SMA-200 Trend + 15% Trailing Stop on TQQQ

ProShares UltraPro QQQ: 3x daily leveraged Nasdaq-100, with large trends and deep drawdowns. Backtest 2021-01-04 to 2026-10-02, $10,000 starting capital, computed by the same engine that runs live DeployQuant strategies.

The SMA-200 trend rule with a 15% trailing stop returned 25.03% a year on TQQQ from 2021-01-04 to 2026-10-02. Buy-and-hold returned 25.35%, so the rule gave up 0.32 points of CAGR. The ending values were $36,064 and $36,585. The difference is the drawdown: 43.94% for the rule and 80.77% for holding, with a Sharpe of 0.75 against 0.69. This is the best result of this template on any of the 59 funds and the 3rd best of the 12 templates on TQQQ.

The rule buys at the open when yesterday's close was above the 200-day average and it holds no position. It then sets a stop that trails 15% below the highest price since entry, and sells when the stop is hit. There is no other exit. On TQQQ it made 29 round trips and had 1 more position open at the end, won 13 of the 29 for a 45% win rate, and was invested 60.7% of sessions. The average win was 25.49% and the average loss 10.15%, with a profit factor of 1.44.

Half the story is a few long winners. The best trades returned 62.1%, 62.02% and 50.2%, and the median trade lost 5.95%. The other half is a long list of small losses taken when the 200-day signal flipped on and the stop closed the trade within days.

This template is backtest-only for now and cannot be deployed to a live account. The numbers are one window with no fees or slippage in the headline run, $10,000 starting cash, no margin and daily decisions. Trade prices are adjusted for splits and dividends.

Year by year

Holding TQQQ returned 88.9% in 2021, −78.1% in 2022, 187.6% in 2023, 57.1% in 2024, 33.9% in 2025 and 54.1% in 2026 so far. The rule returned 18.6%, −20.1%, 82%, 59.7%, 14.5% and 14.4%. It beat holding in 2022 by 58 points and in 2024 by 2.6, and trailed in the other four years. The largest gaps were 2023, at 105.6 points behind, and 2021 at 70.3 behind.

The first trade began on 2021-10-19, because the 200-day average needs history before it can give a signal. That trade returned 10.73% by 2021-12-01, and two short ones followed. The year closed at 18.6% against 88.9% for holding, so the rule missed the early 2021 gains by construction.

The 2022 figure is the case for the stop. The last trade closed on 2022-01-18 at a loss of 10.18%, after entering on 2022-01-06. The rule then made no trade until 2023-02-03. For all of 2022 after January the account sat in cash, and it ended the year at −20.1% where holding lost 78.1%. The January loss accounts for the whole of the rule's 2022 return. The monthly table shows January 2022 at −20.07% and every other month of 2022 at 0%.

In 2023 the rule returned 82% in a year the fund returned 187.6%. It was late: the first trade of the year, from 2023-02-03 to 2023-02-21, lost 8.89%, and the next, from 2023-03-17, gained 62.02% to 2023-08-08. It also took five closed trades that year with 2 wins. The two largest winners of the whole run came in 2023 and early 2024: 62.02% from 2023-03-17 and 62.1% from 2023-10-23 to 2024-04-17.

In 2024 the rule matched the fund at 59.7% against 57.1%. In 2025 it returned 14.5% against 33.9%, and in 2026 so far 14.4% against 54.1%. The recent years had more small losses: 6 closed trades in 2025 with 2 wins and 7 in 2026 with 3 wins.

Month by month

The best month was November 2023 at 33.05%, close to the fund's own strong months. The worst was July 2026 at −20.47%. Holding's best month was April 2026 at 51.68% and worst April 2022 at −36.68%. The rule's April 2022 was 0%, because the account was flat. The rule's April 2026 was 22.62%, against 51.68% for the fund.

The losing months show the cost of the rule. They are January 2022 at −20.07%, September 2023 at −15.66%, April 2024 at −13.4%, November 2025 at −20.17%, June 2026 at −18.9% and July 2026 at −20.47%. The June and July 2026 months were consecutive: the position bought on 2026-06-08 at 76.83 was stopped out a day later at 67.16, a loss of 12.59%, and the buy on 2026-06-25 at 77.59 lost 14.65% by 2026-07-17.

The winning months came in runs: March to July 2023, at 13.62%, 0.07%, 22.44%, 18.15% and 10.37%, and May to June 2024 at 18.25% and 17.91%. The rule earns its return in a few long holds and spends the rest of the time paying small stop losses.

Every trade

trend + trailing stop on TQQQ made 29 closed round trips and one position still open at the end of the test, an average hold of 42 days, an average winner of 25.49%, an average loser of −10.15%, a profit factor of 1.44, a longest losing streak of 3. It held a position at the close on 60.7% of trading days.

Prices are adjusted for splits and dividends, so they sit below the quotes printed at the time. An open position is marked at the last close.

The trade list has 30 entries, 29 closed and 1 open. The median hold was 21 days and the average was 41.7. The shortest was 1 day and the longest 177 days. The median return was −5.95%, which is a poor number next to a 25.03% CAGR, and it says the result depends on the winners. Of the 29 closed trades, 13 won, with an average win of 25.49%.

The best trades were 62.1% from 2023-10-23 at 16.36 to 2024-04-17 at 26.52 over 177 days, 62.02% from 2023-03-17 at 11.93 to 2023-08-08 at 19.329 over 144 days, 50.2% from 2025-05-19 to 2025-11-07 over 172 days and 45.18% from 2026-04-14 to 2026-06-05 over 52 days. The fifth was 34.87% from 2024-04-18 to 2024-07-18. All five were held for more than 50 days. Prices are adjusted for splits and dividends, so these entries are lower than quotes at the time.

The worst trades were all short: −17.03% from 2024-08-02 to 2024-08-05 in 3 days, −14.65% over 22 days from 2026-06-25, −13.58% over 9 days from 2023-08-09, −12.85% over 8 days from 2025-11-10 and −12.59% in 1 day from 2026-06-08. The stop is 15% below the high-water mark, and a 3x fund can fall that far in a day or two after entry, as the 3-day and 1-day losses show.

The clusters of consecutive losers are the signature of the rule. On 2024-07-19 the rule bought at 35.58, was stopped on 2024-07-25 at 31.28 for −12.09%, bought again on 2024-07-26 at 31.69 and was stopped on 2024-08-02 for −7.83%. It re-entered the same day at 29.12 and was stopped on 2024-08-05 for −17.03%. Then it re-entered on 2024-08-05 at 24.12 and held to 2024-09-03 for 26.27%. That is three losses in about two weeks followed by a 26.27% win, the same pattern the strategy's caveat describes: the 200-day average stays satisfied while the stop keeps firing. November 2025 and June to July 2026 show the same sequence.

The open trade at the end was entered on 2026-07-29 at 61.09 and was up 32.61% on 2026-10-02. It is not counted among the 29, and the 25.03% includes its mark-to-market value.

Selling at the stop and buying back higher

The trade list has several cases where the stop sold and the 200-day condition bought again at a higher price within days. The 50.2% trade ended at $51.20 on 2025-11-07. The rule bought again on 2025-11-10 at $54.96, and by 2025-11-18 it had sold at $47.90 for a 12.85% loss. It bought at $48.92 on 2025-11-19, sold at $45.17 on 2025-11-21 for 7.67%, and bought a third time at $48.46 on 2025-11-24. That third trade ended on 2026-02-04 with a gain of 1.15% over 72 days, the only one of the three that did not lose. Over those three entries the rule paid more to re-enter than it received at the first exit.

June 2026 shows the same sequence in a faster form. The 45.18% trade ended at $74.56 on 2026-06-05. The next entry came on 2026-06-08 at $76.83 and ended on 2026-06-09 at $67.16, a one-day loss of 12.59%. The rule bought at $71.87 on 2026-06-10 and sold at $72.04 on 2026-06-24 for 0.23%, then bought at $77.59 on 2026-06-25 and sold at $66.22 on 2026-07-17 for a 14.65% loss. A further entry at $66.53 on 2026-07-17 ended at $60.84 on 2026-07-28 for 8.55%. The last position opened at $61.09 on 2026-07-29.

The winners between these clusters are of the length that pays for them. The trade from 2024-09-04 to 2024-12-20 gained 29.93% over 107 days, and the one from 2024-08-05 gained 26.27% in 29 days. By exit year, the rule won 4 of 7 trades in 2024 and 2 of 6 in 2025, and the longest winning streak in the whole run was 2. A rule with an average loss of 10.15% needs the average win of 25.49% to arrive often enough, and the year table shows that it did in 2023 and 2024 and less often afterwards.

Largest drawdowns

Buy-and-hold's deepest drawdown ran from 2021-11-19 to 2022-12-28 and reached −80.8%.

The largest drawdown was 43.94%, from 2026-06-02 to 2026-07-29, and it had not recovered at the end of the data. It is the drawdown of the June and July 2026 trades: a peak during the 45.18% trade, which ended on 2026-06-05, then a −12.59% trade, a flat one at 0.23%, a −14.65% trade and a −8.55% trade. The second was 37.48% from 2024-07-10 to 2024-08-07, recovered on 2024-12-04, the cluster described above. The third was 34.88% from 2025-10-29 to 2026-03-19, recovered on 2026-05-26.

Holding's three drawdowns were 80.77% from 2021-11-19 to 2022-12-28, 57.4% from 2024-12-16 to 2025-04-08 and 36.73% from 2025-10-29 to 2026-03-30. The rule's worst of 43.94% is 36.83 points shallower than the fund's, and the fund's worst fall, from 2021 to 2022, was one the rule mostly sat out in cash. In later years the rule's drawdowns of 37.48% and 34.88% were close to the fund's own 36.73% in 2025 to 2026, so the stop limits the damage of a long decline and does little in a series of sharp reversals.

The stop's own drawdown is built in: every winner gives back 15% from its peak before the exit. Every winner exits only after a 15% fall from its peak, so the 62.1% trade peaked above that figure. The rule does not try to sell the top, and the trade list shows what the trailing exit costs.

With trading costs

The headline run fills at the bar price. These runs charge slippage on every fill.

At 5 basis points the CAGR was 24.42% with a drawdown of 44.23% and an ending value of $35,054. At 10 basis points it was 23.79% with a drawdown of 44.52% and $34,049. The Sharpe went from 0.75 to 0.738 and 0.725. The rule made 59 fills, and the 10 basis point run took a visible share of the CAGR.

The effect comes from 29 round trips in 5.7 years. TQQQ's liquidity is deep, with average daily dollar volume of $4,150,045,180 and median minute volume of 285,786.5, so the assumed costs are reasonable for a fund this size. At 10 basis points the CAGR of 23.79% is below holding's 25.35%, so the rule trails the fund on return in both cost runs.

Fills, Sharpe and the cost of each stop

The headline run made 59 fills because every stop-out is a sale followed by a fresh purchase, so the stop adds fills that the plain 200-day filter does not make. The Sharpe ratio of 0.75 in the headline run is 0.738 at 5 basis points and 0.725 at 10, and the ending values are $36,064, $35,054 and $34,049. The ratio stays above holding's 0.69 in the 5 basis point run and in the 10 basis point run, so the better risk-adjusted result survives the cost assumption even though the CAGR does not.

Changing the parameters

The three variants change the trail. A 10% trail returned 22.25% with a 40.9% drawdown over 70 trades and 31 wins. A 20% trail returned 21.55% with a 41.69% drawdown over 19 trades and 8 wins. A 25% trail returned 16.84% with a 48.92% drawdown over 11 trades and 5 wins. The published 15% trail returned 25.03% with a 43.94% drawdown and the best CAGR of the four.

The shape is not monotonic. Tightening the trail to 10% lowered the return and trimmed the drawdown, while widening it to 25% lowered the return further and deepened the drawdown. The 15% setting sits at the top of the return column but the 20% setting has a drawdown of 41.69%, shallower than 15%.

With 11 to 70 trades per variant, the differences are driven by a handful of trades, especially the 2023 and 2024 winners. The trade counts moved from 70 at a 10% trail to 11 at 25%. A parameter chosen from this table would be fitted to one window. Median figures across all 59 funds for the variants are on the strategy page.

Sharpe ratios and ending values across the trails

The Sharpe ratios fall in a clear order: 0.75 for the published 15% trail, 0.706 for 10%, 0.679 for 20% and 0.582 for 25%. The ending values are $36,064, $31,684, $30,658 and $24,436. The 10% trail made 70 trades and the best Sharpe among the variants, but it still trailed the 15% rule on CAGR, Sharpe ratio and ending value. The 25% trail made 11 trades and had the deepest drawdown at 48.92%, because a stop that wide gives back a quarter of the peak before it acts. The 15% setting is the best of the four on this fund and this window. A different fund or a different bear market would likely rank the four differently, and no variant here reached the 29.44% of the weekly target or the 39.31% of the RSI(2) snapback on the same fund.

How TQQQ behaved

On TQQQ the rule had long trends to hold. The fund moves three times the daily move of QQQ, with annualized volatility of 67.11% and trends of long duration: 198.32% in 2023 and 58.22% in 2024. The fund was above its 200-day average on 69.48% of sessions and crossed it 28 times. Those crossings are the source of the repeated small losses, and the long periods above the average are the source of the 62% winners.

The leverage numbers add context. TQQQ returned 291.92% in total against 151.07% for QQQ, and a frictionless 3x daily-reset version of QQQ would have returned 568.66%. That ideal leaves out fees, borrowing costs and interest on cash, so the real fund sits below it. The fund came in under the ideal in each of the six years, for example 198.32% against 236.8% in 2023 and 34.36% against 50.54% in 2025. In 2022 the fund lost 79.07% against 77.46% for the ideal. A strategy that holds the fund for 60.7% of sessions earns the compounding in those long holds.

Among leveraged funds, TQQQ is first for this template at 25.03%. SOXL returned 17.82% with an 81.96% drawdown and 97 round trips, QLD 17.16% with 40.15%, ROM 16.46% and TECL 14.4% over 47 round trips. SSO and SPUU returned 11.43% and 11.01% with fewer trades. UST, FAS and TMF lost money at −8.66%, −11.05% and −12.92%. The median for the category was 14.4% and for all 59 funds 2.61%. SOXL needed 97 round trips for its 17.82%.

Calendar-month and weekday averages

TQQQ's average return by calendar month was 14.75% in May and 12.06% in November, with June at 7.24%, October at 6.79% and July at 6.16%. September averaged -5.42%, February -3.82% and December -2.51%. The rule's own monthly table fits this loosely. November 2023 made 33.05% and May 2023 22.44%, while September 2023 lost 15.66% and February 2025 lost 9.93%. November 2025, though, lost 20.17% for the rule. Each month has five or six observations, so the averages describe one window.

Monday averaged 0.53% and Wednesday 0.28%, against 0.03% to 0.06% for the other days. The lag-1 autocorrelation was -0.04. Because 74.87% of the log return came overnight and the stop sells at the open or on a break of the level, the rule is exposed to gaps that a daily-decision test captures only on the next bar. RSI(14) fell below 30 on 18 sessions and the median 5-day return afterwards was 10.3%, against 1.24% for all days. A trailing stop sells after a 15% fall from a peak and so tends to be out of the market for a rebound of that kind.

The rules

Enter when price is above the 200-day average; exit only when a 15% trailing stop is hit.

A hybrid of the regime filter and the trailing stop. The 200-day average decides when to get in, and a 15% trail from the position's high-water mark decides when to get out. Winners can run well past the point where a crossover would have exited, and the position gives back 15% from every peak.

Good for: long trends with moderate pullbacks; the 15% trail is wide enough to survive normal corrections.
Watch out: every exit gives back 15% from the peak, and on choppy assets it re-enters above the 200-day repeatedly after each stop-out.

The two rules interact through the 200-day average. The entry needs yesterday's close above the average, and TQQQ spent 69.48% of sessions there, so after any stop-out the entry condition is often true again the next day. The stop then sits 15% below the new entry price. On a fund with an average daily range of 4.74%, a 15% stop is only a few average days wide, so a short reversal can end the trade.

That mechanism explains the median trade of −5.95% and the win rate of 45%. More than half of the closed trades lost. The ones that worked held for months, with a stop that moved up behind a trend. The trailing stop therefore turns a strategy that would hold through 2023 into a strategy that exits and re-enters several times, and the 2023 and 2024 trade lists show both outcomes.

The comparison with similar rules on the same fund is useful. The 200-day regime filter returned 24.95% with a 36.54% drawdown, a slightly lower return and a shallower drawdown with no stop. The SMA 10/50 trend returned 17.3% with a 53.29% drawdown. The golden cross returned 7.33%. On the dip-buying side, RSI(2) snapback returned 39.31% with a 43.2% drawdown.

Related

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.