There is exactly ONE free lunch in trading.
It’s not an indicator. It’s not a secret setup. It’s not a mentor.
It’s a formula from 1952, and it won a Nobel Prize.
Let me break down the maths of portfolio theory step by step, because once you understand it you’ll never trade one strategy again. 🧵
1/ Risk doesn’t add up the way you think
Say you have one strategy:
→ It makes 10% a year
→ Its volatility (how much it swings) is 15%
Now you add 4 more strategies with the same edge and split your risk equally across all 5.
Most people think: “5 strategies, same edge, so same result.”
Wrong.
Your return stays the same: 10%.
Your risk does NOT.
2/ The formula
Portfolio risk = σ × √( (1 + (N − 1) × ρ) ÷ N )
→ σ = the risk of one strategy
→ N = how many strategies you run
→ ρ = correlation (0 = completely independent, 1 = identical)
That’s it. That’s the whole secret.
3/ Plug in the numbers
5 strategies, completely uncorrelated (ρ = 0):
15% × √(1 ÷ 5) = 15% × 0.447 = 6.7%
Same 10% return. Risk drops from 15% to 6.7%. That’s 55% less risk for the exact same profit.
4/ Now look at the Sharpe ratio
Sharpe = return ÷ risk
→ 1 strategy: 10 ÷ 15 = 0.67
→ 5 uncorrelated strategies: 10 ÷ 6.7 = 1.49
You more than doubled your risk-adjusted return without finding a single better trade.
The multiplier is √N (if the strategies are uncorrelated):
→ 2 strategies: ×1.4
→ 5 strategies: ×2.2
→ 10 strategies: ×3.2
→ 20 strategies: ×4.5
5/ Then use the extra room
Here’s where it gets crazy.
You had 15% risk with one strategy and it felt fine. Now the portfolio only has 6.7%. So scale it back up to 15%.
10% × 2.24 = 22.4% a year. Same risk you were already taking.
You didn’t change a single entry. You just stopped putting all your eggs in one basket.
6/ But correlation is the killer
Real strategies are never perfectly independent. Here’s the same 5-strategy portfolio at different correlations:
→ ρ = 0: Sharpe ×2.2
→ ρ = 0.2: Sharpe ×1.7
→ ρ = 0.5: Sharpe ×1.3
→ ρ = 0.8: Sharpe ×1.1
And the hard ceiling, even with INFINITE strategies, is 1 ÷ √ρ:
→ ρ = 0.2: max ×2.2
→ ρ = 0.5: max ×1.4
→ ρ = 0.8: max ×1.1
Read that again.
If your 5 “different” strategies are all long-NASDAQ breakout systems with ρ = 0.8, you’re basically running ONE strategy five times. You could add 100 more and never get past ×1.1.
This is why “diversification” with 5 gold scalpers does nothing.
7/ The rule that blows people’s minds
Adding a WORSE strategy can make your portfolio BETTER.
The maths: a new strategy improves your portfolio if
its Sharpe > its correlation with your portfolio × your portfolio’s Sharpe
So if your portfolio’s Sharpe is 1.5 and a new strategy has a correlation of 0.1, it only needs a Sharpe above 0.15 to help, added in a small dose.
A boring, mediocre, uncorrelated strategy can be worth more than a brilliant one that moves with everything you already have.
Stop asking “is this strategy good?”
Start asking “is this strategy DIFFERENT?”
8/ Why this matters for prop firms
Prop firms don’t care about your return. They care about your drawdown.
10% max loss. 5% daily loss. That’s the whole game.
Drawdowns shrink roughly in line with volatility. So a strategy with a 12% historical drawdown can’t survive a 10% max loss on its own.
Split the same risk across 5 uncorrelated strategies and that worst drawdown shrinks to roughly 5–6%.
Same edge. Same profit. Now it fits inside the rules.
That’s not luck. That’s maths.
9/ How to actually use it
→ Different markets: indices, metals, FX, crypto
→ Different styles: breakout, mean reversion, intraday
→ Different timeframes: daily swing plus intraday
→ Check the correlation of daily returns before adding anything
→ Size each strategy so no single one can sink the account
10/ The bottom line
Harry Markowitz published this in 1952 and won a Nobel Prize for it. Hedge funds have built empires on it for 70 years.
Retail traders are still looking for the one perfect setup.
There is no perfect setup.
There’s only a portfolio.
@DTommydle I don’t know what you mean with atlas OHLC reviewing. I test with Dukascopy and Darwinex tick data in MT5 with realistic commissions and swaps. I don’t know your rules so I don’t know what I did differently.
How to pass an @FTMO_com prop account in 15 days with an automated portfolio:
Run multiple uncorrelated systems at higher evaluation risk.
Example:
• 5–10 systems
• Different markets
• Different logic
• Higher fixed risk per trade
The goal is to hit the target faster without relying on one strategy.
But higher risk cuts both ways:
You can pass faster.
You can also fail faster.
Once funded, reduce risk aggressively.
High risk for the challenge.
Low risk for longevity.