DeployQuantTQQQ trading strategies, backtested
ProShares UltraPro QQQ: 3x daily leveraged Nasdaq-100, with large trends and deep drawdowns. Every DeployQuant template run on TQQQ over 5.7 years of minute data, same engine, same window, sorted by return.
TQQQ is the ProShares UltraPro QQQ, a fund that aims for three times the daily return of the Nasdaq-100. Each of the 12 DeployQuant templates started with $10,000 on 2021-01-04 and ran to 2026-10-02. Buying and holding TQQQ made 25.35% a year, turned $10,000 into $36,585, and went through an 80.77% max drawdown. The Sharpe was 0.69. The headline runs use no margin and have no fees or slippage. Cost runs add 5 and 10 basis points per trade.
Two of the 12 templates beat holding on CAGR, 11 had a shallower drawdown, and all 12 finished with a positive CAGR. The winner is the RSI(2) snapback at 39.31% a year with a 43.2% drawdown. The weekly 7% target is second at 29.44% with a 49.66% drawdown. Only those two pass buy-and-hold, and one of them does not survive costs.
TQQQ is the fund where the choice of rule matters most in this dataset. It lost 79.07% in 2022 and gained 198.32% in 2023. A rule that was out of the fund in 2022 and in it for 2023 had a very different path from one that was invested throughout, and the year-by-year table shows which templates managed each.
The numbers are one backtest on one window, with daily-decision rules and fills on minute bars. They show how each template behaved on this fund between these dates and do not forecast the next period.
| Strategy | CAGR | max DD | Sharpe | trades | win rate | final value |
|---|---|---|---|---|---|---|
| RSI(2) Dip Snapback | 39.3% | −43.2% | 0.96 | 166 | 68% | $67,076 |
| Weekly Entry + 7% Target | 29.4% | −49.7% | 0.79 | 227 | 50% (+1 open) | $43,997 |
| SMA-200 Trend + 15% Trailing Stop | 25.0% | −43.9% | 0.75 | 29 | 45% (+1 open) | $36,064 |
| 200-Day SMA Regime Filter | 24.9% | −36.5% | 0.76 | 14 | 43% (+1 open) | $35,918 |
| First-to-Last Day of Month | 21.5% | −82.5% | 0.64 | 69 | 59% (+1 open) | $30,574 |
| SMA 10/50 Trend | 17.3% | −53.3% | 0.60 | 20 | 45% (+1 open) | $24,991 |
| EMA 12/26 Trend | 15.3% | −49.9% | 0.55 | 25 | 36% (+1 open) | $22,591 |
| RSI(14) Mean Reversion | 12.1% | −66.6% | 0.48 | 16 | 75% | $19,242 |
| Drawdown Dip Buyer + 8% Target | 11.9% | −79.6% | 0.49 | 9 | 100% (+1 open) | $19,075 |
| 3-Month Momentum Switch | 7.7% | −55.4% | 0.39 | 23 | 48% (+1 open) | $15,273 |
| Golden Cross (SMA 50/200) | 7.3% | −57.4% | 0.38 | 3 | 67% (+1 open) | $15,012 |
| 20-Day Momentum + Trailing Stop | 0.6% | −40.9% | 0.18 | 53 | 34% (+1 open) | $10,334 |
| Buy & hold TQQQ | 25.4% | −80.8% | 0.69 | – | – | $36,585 |
The RSI(2) snapback on a fund that swings 4.7% a day
The RSI(2) snapback buys when the 2-day RSI is very low and sells on the bounce. On TQQQ it made 166 round trips, won 68%, and was invested 36.8% of the time. The average hold was 4.7 days. The average win was 4.62% and the average loss was 5.51%, so the rule won most trades and lost more when it lost. The profit factor was 1.76. The best trade was 33.94% and the worst was negative 23.92%.
The fund moved enough per day to make that profile work. TQQQ had an average intraday range of 4.74% and an annualized volatility of 67.11%. The 2-day RSI closed under 10 on 152 sessions. After those sessions the median 5-day forward return was 1.73% against a baseline of 1.24%, and the median 20-day return was 4.54% against 3.62%. The edge per signal is small, and the rule took 166 of them with short holds. Most of the return came from repeating a small edge many times in a fund with large daily swings.
The snapback beat holding in 2021, 2022, 2024 and 2025 and lost to it in 2023 and 2026. In 2023 the fund returned 187.6% in the buy-and-hold record while the rule returned 39.9%, which is the largest single shortfall. In 2022 the rule lost 27.6% against 78.1% for the fund. Its max drawdown of 43.2% ran from 2021-12-15 to 2022-06-30 and recovered on 2023-06-09. The second drawdown, 31.27%, ran from 2025-02-05 to 2025-04-02.
Costs take part of the return and leave the rule ahead. At 5 basis points the CAGR falls to 35.43%, with end equity of $57,054. At 10 basis points it falls to 31.66%, with $48,506. Both are above the 25.35% for holding. The drawdown goes from 43.2% to 44.04% and 44.96%. On VOOG the same rule lost its edge at 5 basis points. On TQQQ the larger per-trade moves leave room for the cost.
The result is on a fund that fell more than 80% from peak to trough. A rule that buys short-term dips on that fund carried 43.2% of drawdown through 2022. That is still shallower than holding, and the entries are exposed to a fund that can fall 18.3% in one session, as it did on 2025-04-04.
Why the weekly target looks good before costs and weak after
The weekly 7% target buys at the first open of the week, rests a limit order 7% above the entry, and sells on Thursday at 2:00pm if the trade is losing. On TQQQ it made 227 round trips with a 50% win rate, and the average win of 7.4% sat against an average loss of 5.51%. The profit factor was 1.26, which is thin. The average hold was 4.6 days, and the rule was invested 61% of the time. Its CAGR was 29.44% with a 49.66% drawdown.
A profit factor of 1.26 leaves little room for friction, and the cost runs show it. At 5 basis points the CAGR falls to 18.69% with a 54.87% drawdown and end equity of $26,748. At 10 basis points it falls to 15.6% with a 55.47% drawdown and $22,990. At 5 basis points it already trails buy-and-hold at 25.35%. The headline claim that 2 of 12 templates beat holding becomes 1 of 12 once a small cost is included.
The weekly target had a strong 2023 with 101.1% against 187.6% for the fund, and a mild 2022 at negative 12.1%. It beat holding in 2022 and 2025 and trailed in the other four years, by 63.7 points in 2021 and 86.5 in 2023. Its longest drawdown ran from 2022-04-04 to 2022-07-18 and recovered on 2023-05-17. The most recent one, from 2026-06-02, had not recovered by the end of the window and had reached 37.47%.
The contrast with VOOG is useful. On VOOG the same rule won only 32% of its trades but its average win was much larger than its average loss. On TQQQ it wins half its trades and the average win is only a little larger than the average loss. The 7% target is reached easily on a fund with a 4.74% average range, which makes wins common and small. The Thursday exit cuts losers at an average of 5.51%, and the result is a rule that pays out little per trade.
The slow trend rules and the costly ones
The trend plus trailing stop rule made 25.03% with a 43.94% drawdown, and the 200-day regime filter made 24.95% with a 36.54% drawdown. Both trailed holding by a small margin and held the drawdown well below the fund's. The 200-day filter made 14 round trips, won 43%, and had an average win of 28.31% against an average loss of 4.01%. Its profit factor of 4.11 is the highest in the table. Costs barely touched either rule: the 200-day filter fell only to 24.37% at 10 basis points.
Both rules lost 18% to 20% in 2022 against 78.1% for the fund, and both returned 82% to 83.3% in 2023 against 187.6%. They are on the right side of the 2022 decline and then late back into the recovery. In 2021 they returned 17.5% and 18.6% against 88.9% for holding, because the window starts with them flat. The shortfall in 2023 and 2021 is the whole difference between them and holding.
The SMA 10/50 trend rule made 17.3% with a 53.29% drawdown, and the EMA 12/26 trend rule made 15.25% with a 49.92% drawdown. Faster averages whipsaw more on a fund with 4.74% daily ranges. Their 2022 results were negative 45.9% and negative 34.9%, much worse than the slow rules. The EMA rule won only 36% of its 25 trades, though its average win of 31.41% was large.
The monthly cycle rule held the fund 95.2% of the time and returned 21.49%. Its max drawdown was 82.48%, slightly deeper than the 80.77% for holding. It is the one template that did not reduce the drawdown, and it lost 80.3% in 2022.
The RSI(14) mean reversion template made 12.08% and was invested 29.8% of the time. It won 75% of its 16 trades, but its average loss of 25.02% was larger than its average win of 16.71%, and its drawdown was 66.58%. The 14-day RSI closed under 30 on only 18 sessions, and the median 5-day forward return after those was 10.3% against a 1.24% baseline. The rebound was large but the exit rule waits for RSI above 70, which takes weeks, and the position carried the 2022 decline in between. The dip buyer won all 9 of its trades at around 8% each and still made only 11.91% with a 79.56% drawdown, since it was invested 87% of the time with an average hold of 190.7 days.
The 3-month momentum rule made 7.66%, the golden cross made 7.33% with only 3 trades, and the momentum breakout made 0.57% with 53 round trips and a 34% win rate. The breakout's profit factor was 1, and at 10 basis points its CAGR went to negative 1.2%. These three are the weakest on this fund.
How each strategy traded TQQQ
| Strategy | Time in market | Avg hold (days) | Best trade | Worst trade | Profit factor | With 10 bps slippage |
|---|---|---|---|---|---|---|
| RSI(2) snapback | 36.8% | 5 | 33.9% | −23.9% | 1.76 | 31.7% |
| weekly 7% target | 61.0% | 5 | 15.2% | −18.1% | 1.26 | 15.6% |
| trend + trailing stop | 60.7% | 42 | 62.1% | −17.0% | 1.44 | 23.8% |
| 200-day regime filter | 59.8% | 85 | 66.9% | −9.7% | 4.11 | 24.4% |
| monthly cycle | 95.2% | 28 | 46.8% | −32.3% | 1.45 | 18.7% |
| SMA 10/50 trend | 58.8% | 61 | 76.0% | −24.2% | 2.07 | 16.5% |
| EMA 12/26 trend | 61.1% | 51 | 74.5% | −16.9% | 1.63 | 14.3% |
| RSI mean reversion | 29.8% | 39 | 35.1% | −33.6% | 1.83 | 11.5% |
| dip buyer | 87.0% | 191 | 8.0% | 8.0% | – | 11.9% |
| 3-month momentum | 58.2% | 54 | 41.2% | −19.1% | 1.41 | 6.8% |
| golden cross | 60.4% | 372 | 80.3% | −29.7% | 1.95 | 7.2% |
| momentum breakout | 45.5% | 18 | 44.2% | −11.6% | 1.00 | −1.2% |
Trade statistics across the templates
The trade table splits the templates into three groups. The frequent traders are the RSI(2) snapback with 166 round trips, the weekly target with 227, the monthly cycle with 69 and the breakout with 53. Their average holds run from 4.6 to 28.4 days. The middle group is the trend rules with 14 to 29 round trips. The slow group is the golden cross with 3 and the dip buyer with 9.
Costs follow that split. The frequent traders lose the most CAGR at 10 basis points: the weekly target loses the most, the snapback the next most, and the breakout turns negative. The trend rules lose little, and the golden cross and dip buyer barely move. The cost runs here are per-trade slippage added to each fill, so a rule that trades often pays it on every entry and every exit.
The win rates and the average win and loss figures show where each template's return comes from. The golden cross won 2 of 3 trades with an average win of 40.48% and an average loss of 29.73%. The best trade was 80.25%. Three trades is too few to say anything about the rule. The trend and trailing-stop rule won 45% of 29 trades, with an average win of 25.49% against an average loss of 10.15%. The best trade was 62.1%.
The longest loss streaks were 7 trades for the weekly target, 5 for the breakout, and 4 for the SMA, EMA and 200-day rules. A fund that moves this much produces losing runs of that length without any change in the rules. Traders sizing a sleeve should assume losing streaks of that kind and should check the drawdown column as well as the CAGR. The drawdown for every template here is deep compared with the broad index funds such as QQQ, where the same rules earn less and draw down less.
How TQQQ behaved
| Measure | TQQQ |
|---|---|
| Data in this test | 2021-01-04 to 2026-10-02 (1444 sessions) |
| Total return, buy and hold | 291.9% |
| Annualized volatility | 67.1% |
| Deepest drawdown | −81.7% (2021-11-19 to 2022-12-28) |
| Up days | 54.5% |
| Average daily range | 4.74% |
| Average overnight gap | 1.78% |
| Correlation to SPY | 0.94 |
| Correlation to QQQ | 1.00 |
| Correlation to TLT | 0.09 |
| Sessions above the 200-day average | 69.5% |
| Crossings of the 200-day average | 28 |
| Falls of 10% or more from a 20-day high | 56 |
Calendar years
| Year | Return |
|---|---|
| 2021 | 91.0% |
| 2022 | −79.1% |
| 2023 | 198.3% |
| 2024 | 58.2% |
| 2025 | 34.4% |
| 2026 (part) | 54.6% |
Biggest single days
| Best day | Move |
|---|---|
| 2025-04-09 | 35.3% |
| 2022-11-10 | 22.1% |
| 2022-11-30 | 13.7% |
| 2022-07-27 | 12.7% |
| 2025-05-12 | 12.1% |
| Worst day | Move |
|---|---|
| 2025-04-04 | −18.3% |
| 2022-09-13 | −16.6% |
| 2025-04-03 | −16.0% |
| 2022-05-05 | −14.9% |
| 2022-05-18 | −14.8% |
Average return by calendar month
| Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 3.6% | −3.8% | 0.7% | 2.4% | 14.8% | 7.2% | 6.2% | 0.5% | −5.4% | 6.8% | 12.1% | −2.5% |
Most and least correlated funds
| Most correlated | Least correlated | ||
|---|---|---|---|
| QQQ | 1.00 | SQQQ | -1.00 |
| QLD | 1.00 | QID | -1.00 |
| QQQM | 1.00 | PSQ | -1.00 |
| VOOG | 0.98 | TECS | -0.97 |
| ROM | 0.97 | REW | -0.97 |
Leverage and decay against QQQ
| Year | TQQQ | QQQ | 3× QQQ, reset daily |
|---|---|---|---|
| 2021 | 91.0% | 29.2% | 95.1% |
| 2022 | −79.1% | −32.4% | −77.5% |
| 2023 | 198.3% | 54.8% | 236.8% |
| 2024 | 58.2% | 25.6% | 79.5% |
| 2025 | 34.4% | 20.8% | 50.5% |
| 2026 | 54.6% | 22.4% | 67.1% |
The last column compounds 3× QQQ's daily return with no fees or financing: the return a perfect daily-reset fund would have had. It leaves out the fund's fees, its borrowing costs and the interest it earns on cash, which put a real fund below or above this line. A daily-reset fund does not aim for 3× the underlying's return over a year.
Over 2021-01-04 to 2026-10-02, TQQQ returned 291.9% while QQQ returned 151.1% and a perfect daily-reset 3× version would have returned 568.7%. Its measured daily beta to QQQ was 3.00.
What TQQQ did in the window
TQQQ returned 291.92% in total, or 26.86% a year on the fund's own price series, with annualized volatility of 67.11%. Its max drawdown was 81.68%, from the peak on 2021-11-19 to the trough on 2022-12-28. It recovered that peak on 2024-12-04, and the longest time under a prior high was 762 sessions. The calendar years were 90.95% in 2021, negative 79.07% in 2022, 198.32% in 2023, 58.22% in 2024, 34.36% in 2025 and 54.6% in 2026 to date.
Two more drawdowns show how frequent large falls were. The buy-and-hold record has a 57.4% drawdown from 2024-12-16 to 2025-04-08, which recovered on 2025-08-12, and a 36.73% drawdown from 2025-10-29 to 2026-03-30. The fund had 56 drawdown events of 10% or more inside 20 days, spread over 478 sessions. A tight stop would have met many of them.
The daily moves are large. TQQQ rose on 54.47% of sessions, with an average up day of 3.01% and an average down day of negative 3.22%. The best day was 35.25% on 2025-04-09, then 22.13% on 2022-11-10 and 13.74% on 2022-11-30. The worst was negative 18.3% on 2025-04-04, then negative 16.59% on 2022-09-13 and negative 16.04% on 2025-04-03. The 2025-04-09 gain came three sessions after the 2025-04-04 loss, so a rule that exited after the first drop missed the rebound.
The overnight share of log return was 74.87% and the intraday share was 25.13%, with an average overnight gap of 1.78%. TQQQ earns a larger intraday share than VOOG did, with an average intraday range of 4.74%. First-order autocorrelation was negative 0.04, which is near zero. Daily returns carried almost no memory, so the profit in the RSI(2) rule came from the size of the short-term swings and not from a reliable reversal after each one.
The fund closed above its 200-day average on 69.48% of sessions and crossed it 28 times. The 200-day regime filter was invested 59.8% of the time and the trend and trailing-stop rule 60.7%. The 14-day RSI closed above 70 on 120 sessions, which helps explain how a position bought at RSI under 30 was held for 39.2 days on average before the exit signal.
Leverage, decay and the QQQ comparison
TQQQ tracked QQQ with a measured daily beta of 3, an r-squared of 1, and a correlation of 1. Its beta to SPY was 3.86 and to TLT was 0.39. Over the full window TQQQ returned 291.92% while QQQ returned 151.07%. Three times the QQQ return would be 453.22%, so the fund returned less than a simple tripling of the index over the window.
The year-by-year gap shows how path matters. In 2021 TQQQ returned 90.95% against 87.54% for three times QQQ, a gap of 3.42 points. In 2022 it lost 79.07% against 97.18% for three times QQQ, 18.12 points better. In 2023 it returned 198.32% against 164.42%, 33.9 points better. In 2024 it returned 58.22% against 76.77%, 18.56 points worse. In 2025 it returned 34.36% against 62.31%, 27.96 points worse, and in 2026 it returned 54.6% against 67.31%, 12.71 points worse.
The pattern follows how the year moved. Daily rebalancing compounds gains in a steady rise and reduces the base in a fall. Over a year with a big decline followed by a sharp rise, the compounded result can land on either side of a simple tripling. The three-times figures also ignore fees and financing costs inside the fund. These tables describe what happened in each calendar year, and the data here does not isolate fees from path effects.
The practical point for the templates is that a rule on TQQQ holds a fund whose return differs from three times QQQ in each year. A trend rule that avoids the decline avoids the largest period of decay as well. The QLD fund at 2x had a buy-and-hold CAGR of 23.55% in this window, and the RSI(2) snapback was its best template at 26.17%.
Calendar months and other leveraged funds
The average return by calendar month was highest in May at 14.75% and November at 12.06%, and lowest in September at negative 5.42% and February at negative 3.82%. Each month has only 5 or 6 observations, and a single year such as 2023 or 2022 moves each average a lot. These are descriptions of the window. They do not set up a calendar rule.
Among the other leveraged funds the best template differs by fund. The RSI(2) snapback was best on QLD at 26.17% against 23.55% for holding, on SSO at 24.56% against 21.92%, and on SPUU at 24.12% against 22.48%. The golden cross was best on SOXL at 45.93% against 33.32% for holding. The monthly cycle was best on TECL at 35.97%, below holding at 38.16%. On ROM the weekly target made 30.27% against 30.19%, and on FAS the dip buyer made 24.15% against 18.34%.
On the leveraged bond funds every template trailed zero or came close to it. TMF lost 31.16% a year when held, and its best template, the RSI(2) snapback, lost 4%. UST lost 8.51% held, and its best template, the SMA 10/50 trend, made 0.68%. That contrast shows that a leveraged equity fund and a leveraged bond fund do not behave alike in this window, and rules built for one should not be applied to the other.
The test stops at one fund and one window. It does not model margin, borrow, taxes, or the effect of a large order on a fund that trades an average of $4,150,045,180 a day. The median minute volume was 285,786.5 shares, so slippage for a small order should be close to the cost runs.
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.
Frequently asked questions
What was the best strategy for TQQQ?
Of the 12 templates tested on TQQQ over 2021-01-04 to 2026-10-02, the strongest by CAGR was RSI(2) snapback at 39.3% (max drawdown 43.2%), versus 25.4% for buy-and-hold. The best result in hindsight is not a forecast. Check drawdowns and trade counts before drawing conclusions.
Did any strategy beat buying and holding TQQQ?
2 of 12 templates beat TQQQ buy-and-hold (25.4% CAGR) on this window; 11 of 12 had a shallower maximum drawdown than holding (80.8%).
Which strategy worked best on TQQQ?
The RSI(2) snapback returned 39.31% a year with a 43.2% max drawdown, against 25.35% and 80.77% for buy-and-hold. It held up under costs, with 35.43% at 5 basis points and 31.66% at 10. The weekly 7% target was second at 29.44% before costs.
Does the weekly 7% target beat buy-and-hold TQQQ?
Only in the headline run, where it made 29.44% against 25.35%. With 5 basis points of slippage it made 18.69%, and with 10 it made 15.6%. Its profit factor was 1.26, which leaves little room for costs.
How deep did TQQQ fall in 2022?
TQQQ lost 79.07% in calendar 2022. Its max drawdown was 81.68%, from 2021-11-19 to 2022-12-28, and it recovered that peak on 2024-12-04. The longest time under a prior high was 762 sessions.
Do trend rules like the 200-day filter work on TQQQ?
The 200-day regime filter returned 24.95% with a 36.54% drawdown, close to the 25.35% for holding with a much shallower path. It lost 18.3% in 2022 against 78.1% for the fund. It trailed holding in 2021 and 2023, when the fund rose 88.9% and 187.6%.
How does TQQQ compare with three times QQQ?
Over the window TQQQ returned 291.92% and QQQ returned 151.07%. Three times the QQQ return would be 453.22%. By calendar year the gap runs from 33.9 points better in 2023 to 27.96 points worse in 2025.
How many trades did the strategies make on TQQQ?
The weekly target made 227 round trips, the RSI(2) snapback 166, and the monthly cycle 69. The golden cross made 3 and the dip buyer 9. Fast rules were the ones most affected by slippage.
Other leveraged etfs
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.