I Tested a Simple CCI Strategy. The Results Held Up

Finance-specific summary (CCI mean-reversion backtest)

Asset/market tested

• SPY (S&P 500 ETF) using over 33 years of historical data.

Indicator: Commodity Channel Index (CCI)

• CCI is an oscillator that measures how far price deviates from its recent average.
• Very low CCI (especially large negative values) implies unusually weak price relative to its recent history.
• Thesis: mean reversion (i.e., not “CCI magic”).

---

Strategy methodology / step-by-step rules (explicit framework)

Parameters

• Compute CCI on a 17-day lookback (and nearby values were also tested).

Entry rule

• If CCI closes below -180, then buy SPY at the closing price.

Exit rule (primary)

• Hold until CCI rises above -75.
• Exit at that closing price.

Constraints

• No stop loss
• No profit target
• Only one position open at a time
• No hidden filters

---

Backtest performance metrics & key numbers

• Total trades: 150 completed transactions
• Win rate: ~80.7% profitable (about 81%)
• Average profit per trade: +1.04%
• Profit vs loss magnitude: gross profits exceeded gross losses by 3.34×
• Maximum drawdown: ~13.2%
• Return metric reported: “return coefficient”
  • Baseline (17-day): 3.34

Important accounting note

Results do not include:

• commissions
• slippage
• taxes
• dividends

---

Robustness checks (settings, exits, and market periods)

1) Overfitting / lookback robustness

Keeping entry/exit thresholds the same (-180 and -75) while varying CCI lookback:

• CCI 16: return coefficient 2.99
• CCI 17: return coefficient 3.34
• CCI 18: return coefficient 2.61

Conclusion: Similar results across nearby lookbacks (not a single “lucky” parameter).

2) Exit robustness

• Exiting when CCI recovered above -75 was best, but nearby exit levels also worked.
• A 5-day exit (mentioned as an alternative) also remained profitable:
  • odds/score cited: 2.28 (as reported)

3) Regime robustness

Split into three long periods:

• Period 1 return coefficient 2.88
• Period 2 return coefficient 2.64
• Period 3 return coefficient 6.36

Interpretation: Profitability appears across different market eras (not confined to one decade).

---

Explicit cautions / interpretation

• A high win rate alone isn’t sufficient; the author highlights sensitivity testing across:
  • parameter changes,
  • exit changes,
  • multiple historical periods.
• The author cautions against assuming strong recent performance will continue.
• Framing: “historical trend, not a forecast.”

---

Disclosures / disclaimers

• No explicit “not financial advice” disclaimer was included in the provided subtitles excerpt.

---

Tickers/assets/instruments mentioned

• SPY (S&P 500 ETF)
• CCI (indicator; not a traded instrument)

---

Presenters / sources

• No presenter name or external source was provided in the subtitles excerpt.