Core I - Systemic
Share
Translation to English:
PART 2 — Why does the geometric mean produce lower growth?
Because losses cause more damage than equivalent gains create benefit.
Example:
+20% → ×1.20
–20% → ×0.80
Total:
1.2 × 0.8 = 0.96 → –4%
Even though the arithmetic average = 0%, the geometric result = –4%.
This is the core of the Geometric Mean Trap:
When your returns go up and down, losses pull you down faster than gains can lift you up.
🧠 PART 3 — Why is it a “trap”?
Because the brain thinks linearly:
“If I win slightly more often than I lose → the account grows.”
But the account does not follow linear math.
It follows:
log(equity)
That’s why you can have:
-
a positive win rate
-
positive R/R
-
positive average gain
-
small losses
-
a stable strategy
…and still end up with long-term negative growth.
🧠 PART 4 — How is real growth (geometric growth) calculated?
[
E[\ln M] = p \ln(u) + (1-p)\ln(d)
]
Where:
-
( p ) = win rate
-
( u ) = multiplier on a win (1 + risk %)
-
( d ) = multiplier on a loss (1 – risk %)
If:
-
( E[\ln M] > 0 \Rightarrow ) “Profit”
-
( E[\ln M] < 0 \Rightarrow ) “Loss”
🧠 PART 5 — How did the Geometric Mean Trap catch you, specifically?
Here are your numbers:
-
Win rate: 50.32%
-
Gain per win: +2%
-
Loss per loss: –2%
-
Trades: 1101
Your multipliers:
-
( u = 1.02 )
-
( d = 0.98 )
Plugged in:
[
E[\ln M] = 0.5032 \ln(1.02) + 0.4968 \ln(0.98) = -7.202 \times 10^{-5}
]
This is:
-
negative
-
therefore: long-term declining account
Your average growth per trade is:
[
GM = e^{-7.202 \times 10^{-5}} = 0.999928
]
That is:
-
0.0072% loss per trade
-
Accumulated:
( 0.999928^{1101} \approx 0.9135 \rightarrow –8.65% )
Your Excel showed –$861 → –8.61%.
Perfect match.
→ The Geometric Mean Trap explains exactly why you are losing.
🧠 PART 6 — Why is a 50.32% win rate not enough?
The break-even win rate for your strategy is:
[
p_{BE} = \frac{-\ln(d)}{\ln(u) - \ln(d)} \approx 50.50%
]
You are at:
-
50.32%
-
Missing 0.18%
→ the account declines.
Win rate > 50% is not enough in multiplicative growth.
🧠 PART 7 — How does this affect your trades?
Here are the effects you can see directly:
1️⃣ The equity curve slopes slightly downward
Because you lack the microscopic win-rate or RR edge needed to make geometric growth positive.
2️⃣ Drawdowns look large
Losses multiply downward harder than gains pull upward.
3️⃣ You “feel” the strategy should be winning
Because:
-
the number of winning trades is higher
-
average gains look larger
-
RR > 1
-
everything looks good arithmetically
But capital sees things geometrically.
4️⃣ You lose slowly, but consistently
This is not chaos — it is mathematically predictable:
[
GM < 1 \Rightarrow \text{long-term capital erosion}
]
🧠 PART 8 — How do you avoid the Geometric Mean Trap?
You must increase:
-
win rate above 50.50%, or
-
RR higher than 1.16, or
-
risk-per-trade stability, or
-
reduce loss size relative to win size
The easiest paths:
-
Increase average win (raise RR), or
-
Remove your largest losses (cut losers faster)
Both push geometric growth above zero.
🧠 PART 9 — Why is this SO important for traders?
Because:
👉 90% of all strategies that look profitable arithmetically are losing geometrically.
You’ve just seen why.
Only strategies where:
[
E[\ln M] > 0
]
grow.
All others die slowly.
🧠 PART 10 — The ultra-short summary
Geometric Mean Trap = capital decay that occurs because losses multiply harder than gains, even when win rate and RR seem “good enough.”
In your case:
-
Win rate: 50.32%
-
RR = 1.16
-
Risk = 2%
→ Geometric growth < 0
→ Account declines slowly
→ Your Excel was right
2. Arithmetic Mean (AM)
3. Kelly Criterion
4. Expected Value (EV)
[
EV = (\text{win rate} \times \text{avg win}) - (\text{loss rate} \times \text{avg loss})
]
But:
-
This is arithmetic (per trade).
-
Used to evaluate trade quality, not account growth.
A strategy can have positive EV but negative GM (and vice versa).
5. Volatility Drag
Same problem as the Geometric Mean Trap, explained as:
Volatility reduces geometric growth.
The larger the percentage swings (risk per trade), the lower the GM.
Example:
+10%, –10%, +10%, –10%
→ Net loss, even though the average is 0%.
8. Risk of Ruin (RoR)
The probability of going bankrupt (or hitting a defined drawdown limit).
This is simply proper accounting analysis and risk management.
9. Positive Expectancy
EV > 0.
But expected value says nothing about GM.
Many amateurs believe positive expectancy = profit.
Not true — GM determines profit.
10. Central Limit Trap
Many believe that many trades smooth out variance.
But if GM < 1 → more trades = faster account destruction.
11. Margin of Safety (MOS)
How much “buffer” you have between your win rate and the break-even win rate.
If your strategy requires 50.5% to be profitable and you have 50.32% → MOS is negative.
That’s why you lose.
12. Regret Minimization / Utility Function
Risk aversion makes negative GM objectively bad, even if EV is positive.
Used in position-sizing models.