Strategy Report Card — Bollinger 20/2.5 Reversion (QQQ)

VERDICT: REJECT — no robust edge

Do not trade. The edge does not survive out-of-sample + costs.

A Bollinger Band mean-reversion strategy using a 20-period SMA with 2.5 standard-deviation bands. It goes long when the closing price crosses below the lower band and goes short when the closing price crosses above the upper band. It is best suited for range-bound or mean-reverting market regimes.

Family bollinger · direction long_short · params period=20, mult=2.5

Translation warnings

The Pine Script strategy has no explicit exit logic; positions are only reversed when the opposite signal fires. The built-in bollinger template matches this behaviour (flat inside bands), but verify that the engine handles the 'no exit until opposite signal' case correctly.
The script uses strategy.entry which means a new Long entry while already long (or Short while already short) is a no-op in TradingView, so there is no pyramiding — the engine should behave the same way.

Metric	Original ★	Improved +low_vol+vol_target
OOS Sharpe	-0.141	-0.408
Stress Sharpe (3x cost)	-0.417	-0.614
Ann return	-0.3%	-0.8%
Max drawdown	-4.3%	-5.7%
Ann turnover	6.2	4.7
PSR (P[Sharpe>0])	0.354	0.070
Bootstrap p(mean<=0)	0.705	0.940

Recommended: original — OOS Sharpe -0.141, PSR 0.354, bootstrap p 0.705.

Breadth — does the same edge hold elsewhere?

MIXED — works in some places; likely regime/asset specific

Across assets: 3/8 positive (3/8 under stress), median Sharpe -0.087

Asset	Sharpe	Stress
SPY	-0.03	-0.30
QQQ	-0.14	-0.42
IWM	-0.26	-0.50
EEM	0.80	0.58
EWZ	0.40	0.29
GLD	-0.96	-1.34
AAPL	-0.16	-0.33
MSFT	0.74	0.56

Timeframe	Sharpe	Stress
D	-0.14	-0.42
W	0.40	0.38
M	0.38	0.38

Want your strategy checked like this — out-of-sample, net of cost?

Request a free analysis of any strategy →

Out-of-sample, net of cost. PSR / bootstrap / stress-test robustness is what a TradingView backtest cannot show. Not investment advice.