Is the Supertrend indicator profitable?

Not against buying and holding. On SPY since 1993, all 24 settings tested had a lower Sharpe ratio than holding, and the default 10 × 3 earned 6.6% a year against 10.7%.

What are the best Supertrend settings?

None held up. The best setting on 1993–2009 (ATR 21, multiplier 4) earned 6.0% a year from 2010, against 13.9% for holding SPY.

Is Supertrend better than a 200-day moving average?

No. On SPY since 1993 the 200-day rule earned 8.9% a year with a 23% worst fall, against 6.6% and 27% for Supertrend 10 × 3.

Market factPublished Nov 2026 · SPY and FRED to Nov 6, 2026; company prices to Sep 25, 2026

We tested 24 Supertrend settings. None beat holding the S&P 500

On SPY since 1993, every Supertrend setting had a lower Sharpe ratio than buying and holding. The setting that looked best before 2010 earned 6.0% a year afterwards, against 13.9% for holding.

Supertrend settings that beat holding SPY: 0 of 24
Return, 10 × 3: 6.6%/yr
Return, held: 10.7%/yr
Best pre-2010 setting, after: 6.0%/yr
Held, after 2010: 13.9%/yr

What we found

Supertrend draws a band of average true ranges (ATR) around the price and flips between uptrend and downtrend when the price crosses it. Long SPY in an uptrend and in T-bills in a downtrend, it had a lower Sharpe ratio than holding at all 24 settings tested: ATR periods of 7, 10, 14 and 21 days and multipliers from 1.5 to 4. They ran from −0.08 to 0.48, against 0.50 for holding; the popular 10 × 3 ranked 10th, at 0.40.

The default 10 × 3 earned 6.6% a year against 8.9% for the 200-day rule and 10.7% for holding, with a worst fall of 27%. It trailed even before trading costs.

Picking the best setting with hindsight doesn’t rescue it. The best one on 1993–2009 (ATR 21 × 4) then earned 6.0% a year from 2010 against 13.9% for holding, and 9.3% against 18.2% on QQQ. On single stocks it beat holding’s Sharpe ratio for only 16% to 27% of them, the most volatile included.

A grid of Sharpe ratios for 24 Supertrend settings, all below holding SPY’s 0.50; beside it, returns from 2010 to 2026: holding 13.9% a year, the 200-day rule 8.7%, Supertrend 10 × 3 7.5%, and the best setting before 2010 6.0%.

The full results

Sharpe ratio of each setting (holding SPY: 0.50)

Darker is better; none reaches holding.

Sharpe ratio of each setting, SPY 1993–2026

ATR period	1.5×	2×	2.5×	3×	3.5×	4×
7 days	−0.08	0.36	0.48	0.47	0.34	0.42
10 days	−0.04	0.35	0.40	0.40	0.41	0.35
14 days	0.02	0.19	0.41	0.38	0.41	0.39
21 days	0.02	0.15	0.32	0.37	0.40	0.44

Holding SPY: 0.50. Cells beating it: 0 of 24. None differs significantly from holding at a Bonferroni-corrected 5% level.

Supertrend 10 × 3 against the 200-day rule, SPY

Measure	Supertrend 10 × 3	200-day rule	Always invested
Return a year	6.6%	8.9%	10.7%
Worst fall	−27%	−23%	−55%
Sharpe ratio	0.41	0.56	0.50
Time in the market	62%	77%	100%
Round trips a year	3.9	3.2	0
Trips won	51%	44%	—
Average win / loss	+6.0% / −2.8%	+9.2% / −1.8%	—
Cost per switch that erases the Sharpe edge	none (behind at 0 bp)	16 bp	—

Supertrend minus the 200-day rule, Sharpe: −0.15 (−0.36 to +0.06).

Sharpe ratio before and after 2010

1993–2009, where the best setting was chosen
2010–2026

Out of sample: the best setting on 1993–2009 (21 × 4), then 2010–2026

Strategy	Sharpe 1993–2009	Sharpe 2010–2026	Return a year 2010–2026
SPY: best setting	0.43	0.46	6.0%
SPY: 10 × 3	0.21	0.60	7.5%
SPY: 200-day rule	0.49	0.63	8.7%
QQQ: best setting	0.43	0.62	9.3%
QQQ: 10 × 3	0.21	0.64	9.0%
QQQ: 200-day rule	0.17	0.69	12.1%
SPY: held	0.28	0.76	13.9%
QQQ: held	−0.09	0.84	18.2%

QQQ starts in 1999, so its first period is 1999–2009.

Single stocks: share where the rule beat holding’s Sharpe ratio, by volatility

Supertrend 10 × 3
200-day rule

Single stocks by volatility: share where the rule beat holding’s Sharpe ratio

Volatility third	Median volatility	Supertrend 10 × 3	200-day rule
Low	18%	16%	16%
Mid	24%	23%	21%
High	35%	27%	31%

The 500 largest US stocks each January 2011–2021, held 5 years; volatility over the year before. Survivors only.

Why it trails

Supertrend reacts faster than a 200-day average, so it trades more (3.9 round trips a year on SPY against 3.2) and spends more time in cash (38% against 23%). Stocks rise most of the time, so every extra day out costs return, and faster signals are fooled more often by short dips.

Narrow bands are worst: with a 1.5× multiplier it switched 21 times a year and earned under 2% a year. Wider bands trade less and do better, but none caught up with holding.

The trap of the best setting

Test enough settings and one will look good by chance. The best of the 24 on 1993–2009 had a Sharpe ratio of 0.43 against 0.29 for holding, which looks like an edge. From 2010 it had 0.46 against 0.76. Choosing a setting on past data and assuming it will keep working is how most indicator backtests mislead.

How it was measured

SPY with dividends from 1993, with high and low prices adjusted for dividends. Supertrend on the high–low midpoint with Wilder’s ATR.
Traded at the next day’s close, 0.05% a switch, cash at the 3-month T-bill rate. Sharpe ratios from daily returns over T-bills.
With 24 settings tested, a setting would need to clear a Bonferroni-corrected 5% level to count. None comes close.

Caveats

The S&P 500 since 1962 tells the same story on prices alone (3.3% to 5.2% a year against 7.5% held). Its high and low prices before 1962 aren’t real, so the test can’t go further back.
The single stocks are companies still listed today.

Questions

Is the Supertrend indicator profitable?: Not against buying and holding. On SPY since 1993, all 24 settings tested had a lower Sharpe ratio than holding, and the default 10 × 3 earned 6.6% a year against 10.7%.
What are the best Supertrend settings?: None held up. The best setting on 1993–2009 (ATR 21, multiplier 4) earned 6.0% a year from 2010, against 13.9% for holding SPY.
Is Supertrend better than a 200-day moving average?: No. On SPY since 1993 the 200-day rule earned 8.9% a year with a 23% worst fall, against 6.6% and 27% for Supertrend 10 × 3.

More research

All research →

Historical analysis. Past performance does not guarantee future results. Not investment advice. Published Nov 2026. Data: SPY and FRED to Nov 6, 2026; company prices to Sep 25, 2026.