DeployQuant20-Day Momentum + Trailing Stop
Buy after a +10% four-week burst and let a 10% trailing stop manage the exit.
This page tests one rule set on 59 ETFs from 2021-01-04 to 2026-10-02, a window of 5.7 years. The rule buys when an ETF's 20-day return is above +10%, puts 98% of the sleeve into it, and then a trailing stop 10% below the highest close manages the exit. There is no profit target. The starting cash is $10,000 and the headline run has no fees or slippage.
The median result across all 59 funds is a CAGR of 0.0% with a median max drawdown of 20.1%. The strategy finished ahead of buy-and-hold on 21 of the 59 funds and had a shallower drawdown than buy-and-hold on 56. Only 27 of the 59 funds ended with a positive CAGR. Those figures describe a rule that sits out most of the time: the median exposure is 21.3%, and the median fund saw 6 round trips in the whole window.
A median of zero hides two different groups. Bond and currency funds produced zero trades because the entry condition never came true, so their result is a flat line at $10,000. The leveraged and sector funds produced the large positive and negative numbers. The trailing stop is a backtest-only feature for now, so the results here describe simulated history and not something that can be switched on in a live account.
The rankings table above lists every fund. Each row links to its own page with the trade list, the drawdown dates and the cost runs.
The rules
- WHEN the market opens · IF not invested AND the 20-day return > +10% · THEN buy with 98% of the sleeve
- WHILE invested · a managed trailing stop follows 10% below the position's high-water mark
Momentum entry, trailing-stop exit. A +10% gain over 20 sessions marks a burst of momentum. The template buys the strength and exits on a 10% trailing stop that ratchets up beneath the highest close and never moves down. There is no profit target, so winners run until the trail is hit.
Good for: high-momentum assets where trends extend, such as semiconductor and leveraged tech ETFs.
Watch out: buying strength means buying high; when a burst immediately reverses, the trail exits about 10% below the entry.
What the entry and the stop do together
The entry is a single test. At the market open, if the strategy is flat and the fund has gained more than 10% over the previous 20 sessions, it buys. The trailing stop then follows 10% below the position's high-water mark. It ratchets up beneath the highest close and never moves down. When price closes through that line the position is sold, and the strategy waits for the next burst.
These two parts pull in opposite directions. The entry waits for a fast move, so the first purchase is made after the fund has already risen about a tenth. The exit allows a fall of a tenth from the peak. On a fund that trends in a clean line, the stop stays far from price and the position rides. On a fund that rises in jumps and gives half of them back, the stop sits close to price after every pop, and a normal pullback ends the trade.
The result is a rule with a narrow band of conditions where it works. It needs bursts that keep going for weeks, and it needs the gaps between bursts to be shallower than 10%. The strategy page describes the fit as high-momentum assets where trends extend, such as semiconductor and leveraged tech ETFs. The data supports part of that and contradicts part of it. QLD ran at a 22.5% CAGR against 23.6% for buy-and-hold with a 20.1% drawdown, which is the closest the rule came to matching a fund on a return basis. SOXL and TQQQ went the other way, as the next section shows.
The caveat in the template is also visible in the data. When a burst reverses at once, the trail exits about 10% below the entry. The funds with the lowest win rates, such as TQQQ at 34% and TECL at 42%, are the ones where this happened over and over.
Build it from blocks (or type it in English), backtest it on 5.7 years of minute data in seconds, tweak any parameter, then paper trade it on live data. No card, no broker needed to start.
Results on every ETF
| ETF | CAGR | buy & hold | max DD | Sharpe | trades | win rate |
|---|---|---|---|---|---|---|
| QLD | 22.5% | 23.6% | −20.1% | 0.97 | 20 | 50% (+1 open) |
| SOXL | 19.1% | 33.3% | −67.5% | 0.59 | 138 | 37% (+1 open) |
| SPUU | 17.3% | 22.5% | −17.6% | 0.91 | 12 | 75% (+1 open) |
| SOXX | 16.7% | 31.1% | −30.2% | 0.78 | 18 | 61% (+1 open) |
| SSO | 15.4% | 21.9% | −24.1% | 0.83 | 13 | 62% (+1 open) |
| ROM | 13.3% | 30.2% | −36.2% | 0.58 | 33 | 42% (+1 open) |
| XLK | 11.6% | 22.1% | −26.1% | 0.80 | 9 | 56% |
| TECL | 10.2% | 38.2% | −48.5% | 0.45 | 64 | 42% (+1 open) |
| CLSE | 10.1% | 19.5% | −4.7% | 1.46 | 0 | – (+1 open) |
| VV | 9.3% | 14.3% | −13.6% | 0.91 | 3 | 67% (+1 open) |
| VOO | 9.2% | 14.4% | −13.2% | 0.91 | 3 | 67% (+1 open) |
| SPY | 9.1% | 14.6% | −13.1% | 0.91 | 3 | 67% (+1 open) |
| QQQM | 7.9% | 16.7% | −27.0% | 0.62 | 8 | 63% |
| QQQ | 7.8% | 16.7% | −26.5% | 0.61 | 8 | 63% |
| IOO | 7.3% | 16.7% | −19.6% | 0.72 | 4 | 50% (+1 open) |
| IAU | 7.1% | 13.7% | −18.5% | 0.58 | 6 | 50% |
| XLF | 7.0% | 12.6% | −17.6% | 0.60 | 5 | 60% |
| VOOV | 6.4% | 12.6% | −11.8% | 0.80 | 2 | 100% (+1 open) |
| VOOG | 6.1% | 15.7% | −28.7% | 0.56 | 6 | 50% (+1 open) |
| FAS | 3.9% | 18.3% | −45.4% | 0.28 | 45 | 42% |
| XLY | 3.8% | 6.3% | −18.4% | 0.33 | 9 | 56% |
| EEM | 2.4% | 6.6% | −24.4% | 0.27 | 5 | 60% |
| VOX | 2.2% | 9.1% | −27.7% | 0.23 | 8 | 63% |
| KMLM | 2.1% | 7.1% | −10.0% | 0.36 | 2 | 50% |
| QQQE | 1.8% | 9.8% | −21.2% | 0.21 | 6 | 50% (+1 open) |
| IWM | 1.4% | 7.5% | −24.1% | 0.17 | 7 | 29% (+1 open) |
| TQQQ | 0.6% | 25.4% | −40.9% | 0.18 | 53 | 34% (+1 open) |
| AGG | 0.0% | −0.8% | −0.0% | 0.00 | 0 | – |
| BND | 0.0% | −0.8% | −0.0% | 0.00 | 0 | – |
| IEF | 0.0% | −2.3% | −0.0% | 0.00 | 0 | – |
| IEI | 0.0% | −0.4% | −0.0% | 0.00 | 0 | – |
| IGIB | 0.0% | 0.2% | −0.0% | 0.00 | 0 | – |
| SGOV | 0.0% | 3.2% | −0.0% | 0.00 | 0 | – |
| FXE | 0.0% | −0.9% | −0.0% | 0.00 | 0 | – |
| UDN | 0.0% | −1.1% | −0.0% | 0.00 | 0 | – |
| USDU | 0.0% | 5.2% | −0.0% | 0.00 | 0 | – |
| QAI | 0.0% | 3.9% | −0.0% | 0.00 | 0 | – |
| ALTY | 0.0% | 7.6% | −0.0% | 0.00 | 0 | – |
| VTV | −0.3% | 13.4% | −9.5% | -0.03 | 1 | 0% |
| TLT | −1.2% | −8.2% | −10.3% | -0.14 | 2 | 0% |
| XLP | −1.9% | 5.8% | −14.5% | -0.36 | 2 | 0% |
| SPDN | −2.2% | −9.7% | −17.9% | -0.22 | 4 | 25% |
| SH | −2.2% | −10.0% | −17.9% | -0.23 | 4 | 25% |
| RINF | −2.3% | 6.6% | −17.1% | -0.32 | 2 | 0% |
| TBF | −2.3% | 12.2% | −16.4% | -0.29 | 4 | 25% |
| CTA | −2.8% | 8.9% | −27.4% | -0.20 | 4 | 25% (+1 open) |
| QID | −3.1% | −29.5% | −45.7% | 0.00 | 33 | 39% |
| UST | −4.3% | −8.5% | −24.6% | -0.74 | 3 | 0% |
| PSQ | −5.7% | −13.8% | −33.1% | -0.47 | 8 | 38% |
| REW | −8.1% | −36.1% | −62.8% | -0.13 | 44 | 34% |
| SDS | −9.9% | −21.6% | −48.6% | -0.48 | 23 | 30% |
| TMF | −11.1% | −31.2% | −55.7% | -0.46 | 27 | 26% |
| SOXS | −13.6% | −48.3% | −77.5% | -0.01 | 129 | 31% |
| EEV | −14.8% | −16.1% | −61.9% | -0.61 | 25 | 24% |
| VIXM | −15.9% | −16.1% | −65.0% | -1.03 | 17 | 0% |
| SQQQ | −16.0% | −42.3% | −67.5% | -0.28 | 72 | 29% |
| VXZ | −17.3% | −15.1% | −67.9% | -1.13 | 18 | 6% |
| TECS | −18.3% | −46.7% | −72.4% | -0.32 | 83 | 29% |
| UVXY | −36.2% | −48.7% | −93.3% | -0.60 | 125 | 28% |
Where the rule worked and where it failed
The top of the ranking is led by leveraged funds that track broad indexes. QLD returned a 22.5% CAGR, SPUU returned 17.3% with a 17.6% drawdown, and SSO returned 15.4% with a 24.1% drawdown. All three had 20 trades or fewer, with win rates of 50%, 75% and 62%. Their exposure was between 50.3% and 59.3%. These funds rose in steady stretches without the violent daily reversals of the single-sector leveraged funds, so the 10% stop was rarely hit by noise.
The semiconductor funds show the cost of a stop that is too tight for the underlying volatility. SOXL traded 138 times in the window with a 37% win rate, and its CAGR of 19.1% sat below the 33.3% for buy-and-hold. The max drawdown was 67.5%. A fund that moves 10% in a day or two will trigger the stop on ordinary swings, and the rule then re-enters on the next burst. Each cycle buys after a rise and sells after a fall. SOXX, the unlevered version, had 18 trades and a 16.7% CAGR with a 30.2% drawdown, against 31.1% for buy-and-hold. The same sector, with less amplitude, gave the rule fewer false exits.
The worst pairing is TQQQ. It traded 53 times, won 34% of them, and finished at a CAGR of 0.6% against 25.4% for buy-and-hold, with a 40.9% max drawdown. TECL had 64 trades and a CAGR of 10.2% against 38.2%. In both cases buy-and-hold was far ahead. The rule reads a fast move in a three-times fund as a signal, and the trail gives it only a 10% cushion on an instrument that can fall that far in a week.
The broad unleveraged index funds sit in the middle. VV, VOO and SPY each had 3 round trips and about 9% CAGR against roughly 14% for buy-and-hold, with drawdowns between 13.1% and 13.6%. QQQ managed 7.8% against 16.7% with a 26.5% drawdown. The rule captured part of each trend and was out of the market for the rest.
One row looks odd. CLSE shows a 10.1% CAGR with a 4.7% drawdown and 0 round trips, at an exposure of 30.9%. A fund can show a return with no completed round trip when the strategy buys once and is still holding at the end of the window. The 0 in the trades column counts closed trades.
The inverse and volatility funds show the other side. REW lost at a CAGR of 8.1% negative against 36.1% negative for buy-and-hold, and TMF lost 11.1% a year against 31.2%. The strategy beat buy-and-hold on 10 of the 11 inverse funds. That statistic reflects how badly those funds decay when held through the window, because the rule was out of them for most of the time. It does not mean the rule made money on them: the median CAGR for the group was negative.
Results by fund type
| Fund type | ETFs | Median CAGR | Median buy & hold | Median max DD | Beat holding |
|---|---|---|---|---|---|
| Broad index ETFs | 12 | 7.3% | 14.4% | −21.2% | 0 of 12 |
| Sector ETFs | 6 | 7.0% | 12.6% | −26.1% | 0 of 6 |
| Leveraged ETFs | 10 | 13.3% | 23.6% | −40.9% | 2 of 10 |
| Inverse ETFs | 11 | −8.1% | −21.6% | −48.6% | 10 of 11 |
| Bond ETFs | 7 | 0.0% | −0.8% | −0.0% | 5 of 7 |
| Commodity ETFs | 1 | 7.1% | 13.7% | −18.5% | 0 of 1 |
| Currency ETFs | 3 | 0.0% | −0.9% | −0.0% | 2 of 3 |
| Volatility products | 3 | −17.3% | −16.1% | −67.9% | 2 of 3 |
| Alternative-strategy ETFs | 6 | 0.0% | 7.6% | −10.0% | 0 of 6 |
Results by fund type
Grouping the funds shows where the median of zero comes from. Bond funds have a median CAGR of 0.0% and a median max drawdown of 0.0%, and the rule beat buy-and-hold on 5 of the 7. The comparison looks good because most bond funds had a negative buy-and-hold CAGR over the window, and the strategy never entered any of them except TLT, which had 2 round trips and a CAGR of 1.2% negative against 8.2% negative for buy-and-hold. The currency funds behave the same way: 3 funds, a median CAGR of 0.0%, and a beat of buy-and-hold on 2 of 3.
Leveraged ETFs are the largest and most uneven group. There are 10 of them, with a median CAGR of 13.3% against 23.6% for buy-and-hold, and a median drawdown of 40.9%. Only 2 of the 10 beat buy-and-hold. This is the category the template names as a good fit, and the data says the fit is partial. The index-tracking funds QLD, SPUU and SSO work, and the single-sector and three-times funds do not.
Sector funds had a median CAGR of 7.0% against 12.6% for buy-and-hold and a median drawdown of 26.1%, with 0 of 6 beating buy-and-hold. Broad index funds had a median of 7.35% against 14.42% and a median drawdown of 21.2%, with 0 of 12 ahead. Commodity ETFs hold one fund, IAU, at 7.1% against 13.7% and an 18.5% drawdown. Alternative-strategy funds have a median CAGR of 0.0% against 7.6% for buy-and-hold, with a median drawdown of 10.0%, and 0 of 6 ahead.
The inverse group has a median CAGR of 8.12% negative against 21.57% negative, and a median drawdown of 48.61%. These are also the funds with the most trades. SOXS made 129 round trips with a CAGR of 13.60% negative against 48.27% negative held and a 77.45% max drawdown. TECS made 83 round trips and returned 18.33% negative against 46.67% negative held, and SQQQ made 72 and returned 15.97% negative against 42.33% negative held. Volatility products are the worst, with a median CAGR of 17.28% negative and a median drawdown of 67.94%, and 2 of the 3 beat holding. UVXY made 125 round trips with a 28% win rate, a CAGR of 36.17% negative and a 93.34% max drawdown. The stop did not protect against the structural decay of those products, because the rule kept buying each new burst.
The pattern across categories is consistent. The groups where the rule looks good against buy-and-hold are the ones where buy-and-hold lost money: bond, currency, inverse and volatility funds. In every group where the underlying fund made money, the strategy trailed it. That is the cost of a rule that is invested about a fifth of the time at the median.
Year by year, median across all ETFs
| Year | momentum breakout | Buy & hold | ETFs with a gain |
|---|---|---|---|
| 2021 | 0.0% | 4.1% | 16 of 59 |
| 2022 | −5.7% | −12.7% | 8 of 59 |
| 2023 | 0.0% | 8.9% | 28 of 59 |
| 2024 | 0.0% | 9.7% | 19 of 59 |
| 2025 | 1.0% | 11.1% | 30 of 59 |
| 2026 | 0.0% | 3.7% | 22 of 59 |
Year by year across the 59 funds
The yearly medians show the same story in miniature. In 2021 the median strategy return was 0.0% against 4.1% for buy-and-hold, with 16 funds positive. In 2022 the strategy median was 5.7% negative against 12.7% negative for buy-and-hold, and 8 funds were positive. That is the one year where the rule's median came out clearly ahead of the median fund held through the year, and the strategy's low exposure kept it out of part of the decline.
In 2023 the median strategy return was 0.0% while buy-and-hold had 8.9% and 28 funds were positive. The market recovered in steady stages, and a 20-day return above 10% is a rare event for a broad index fund in a year like that. The rule was waiting for a burst while the funds rose quietly. In 2024 the pattern repeated: 0.0% for the strategy, 9.7% for buy-and-hold, and 19 funds positive.
In 2025 the strategy median was 1.0% against 11.1% for buy-and-hold, with 30 funds positive. That was the year with the most positive funds, and the rule still captured little of the move at the median. The 2026 row covers only the part of the year up to 2026-10-02, and it shows 0.0% against 3.7% with 22 funds positive.
Two things follow from the table. First, the strategy was flat in the median fund in four of the six rows, so the median is dominated by funds that did not trade that year. Second, the one year where it lost money at the median, 2022, is also the year where buy-and-hold lost more. The rule's drawdown figures look good against buy-and-hold on 56 of 59 funds because it spends most of its time out of the market, and the cost shows up as lower return in each rising year.
The window is one stretch of 5.7 years and it includes a single bear market. Any statement about how the rule behaves across cycles would need more history than this test holds.
Changing the parameters
| Version | Median CAGR | Median max DD | Median round trips |
|---|---|---|---|
| Published rules | 0.0% | −20.1% | 6 |
| 5% trailing stop | 0.3% | −12.5% | 8 |
| 15% trailing stop | 0.0% | −28.6% | 3 |
| 20% return entry | 0.0% | −0.0% | 0 |
What changing the stop and the entry did
Three variants were run on all 59 funds and the page reports the median of each.
The 5% trailing stop gave a median CAGR of 0.25%, a median max drawdown of 12.5% and a median of 8 round trips. Tightening the stop raised the number of trades and lowered the depth of the median loss, because positions were closed sooner. The CAGR stayed near zero, so the shallower drawdown came from being out of the market more often, and not from a better edge. Compare that with the base rule at 20.1% median drawdown and 6 median round trips.
The 15% trailing stop gave a median CAGR of 0.0%, a median drawdown of 28.6% and only 3 median round trips. A wider stop kept positions alive through pullbacks and so made fewer trades, and the drawdown rose because each position was allowed to give back more before the exit. The median return did not improve. On the median fund the 10% setting did not look like an obvious midpoint of a smooth curve either: the three stop widths give three different trade counts and three different drawdowns with CAGR near zero at each.
The 20% return entry produced a median CAGR of 0.0%, a median drawdown of 0.0% and a median of 0 round trips. At that threshold most funds never reached a 20% gain in 20 sessions, so the strategy never entered. The only funds that would trigger it are the leveraged and volatility products. For a rule meant to run on the 59-fund universe, an entry that fires on fewer than half of them is a very selective filter.
These are median results, and individual funds can differ. The variants do show that the stop width trades drawdown for trade count, and that the entry threshold decides whether the rule does anything at all. Neither parameter turned the median into a positive CAGR. The test varied one setting at a time and did not search combinations, so it says nothing about whether some pairing of the two would do better.
The base rule is one of 12 templates tested on the same universe. The strongest medians among the others are the weekly 7% target at 6.61% and the monthly cycle at 5.46%, then the RSI(2) snapback at 4.74%. This strategy's median sits at 0.0%. The trend + trailing stop template, which adds a trend filter in front of a stop of the same kind, has a median of 2.61%, and the RSI mean reversion template has 2.98%. The 200-day regime filter is at 1.84%.
Frequently asked questions
What is the momentum breakout strategy?
Buy after a +10% four-week burst and let a 10% trailing stop manage the exit. Momentum entry, trailing-stop exit. A +10% gain over 20 sessions marks a burst of momentum. The template buys the strength and exits on a 10% trailing stop that ratchets up beneath the highest close and never moves down. There is no profit target, so winners run until the trail is hit.
Does momentum breakout beat buy-and-hold?
Across 59 ETFs backtested 2021-01-04 to 2026-10-02, it beat same-ETF buy-and-hold on 21 of 59 (36%). Median CAGR was 0.0% with a median max drawdown of 20.1%. Per-ETF results vary widely; the table lists every one.
How does the trailing stop work in the backtest?
The stop sits 10% below the highest price since entry and only ratchets up. When a bar trades through it, the position sells at the stop, or at the open if the price gaps below it. Trailing-stop exits can be backtested now; deploying them live is not available yet.
What does the 20-day momentum plus trailing stop strategy do?
It buys with 98% of the sleeve when a fund's 20-day return is above +10% and the strategy is flat. A trailing stop then follows 10% below the highest close and never moves down. When price falls through the stop, the position is sold and the rule waits for the next burst.
How did it do across the 59 ETFs?
The median CAGR was 0.0% and the median max drawdown was 20.1%, over 2021-01-04 to 2026-10-02. It beat buy-and-hold on 21 of 59 funds and ended with a positive CAGR on 27. Bond and currency funds produced no trades, which is why the median is zero.
Which ETFs suited this rule best?
QLD had the highest CAGR at 22.5% against 23.6% for buy-and-hold, with a 20.1% drawdown. SPUU returned 17.3% and SSO 15.4%. These are leveraged index funds that trended in long stretches. Fast single-sector funds such as SOXL and TECL traded far more often and finished well below buy-and-hold.
Why did it trade SOXL 138 times?
SOXL moves fast enough that a 10% fall from the high is a normal swing. The stop was hit often and the 20-day burst test then triggered a new entry. The win rate was 37% and the max drawdown was 67.5%.
What happens if the trailing stop is 5% or 15%?
A 5% stop gave a median CAGR of 0.25%, a median drawdown of 12.5% and 8 median round trips. A 15% stop gave a median CAGR of 0.0%, a median drawdown of 28.6% and 3 median round trips. The stop width changed the trade count and drawdown and left the median return near zero.
Can this strategy be run live?
Not yet. The trailing-stop templates are backtest-only for now, so the figures on this page come from simulated history. The headline run has no fees or slippage, and the page's cost runs add 5 and 10 basis points.
Does a 20% entry threshold work better?
At a 20% threshold the median fund had 0 round trips, so the strategy never entered on most funds. The median CAGR and median drawdown were both 0.0%.
Compare with other strategies
Backtests are hypothetical, computed by DeployQuant's engine on minute-resolution consolidated US market data (2021-01-04 to 2026-10-02, $10,000 starting capital, no margin, no fees or slippage in the headline run; buy-and-hold puts 98% of the account in at the first open, as the templates do) and do not guarantee future results. Nothing on this page is investment advice. Live trading involves risk of loss.