groups, shows you directional alignment in real time, and flags when multiple EAs are expressing the same trade. The goal isn't to tell you what to do — it's to make invisible concentration visible, so you can make an informed decision before the market makes it for you.

See this in your own portfolio

AlgoLens gives you every metric and visualization mentioned in this article — live, from your real trading data.

Start free trial →See all features