DeployQuantSPUU trading strategies, backtested
Direxion Daily S&P 500 Bull 2X Shares: Direxion's 2x daily S&P 500 fund. Every DeployQuant template run on SPUU over 5.7 years of minute data, same engine, same window, sorted by return.
SPUU is Direxion's 2x daily S&P 500 fund. From 2021-01-04 to 2026-10-02, holding it returned 22.48% a year, turned $10,000 into $32,040, and fell as far as 46.08% from the 2021-12-29 peak to the 2022-10-12 low. The fund took until 2024-02-22 to get back to that peak. That is a leveraged index tracker with a deep drawdown, and the 12 templates tested here show what rules can do with a fund like that.
Two of the 12 templates beat holding on CAGR: RSI(2) snapback at 24.12% and the weekly 7% target at 23.28%. All 12 had a shallower drawdown than holding, and all 12 finished with a positive CAGR. The best drawdown was 17.60% on 20-day momentum with a trailing stop.
The headline runs have no fees or slippage. SPUU is thinly traded in the data, with an average daily dollar volume of $2,875,224 and a median of 233 shares a minute, so the cost runs matter more here than on a large fund. The window is one path of 1,444 sessions.
| Strategy | CAGR | max DD | Sharpe | trades | win rate | final value |
|---|---|---|---|---|---|---|
| RSI(2) Dip Snapback | 24.1% | −31.2% | 1.09 | 171 | 68% (+1 open) | $34,576 |
| Weekly Entry + 7% Target | 23.3% | −27.0% | 0.94 | 146 | 38% (+1 open) | $33,253 |
| First-to-Last Day of Month | 19.5% | −44.1% | 0.75 | 69 | 65% (+1 open) | $27,825 |
| EMA 12/26 Trend | 19.1% | −22.9% | 0.95 | 21 | 48% (+1 open) | $27,307 |
| 20-Day Momentum + Trailing Stop | 17.3% | −17.6% | 0.91 | 12 | 75% (+1 open) | $24,932 |
| Drawdown Dip Buyer + 8% Target | 14.2% | −41.3% | 0.63 | 10 | 100% | $21,403 |
| Golden Cross (SMA 50/200) | 13.2% | −34.7% | 0.65 | 3 | 67% (+1 open) | $20,402 |
| SMA 10/50 Trend | 11.2% | −31.4% | 0.62 | 14 | 57% (+1 open) | $18,375 |
| SMA-200 Trend + 15% Trailing Stop | 11.0% | −39.6% | 0.61 | 8 | 50% (+1 open) | $18,216 |
| 200-Day SMA Regime Filter | 10.5% | −37.3% | 0.61 | 17 | 24% (+1 open) | $17,776 |
| 3-Month Momentum Switch | 9.8% | −38.6% | 0.58 | 18 | 56% (+1 open) | $17,061 |
| RSI(14) Mean Reversion | 5.4% | −34.6% | 0.34 | 12 | 75% | $13,507 |
| Buy & hold SPUU | 22.5% | −46.1% | 0.81 | – | – | $32,040 |
Where the templates finished against a 2x fund
Holding SPUU had a Sharpe of 0.81. The RSI(2) snapback template had 1.09 with a 31.17% drawdown, and the weekly 7% target had 0.94 with 26.96%. EMA 12/26 had 0.95 with a 22.88% drawdown and a 19.12% CAGR. In risk-adjusted terms those three were ahead of holding. EMA 12/26 had the shallowest drawdown of the three at 22.88%, against 46.08% for holding.
RSI(2) snapback made 171 round trips and won 68% of them. It spent 37.9% of the time invested, with an average hold of 4.7 days and a profit factor of 2.03. Its yearly returns were 54.2% in 2021, negative 15.0% in 2022, 20.6% in 2023, 39.0% in 2024, 29.9% in 2025 and 21.1% in 2026. Against holding, it was 23.2 points ahead in 2022 and 3.6 points ahead in 2025, and behind in the other four years, by 25.6 points in 2023. On a 2x fund a 2-day RSI under 10 is a sharp washout. The fund's median forward 5-day return after RSI(2) fell under 10 was 1.38% against a baseline of 0.72%, and the 20-day return was 4.11% against 2.94%. Its best trade, entered 2025-04-07 and closed 2025-04-10, made 17.79% in 3 days. The worst, from 2025-02-24 to 2025-03-18, lost 12.54%. A 2x fund magnifies each, so the average win was 2.34% and the average loss 2.47%.
The weekly 7% target won only 38% of 146 round trips, with a losing streak of 7. Its average win was 7.39% and its average loss was 2.92%. It was invested 77.2% of the time and made 23.28%. It lost 4.3% in 2022 while holding lost 38.2%, and that year decided the comparison: it gained 33.9 points on holding in 2022 and 3.5 points in 2024, and trailed by double digits in 2021, 2023 and 2026.
The monthly cycle returned 19.51% but its drawdown was 44.11%, nearly the same as holding. It was invested 95.2% of the time. Its worst trade was June 2022, entered 2022-06-01 and closed 2022-06-30 for negative 18.01%, and its best was April 2026 at 18.52%. It lost 41.4% in 2022 against 38.2% held, so it gave no protection in the bear year.
The trend templates lagged in return and cut the drawdown. EMA 12/26 made 21 round trips with a profit factor of 4.99. Its three best trades returned 36.78%, 32.51% and 26.28%, each held about six months, and its average win was 14.78% against an average loss of 2.67%. The golden cross returned 13.22% on 3 round trips, one of which made 63.19% over 770 days. SMA 10/50 returned 11.18%, the trailing-stop template 11.01% and the 200-day filter 10.54%. The 200-day filter won 24% of its 17 round trips and relied on one trade of 65.38% over 488 days.
The dip buyer returned 14.17% from 10 trades, all winners at the 8% target, and still carried a 41.32% drawdown. Its worst year was 2022 at negative 31.9%. One trade, entered 2022-01-20 and closed 2024-02-07, was held for 748 days to reach 8.02%. A 100% win rate here came with a drawdown close to holding's, because the one long trade sat through the whole 2022 decline.
At the bottom, RSI(14) mean reversion returned 5.38%. It won 75% of its 12 trades, but its average win was 8.23% and its average loss was 12.17%, and the two big losses were 18.51% from 2022-04-25 to 2022-07-11 and 14.98% from 2024-12-20 to 2025-04-30. A high win rate with an average loss larger than the average win produced a drawdown of 34.59%.
For a leveraged fund, the lesson from the table is that exposure control mattered more than signal quality. Every template had a shallower drawdown than holding, and the ones that sat out most of 2022 did best on that measure.
The 2022 column shows the split. Holding lost 38.2% that year. The weekly target lost 4.3%, momentum 6.7%, RSI(2) snapback 15.0%, EMA 12/26 15.9%, and the 200-day filter 26.2%. The monthly cycle lost 41.4% and the dip buyer 31.9%. In the rebound years, 2023 and 2024, holding made 46.2% and 43.6%, and most templates trailed it. The golden cross and the 200-day filter were the two that made 43.6% and 43.1% in 2024 but only 41.9% and 8.7% in 2023, because their entries came after the rebound had started.
Year by year, and the positions still open at the end
Holding SPUU made 64.3% in 2021, lost 38.2% in 2022, then made 46.2%, 43.6%, 26.3% and 22.9% in 2023, 2024, 2025 and 2026 to date. In 2021 the monthly cycle made 56%, the EMA 12/26 27.7% and the RSI(2) snapback 54.2%, while the slower templates sat near 11.7% because the 200-day average needed time to form and they entered in October. The weekly 7% target made 47.8%.
The years after 2022 are where the slow rules caught up. The 200-day filter made 43.1% in 2024, the golden cross and the trend plus trailing stop each made 43.6%, and the dip buyer made 47.6%, all close to holding's 43.6%. In 2025 the order changed. The golden cross lost 6.1% and RSI(14) mean reversion lost 9.7%, against a gain of 26.3% for holding. Both had long positions open when the fund fell from its 2025-02-19 peak to the 2025-04-08 trough, a 34.69% drawdown, and both gave up the gain. The golden cross's 63.19% trade, from 2023-03-02 to 2025-04-10, closed two days after that trough.
Several templates end the window inside a large open trade. The 200-day filter and the trend plus trailing stop each entered on 2026-04-09 at 177.43 and were up 27.23%. The golden cross has been in since 2025-07-11 at 156.17 and was up 44.55%. The momentum breakout entered on 2026-04-17 at 193.51 and was up 16.66%. Those open gains are included in the CAGR figures above, so the CAGR of the slow templates includes gains that no sell signal had yet closed on 2026-10-02.
The best and worst months show the same thing from the fund's side. Holding made 21.23% in April 2026 and lost 18.06% in September 2022. The monthly cycle reached 18.02% in April 2026 and -17.6% in June 2022, the best of 69 monthly trades being 18.52%. EMA 12/26 and the 3-month momentum switch had their worst months in December 2022 and September 2022 at -12.7% and -14.22%. The RSI(2) snapback had its best month in April 2025 at 19.26%, the 17.79% trade from 2025-04-07 to 2025-04-10 that caught the rebound after the trough.
How each strategy traded SPUU
| Strategy | Time in market | Avg hold (days) | Best trade | Worst trade | Profit factor | With 10 bps slippage |
|---|---|---|---|---|---|---|
| RSI(2) snapback | 37.9% | 5 | 17.8% | −12.5% | 2.03 | 17.1% |
| weekly 7% target | 77.2% | 10 | 11.6% | −8.8% | 1.47 | 17.5% |
| monthly cycle | 95.2% | 28 | 18.5% | −18.0% | 1.72 | 16.7% |
| EMA 12/26 trend | 69.1% | 69 | 36.8% | −4.3% | 4.99 | 18.3% |
| momentum breakout | 59.2% | 89 | 34.4% | −6.0% | 11.14 | 16.8% |
| dip buyer | 47.9% | 100 | 8.3% | 8.0% | – | 14.1% |
| golden cross | 65.2% | 307 | 63.2% | −13.4% | 4.23 | 13.0% |
| SMA 10/50 trend | 66.8% | 97 | 30.4% | −15.7% | 2.77 | 10.6% |
| trend + trailing stop | 66.6% | 153 | 38.0% | −12.4% | 2.51 | 10.7% |
| 200-day regime filter | 63.4% | 68 | 65.4% | −8.2% | 1.99 | 9.8% |
| 3-month momentum | 65.9% | 76 | 31.1% | −14.6% | 2.22 | 9.1% |
| RSI mean reversion | 29.7% | 52 | 17.1% | −18.5% | 1.85 | 5.0% |
Costs, thin volume and the trades that carried each result
SPUU has the thin liquidity of a niche fund, with a median of 233 shares a minute. The cost runs use a flat charge per fill, so they describe sensitivity and do not model that thinness. They matter most for the templates that trade often.
RSI(2) snapback made 343 fills. At 5 basis points its CAGR fell from 24.12% to 20.53%, and at 10 basis points to 17.06%, with the drawdown rising to 34.12%. At 10 basis points it was below the weekly 7% target at 17.51% and below EMA 12/26 at 18.29%, and just above the monthly cycle at 16.73%. In that run it was no longer ahead of holding, which was at 22.48%. The weekly target made 293 fills and went from 23.28% to 20.24% and 17.51%. Neither would have beaten holding at 10 basis points. The monthly cycle went from 19.51% to 16.73%.
The templates with few fills barely changed. The golden cross went from 13.22% to 13.04%, the dip buyer from 14.17% to 14.12%, and the momentum template from 17.25% to 16.80%. The effect on the ranking is that at 10 basis points the order of the top group becomes EMA 12/26 at 18.29%, weekly target at 17.51%, RSI(2) at 17.06%, momentum at 16.80% and monthly cycle at 16.73%. The five are within two points of each other, which is inside the noise of one window.
The momentum template deserves a separate look. 20-day momentum with a trailing stop made 12 round trips, won 9, and had a profit factor of 11.14. Its drawdown was 17.60%, the shallowest in the table, and it returned 17.25%. It was invested 59.2% of the time. Its best trades were 34.38% from 2025-05-06 to 2025-11-20 and 20.67% from 2023-11-17 to 2024-04-19. It lost 6.7% in 2022, against 38.2% for holding. It currently holds a position entered 2026-04-17 at 193.51, up 16.66%. The trade count is small and the result depends on a handful of winners.
The long holds in this fund were large trades. The golden cross had one trade of 63.19%, and the 200-day filter one of 65.38%. Single trades of that size make the results sensitive to one exit date. The 200-day filter exited on 2025-03-05 and re-entered on 2025-05-30.
Hold times ran from 4.7 days on RSI(2) to 307 days on the golden cross. The trailing-stop template made 8 round trips with an average hold of 152.5 days. It lost 30.6% in 2022 and its worst closed trade was 12.37% from 2022-02-01 to 2022-02-23. It had three losing trades in a row that year. Its best trade was 38.02% over 312 days.
How SPUU behaved
| Measure | SPUU |
|---|---|
| Data in this test | 2021-01-04 to 2026-10-02 (1444 sessions) |
| Total return, buy and hold | 236.4% |
| Annualized volatility | 32.8% |
| Deepest drawdown | −46.7% (2021-12-29 to 2022-10-12) |
| Up days | 53.6% |
| Average daily range | 2.00% |
| Average overnight gap | 0.90% |
| Correlation to SPY | 0.99 |
| Correlation to QQQ | 0.94 |
| Correlation to TLT | 0.07 |
| Sessions above the 200-day average | 73.6% |
| Crossings of the 200-day average | 34 |
| Falls of 10% or more from a 20-day high | 30 |
Calendar years
| Year | Return |
|---|---|
| 2021 | 65.8% |
| 2022 | −38.7% |
| 2023 | 47.3% |
| 2024 | 44.2% |
| 2025 | 26.6% |
| 2026 (part) | 23.1% |
Biggest single days
| Best day | Move |
|---|---|
| 2025-04-09 | 18.4% |
| 2022-11-10 | 10.9% |
| 2025-05-12 | 6.6% |
| 2022-11-30 | 6.4% |
| 2022-10-04 | 6.1% |
| Worst day | Move |
|---|---|
| 2025-04-04 | −11.9% |
| 2025-04-03 | −9.7% |
| 2022-09-13 | −8.8% |
| 2022-05-18 | −7.9% |
| 2022-06-13 | −7.6% |
Average return by calendar month
| Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 2.3% | −0.5% | 1.2% | 0.8% | 5.5% | 2.5% | 5.7% | 0.9% | −4.7% | 5.4% | 7.5% | 0.1% |
Most and least correlated funds
| Most correlated | Least correlated | ||
|---|---|---|---|
| SPY | 0.99 | SDS | -0.99 |
| VOO | 0.99 | SH | -0.99 |
| SSO | 0.99 | SPDN | -0.99 |
| VV | 0.99 | SQQQ | -0.94 |
| IOO | 0.96 | QID | -0.94 |
Leverage and decay against SPY
| Year | SPUU | SPY | 2× SPY, reset daily |
|---|---|---|---|
| 2021 | 65.8% | 30.5% | 67.4% |
| 2022 | −38.7% | −18.2% | −36.8% |
| 2023 | 47.3% | 26.2% | 56.5% |
| 2024 | 44.2% | 24.9% | 53.4% |
| 2025 | 26.6% | 17.7% | 34.0% |
| 2026 | 23.1% | 13.8% | 27.8% |
The last column compounds 2× SPY'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 2× the underlying's return over a year.
Over 2021-01-04 to 2026-10-02, SPUU returned 236.4% while SPY returned 125.3% and a perfect daily-reset 2× version would have returned 335.0%. Its measured daily beta to SPY was 1.99.
How SPUU behaved in this window
SPUU's annualized volatility was 32.82%, high for an S&P 500 fund. The unleveraged comparison is VOO, and SSO is the other 2x S&P 500 fund. The average up day was 1.48% and the average down day negative 1.50%. The average intraday range was 2.00% and the average overnight gap 0.90%. The fund rose on 53.64% of days. Of its log return, 67.53% came overnight and 32.47% intraday.
Correlation to SPY was 0.99 and the realized beta was 1.99. Correlation to QQQ was 0.94 and to TLT 0.07. The most correlated funds were SPY, VOO, SSO and VV at 0.99.
The leverage table compares the fund with SPY over the same window. SPUU returned 236.37% in total and SPY returned 125.29%. The realized beta to SPY was 1.99 with an R squared of 0.99. A 2x position on SPY that resets daily would have returned 334.97% over the window. That line is frictionless: it leaves out fees, borrowing costs and interest on cash, which put a real fund below or above it, and the data does not split the shortfall into those parts. Year by year, SPUU returned 65.76% in 2021 against 67.39% for the daily-reset ideal, and negative 38.71% in 2022 against negative 36.81%. In 2023 it returned 47.26% against 56.55%, in 2024 44.24% against 53.43%, in 2025 26.59% against 33.98% and in 2026 23.13% against 27.78%. The fund tracked the ideal closely in 2021 and 2022 and sat further below it in 2023, 2024 and 2025.
The calendar years were 65.76% in 2021, negative 38.71% in 2022, 47.26% in 2023, 44.24% in 2024, 26.59% in 2025 and 23.13% in 2026 through 2026-10-02. SPY returned 30.46%, negative 18.16%, 26.18%, 24.86%, 17.72% and 13.77%. In 2022 the fund fell about twice as much as SPY, and in the up years it rose by less than twice as much in 2023 to 2026.
Holding's drawdowns were 46.08% from 2021-12-29 to 2022-10-12 with recovery on 2024-02-22 after 498 days, 34.69% from 2025-02-19 to 2025-04-08 with recovery on 2025-07-03, and 18.02% from 2026-01-12 to 2026-03-30. The longest drawdown lasted 538 sessions. The profile table shows 46.69% for the same peak and low. There were 30 falls of 10% or more from a 20-day high, over 172 days. Dip rules have many chances to trigger on this fund.
The best days were 2025-04-09 at 18.40%, 2022-11-10 at 10.88%, 2025-05-12 at 6.58%, 2022-11-30 at 6.39% and 2022-10-04 at 6.13%. The worst were 2025-04-04 at negative 11.88%, 2025-04-03 at negative 9.69%, 2022-09-13 at negative 8.84%, 2022-05-18 at negative 7.94% and 2022-06-13 at negative 7.61%. A single day moved the fund by 18.40%, and 2025-04-09 followed 2025-04-08, the drawdown low. Any rule that was out of the fund that day missed it.
Monthly averages show November at 7.50%, July at 5.70%, May at 5.53% and October at 5.38%, with September at negative 4.72% and February at negative 0.47%. With 5 or 6 observations per month, these describe the sample. The fund was above its 200-day average in 73.57% of sessions and crossed it 34 times. Autocorrelation at lag 1 was negative 0.02.
RSI(14) fell under 30 in 20 sessions, and the median forward 5-day return was 6.69% against a 0.72% baseline. The 20-day figure was 2.39%, below the 2.94% baseline. An oversold reading gave a quick bounce and then no extra gain by 20 days. That fits RSI(2) snapback holding for about 5 days, and fits RSI(14) mean reversion, which holds for weeks, doing badly here.
The test is one window, with daily decisions and no margin. A 2x fund can lose more than the S&P 500 in a bad year, and 2022 is the only extended bear market in the window.
SPUU against the other leveraged funds
Overnight returns made up 67.53% of SPUU's log return, with 79.53% overnight and 38.24% intraday. The average intraday range was 2% and the average overnight gap was 0.9%. A rule that holds through the close collects both parts, and a rule that trades only inside the session gets about a third. The RSI(2) snapback holds for 4.7 days on average, so most of its trades span at least one overnight gap.
The same snapback rule led the 2x and 3x peers on several of the other leveraged funds. TQQQ held at 25.35% and the RSI(2) snapback made 39.31%. QLD held at 23.55% and it made 26.17%, and SSO held at 21.92% and it made 24.56%. On SPUU it made 24.12% against 22.48% for holding. SOXL behaved differently: the golden cross made 45.93% against 33.32% for holding, and TECL held at 38.16% with its monthly cycle at 35.97%.
SPUU's average daily dollar volume of $2,875,224 is small, and that is the main caveat for the RSI(2) result. The rule made 343 fills over the window. The cost runs charge 5 and 10 basis points on each fill and bring the CAGR to 20.53% and 17.06%. With a median of 233 shares a minute, a position of a few hundred shares would be a large share of a typical minute, and the test does not model that.
Six templates finished with a CAGR between 9.75% and 14.17%, and five of them were the slow trend and momentum rules. Their drawdowns ran from 31.38% to 41.32%, below the fund's 46.08% but still deep. Their yearly results for 2022 were -26.2%, -22.2% and -30.6% for the 200-day filter, golden cross and trailing stop. The only templates that avoided most of the 2022 loss were the weekly 7% target at -4.3% and the momentum breakout at -6.7%.
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 SPUU?
Of the 12 templates tested on SPUU over 2021-01-04 to 2026-10-02, the strongest by CAGR was RSI(2) snapback at 24.1% (max drawdown 31.2%), versus 22.5% 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 SPUU?
2 of 12 templates beat SPUU buy-and-hold (22.5% CAGR) on this window; 12 of 12 had a shallower maximum drawdown than holding (46.1%).
What was the best strategy for SPUU?
[RSI(2) snapback](/learn/strategies/rsi2-dip-snapback/spuu/) had the highest CAGR at 24.12%, with a 31.17% drawdown and a Sharpe of 1.09. Holding returned 22.48% with a 46.08% drawdown. At 10 basis points of slippage the RSI(2) result fell to 17.06%.
Did any strategy beat buying and holding SPUU?
Two of 12 did on CAGR, RSI(2) snapback and the weekly 7% target at 23.28%. All 12 had a shallower maximum drawdown than the 46.08% for holding.
How much does SPUU fall in a bear market?
In 2022 SPUU lost 38.71% for the calendar year and fell 46.69% in the profile data, or 46.08% in the buy-and-hold run, from 2021-12-29 to 2022-10-12. SPY lost 18.16% that year.
Does SPUU return exactly twice the S&P 500?
Over the window SPUU returned 236.37% against 125.29% for SPY. Twice SPY's total return is 250.58%. The realized beta to SPY was 1.99.
Is SPUU liquid enough to trade?
The data shows an average daily dollar volume of $2,875,224 and a median minute volume of 233 shares. That is low, and the cost runs use a flat charge per fill that does not model it.
Which trend strategy worked best on SPUU?
EMA 12/26 returned 19.12% with a 22.88% drawdown and a profit factor of 4.99. The [golden cross](/learn/strategies/golden-cross/spuu/) returned 13.22% on 3 round trips, and [SMA 10/50](/learn/strategies/sma-10-50-trend/spuu/) returned 11.18%.
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.