Learn › Strategies › RSI(2) Dip Snapback › QQQE

RSI(2) Dip Snapback on QQQE

Direxion NASDAQ-100 Equal Weighted Index Shares: the Nasdaq-100 with every stock weighted equally, so less mega-cap concentration. Backtest 2021-01-04 to 2026-10-02, $10,000 starting capital, computed by the same engine that runs live DeployQuant strategies.

Result: RSI(2) snapback on QQQE turned $10,000 into $16,825 (68.3% total, 9.5% CAGR): it trailed buy-and-hold by 0.3% per year, with a maximum drawdown 18.3 points shallower than holding (13.2% vs 31.4%).

Buying QQQE on an extreme two-day RSI reading and selling the bounce produced a CAGR of 9.49% from 2021-01-04 to 2026-10-02, and turned $10,000 into $16,825. Holding QQQE, the Direxion NASDAQ-100 Equal Weighted Index Shares, produced 9.83% and $17,128. The two ended within $303 of each other. What differs is the path: the rule's maximum drawdown was 13.16% against 31.44% for the fund, and its Sharpe ratio was 0.73 against 0.59.

The RSI(2) snapback buys with 98% of the sleeve at the open when the 2-day RSI is below 10 and sells everything when it climbs above 70. It closed 156 round trips from 312 fills, 96 of them winners, a win rate of 62%. The average trade held 5.4 days and was invested on 40.3% of trading days. Winners averaged 1.54% and losers -1.52%, so the edge sits in the win rate, and the profit factor was 1.58.

This is the second-best template on QQQE, behind only the weekly 7% target, and QQQE is 22nd of 59 funds for this rule. The rule beat the fund in 2022, 2024 and 2025, and trailed in 2021, 2023 and 2026. The biggest caveat is cost. The headline run charges nothing per fill, and with 156 round trips the 5 and 10 basis point runs take the CAGR down to 6.62% and 3.81%.

9.5%CAGR
9.8%buy & hold CAGR
−13.2%max drawdown
0.72Sharpe ratio
156round trips
62%win rate
■ RSI(2) snapback   ■ buy & hold, $10,000 invested 2021-01-04

Year by year

YearRSI(2) snapbackbuy & hold
202118.3%19.2%
2022−2.6%−23.9%
20239.0%32.8%
202413.4%6.8%
202515.8%14.3%
20262.0%18.8%

What each year looked like

2021 returned 18.3% against 19.2% for the fund. The test opened badly: the first trade began on 2021-01-25 at $72.40 and sold on 2021-02-03 at $72.16, a 0.33% loss, and January lost 3.81% for the sleeve. March made 11.27%, the best month of the entire test, and May added 4.79%. The year closed 25 round trips, 15 of them winners. November lost 4.43%, which put a dent in an otherwise steady second half.

2022 is where the rule separated from the fund: -2.6% against -23.9%. Thirty round trips closed and 15 were winners, so half of the trades made money, the lowest hit rate of the six years along with 2026. The rule was never far from flat. February made 4.79%, April 2.01% and July 6.23%, while January lost 5.51%, June lost 5.92% and August and September lost 4.55% and 4.82%. October 2022 made 5.93% as the market turned up from its low.

Several of the worst trades came in 2022, which makes sense for a rule that buys drops. The entry on 2022-06-10 at $63.62 sold on 2022-06-16 at $59.60 for a 6.32% loss. The entry on 2022-03-02 at $71.94 lost 4.36% by 2022-03-10, the entry on 2022-09-19 lost 4.05% and the entry on 2022-08-18 lost 3.24%. A fall that kept falling is the case the template's caveat warns about, and 2022 supplied a series of them. Even so, the year ended at -2.6% because the winners, such as the 4.56% from 2022-03-14 to 2022-03-17 and the 3.89% from 2022-10-03 to 2022-10-05, came quickly after the drops.

2023 returned 9% against 32.8%, the widest shortfall of the six years. The fund rose in a long, mostly uninterrupted run. Twenty-six trades closed, 15 won, and the monthly table is a mix of small gains and losses: 3.56% in January, -2.45% in February, 5.15% in May, and losses of 2.36%, 2.24% and 1.19% from August to October. A rule that buys weakness and sells strength is badly placed in a market that rarely gives weakness to buy.

2024 returned 13.4% against 6.8%, one of the three years the rule led. Twenty-eight trades closed and 20 were winners. February made 4.79%, June 3.35% and August 3.35%. The fund itself had a weak year with a deep, sharp August drop and a recovery, which suits a rule that buys the drop and sells the recovery within a week. November and December lost 1.38% and 2.66%.

2025 returned 15.8% against 14.3%. Twenty-five trades closed with 20 winners. The path inside the year was not smooth. February lost 3.05% and March lost 5.26%, including the longest and worst trade of the test: an entry on 2025-02-21 at $96.60 that was sold on 2025-03-18 at $88.25 for an 8.64% loss after 25 days. April then made 8.65% and January had made 5.67%. The best trade of the test came in April: an entry on 2025-04-07 at $75.57 sold on 2025-04-10 at $82.65 for 9.37% in 3 days.

2026 is a partial year, to early October. The rule returned 2% against 18.8%, the widest shortfall in a year that is still running. Twenty-two trades closed and 11 won. The weakest months were March at -2.21% and July at -1.94%, and the best April at 4.35%. The final trade began on 2026-09-25 at $120.80 and sold on 2026-10-02 at $121.67 for 0.72%. No position was open at the end of the data.

Month by month

YearJanFebMarAprMayJunJulAugSepOctNovDec
2021−3.8%−0.5%11.3%1.4%4.8%2.8%0.0%0.2%−1.1%3.9%−4.4%3.5%
2022−5.5%4.8%0.0%2.0%0.3%−5.9%6.2%−4.5%−4.8%5.9%0.7%−0.6%
20233.6%−2.5%2.7%−0.2%5.2%−0.4%3.3%−2.4%−2.2%−1.2%3.5%−0.3%
20242.7%4.8%2.3%0.6%1.9%3.4%−0.2%3.4%−0.8%−1.0%−1.4%−2.7%
20255.7%−3.0%−5.3%8.7%4.0%1.4%−0.5%1.6%2.2%1.8%−0.8%−0.2%
20262.2%−0.5%−2.2%4.3%0.1%−0.2%−1.9%0.5%−1.3%1.3%––

The monthly table

The monthly table has fewer zeros than a trend rule would show. March 2022 shows 0% because a 4.36% loss and a 4.56% gain in that month netted out, and every other month shows a non-zero result. The best month was March 2021 at 11.27% and the next best April 2025 at 8.65%. The worst was June 2022 at -5.92%, with January 2022 at -5.51% and March 2025 at -5.26% next. The fund's own worst month was April 2022 at -10.27% and its best April 2026 at 10.95%. The rule's range of monthly outcomes was much narrower.

Month after month the rule made small gains, which fits an average hold of 5.4 days. The count of months with a gain is large. The larger losses come in clusters, such as the stretch from January to June 2022, with 5.51% lost in January and 5.92% in June, or February and March 2025. Those clusters are the data behind the drawdowns in the next section.

Every trade

RSI(2) snapback on QQQE made 156 closed round trips, an average hold of 5 days, an average winner of 1.54%, an average loser of −1.52%, a profit factor of 1.58, a longest losing streak of 4. It held a position at the close on 40.3% of trading days.

Best 10 round trips

EntryEntry priceExitExit priceReturnDays held
2025-04-07$75.572025-04-10$82.659.4%3
2021-05-19$70.142021-05-21$73.354.6%2
2022-03-14$67.482022-03-17$70.564.6%3
2022-10-03$59.082022-10-05$61.383.9%2
2023-05-25$69.562023-05-30$72.183.8%5
2024-01-17$81.482024-01-22$84.493.7%5
2023-03-02$66.372023-03-06$68.703.5%4
2022-11-29$64.512022-12-01$66.753.5%2
2024-08-05$80.472024-08-09$83.223.4%4
2023-01-19$64.482023-01-24$66.673.4%5

Worst 10 round trips

EntryEntry priceExitExit priceReturnDays held
2025-02-21$96.602025-03-18$88.25−8.6%25
2022-06-10$63.622022-06-16$59.60−6.3%6
2022-03-02$71.942022-03-10$68.80−4.4%8
2022-09-19$62.402022-09-29$59.87−4.0%10
2022-08-18$69.732022-08-25$67.47−3.2%7
2021-11-19$83.872021-12-08$81.23−3.1%19
2022-01-14$77.452022-02-01$75.08−3.1%18
2023-09-19$74.702023-09-26$72.47−3.0%7
2024-12-16$93.312024-12-24$90.54−3.0%8
2023-10-19$73.272023-10-25$71.17−2.9%6

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.

156 short trades

The 156 trades were 96 winners and 60 losers. The average winner returned 1.54% and the average loser -1.52%, and the median trade returned 0.48%. The median hold was 5 days, the longest 25 and the shortest 1. The profit factor was 1.58. The longest winning streak was 15 trades and the longest losing streak 4.

Almost every trade is small, so a few large ones stand out. The best, 9.37%, came in April 2025. After it were 4.58% from 2021-05-19 to 2021-05-21, 4.56% from 2022-03-14, 3.89% from 2022-10-03 and 3.77% from 2023-05-25 to 2023-05-30. The worst were 8.64% in March 2025, 6.32% in June 2022, 4.36% and 4.05% in 2022 and 3.24% in August 2022. The two worst losses were both larger than the best-trade list's second place, which says that the downside tail of this rule is longer than its upside tail.

By exit year the counts were 25 trades and 15 wins in 2021, 30 and 15 in 2022, 26 and 15 in 2023, 28 and 20 in 2024, 25 and 20 in 2025, and 22 and 11 in 2026. A win rate of 62% averages over years that ranged from 11 of 22 to 20 of 28.

Largest drawdowns

PeakLow pointDepthDays to lowRecoveredDays to recover
2022-06-022022-09-26−13.2%1162023-05-26242
2021-11-082022-01-27−12.1%802022-04-1982
2025-02-202025-03-13−11.4%212025-04-0927

Buy-and-hold's deepest drawdown ran from 2021-11-16 to 2022-10-14 and reached −31.4%.

Drawdowns

The deepest drawdown was 13.16%, from a peak on 2022-06-02 to a low on 2022-09-26, 116 days later. It recovered on 2023-05-26, 242 days after the low. The second was 12.06%, from 2021-11-08 to 2022-01-27, recovered by 2022-04-19. The third was 11.37%, over 21 days from 2025-02-20 to 2025-03-13, recovered on 2025-04-09 after 27 days.

The fund's drawdowns were deeper and longer. It fell 31.44% from 2021-11-16 to 2022-10-14 and recovered on 2024-01-19. A second fall of 20.93% ran from 2025-02-18 to 2025-04-08, and a third of 10.15% from 2024-07-16 to 2024-08-05. The rule's three drawdowns all sit in the 11% to 13% range, so no single episode dominated. Its worst, June to September 2022, is a stretch when the fund itself kept falling for months.

The 2025 episode shows how the rule fails. The fund fell for 49 days from 2025-02-18, and the rule bought into the fall on 2025-02-21 and held for 25 days to a loss of 8.64%. The sell condition, RSI(2) above 70, never arrived in a falling market, so the position stayed open. The rule has no stop. Its protection against a long decline is only that RSI(2) tends to rise at some point.

Where the largest losses sat inside the drawdowns

Three of the five worst trades began inside the rule's deepest drawdown, which ran from 2022-06-02 to 2022-09-26. The entry of 2022-06-10 at 63.62 lost 6.32% by 2022-06-16. The entry of 2022-08-18 at 69.73 lost 3.24%. The entry of 2022-09-19 at 62.4 lost 4.05% and closed on 2022-09-29, three days after the equity low. Each bought a drop that went on to fall further, and the exit condition of RSI(2) above 70 did not arrive until the price had already given back the loss.

The 2025 episode ran in the opposite order. The longest and worst trade bought at 96.60 on 2025-02-21 and sold at 88.25 on 2025-03-18. The best trade bought at 75.57 on 2025-04-07, after the fund had fallen further, and sold at 82.65 on 2025-04-10. Its three days cover 2025-04-09, the fund's best session of the window at +10.63%. The fund's drawdown low was 2025-04-08, one day after that entry. The same rule lost 8.64% on the first leg of the decline and made 9.37% on the last leg, and the second trade only existed because the first had already sold.

The pair shows the cost of having no stop. A position bought into a fall waits for RSI(2) to recover. When the recovery comes within days, as in 2022-03-14 to 2022-03-17 for 4.56%, the rule collects it. When the fall continues for weeks, the account carries the open loss, and the three drawdowns of 11% to 13% in the table are what that looked like.

With trading costs

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

Slippage per fillCAGRMax drawdownFinal valueSharpe
None (headline)9.5%−13.2%$16,8250.72
5 basis points6.6%−13.9%$14,4500.53
10 basis points3.8%−14.8%$12,3940.34

Why costs matter more here

With 5 basis points of slippage on each fill the CAGR is 6.62%, the final value $14,450, the maximum drawdown 13.86% and the Sharpe ratio 0.533. At 10 basis points the CAGR is 3.81%, the final value $12,394, the drawdown 14.85% and the Sharpe ratio 0.339. Each step removed a large share of the headline CAGR.

The reason is the trade count and size. The rule made 312 fills, and its average trade gained or lost about 1.5%. The charge applies to the entry and to the exit, so it comes out of every round trip, and the average win is only 1.54%. The headline run is therefore an upper bound.

QQQE trades about $14,399,103 a day on average and its median minute volume is 308 shares. It is a thin fund. A fixed charge understates a wide quote on a small, quiet fund, and the opening minutes are when this rule trades. A rule that trades less often, such as the EMA 12/26 trend on the same fund, pays the same charge far fewer times.

Changing the parameters

VersionCAGRMax drawdownRound tripsWin rateFinal value
Published rules9.5%−13.2%15662%$16,825
RSI(2) < 5 / > 708.9%−13.2%15161%$16,342
RSI(2) < 15 / > 7010.1%−15.6%16563%$17,338
RSI(2) < 10 / > 607.3%−16.4%16462%$14,984
RSI(2) < 10 / > 8011.2%−12.1%14963%$18,403

Changing the thresholds

Four variants moved one threshold at a time. Buying below 5 returned 8.93% with a 13.16% drawdown, 151 trades and 92 wins, and a Sharpe ratio of 0.693. Buying below 15 returned 10.06% with a 15.63% drawdown, 165 trades and 104 wins. Selling above 60 returned 7.3% with a 16.36% drawdown over 164 trades. Selling above 80 returned 11.21% with a 12.05% drawdown over 149 trades and a Sharpe ratio of 0.829.

Two directions are visible. A higher sell level held the position longer and made more, with the shallowest drawdown of any run, 12.05%. A lower sell level at 60 cut the hold, with the lowest return of the five runs. A looser entry level of 15 raised the CAGR to 10.06% but added to the drawdown, and a stricter level of 5 lowered it. These are single runs, and the CAGR range from 7.3% to 11.21% is about as wide as the gap between the rule and the fund.

The published settings, below 10 and above 70, sit in the middle. The results show that the exit level mattered more than the entry level on this fund, and they come from one window and one fund, so they can be sensitive to the sample.

How QQQE behaved

MeasureQQQE
Data in this test2021-01-04 to 2026-10-02 (1444 sessions)
Total return, buy and hold76.6%
Annualized volatility20.2%
Deepest drawdown−32.1% (2021-11-16 to 2022-10-14)
Up days52.0%
Average daily range1.35%
Average overnight gap0.53%
Correlation to SPY0.94
Correlation to QQQ0.95
Correlation to TLT0.12
Sessions above the 200-day average74.5%
Crossings of the 200-day average30
Falls of 10% or more from a 20-day high23

How QQQE behaved

QQQE returned 76.64% with a CAGR of 10.42% and annualized volatility of 20.24%. Its deepest drawdown was 32.14% from 2021-11-16 to 2022-10-14, and the longest was 544 sessions. Up days were 52.04% of sessions, with an average gain of 0.94% and an average loss of 0.93%. The average range was 1.35% and the overnight gap 0.53%.

The overnight gap carried 148.4% of the log return and the trading day was negative at -48.4%. In plain terms, the fund's gain came from overnight moves while its intraday moves lost money in total. A rule that buys at the open and sells at a later open takes both parts of the day on trades that last 5 days.

QQQE had 166 sessions with RSI(2) below 10. After those sessions the median 5-day return was 0.74% against 0.41% for all days, and the median 20-day return 2.23% against 1.22%, from 166 and 162 observations. The snapback signal therefore preceded a better than average short-term return on this fund, though by under a point over 5 days. That small edge is what the rule collects 156 times. The fund's RSI(14) fell below 30 on 18 sessions with a median 5-day return of 3.82%.

The fund closed above its 200-day average on 74.54% of sessions and crossed it 30 times. Correlation to QQQ was 0.95 and to SPY 0.94, with a beta to SPY of 1.15. The best days were 10.63% on 2025-04-09 and 7.04% on 2022-11-10. The worst were -6.23% on 2025-04-04 and -5.03% on 2025-04-03. Lag-1 autocorrelation was -0.03. The data counts 23 events of a fall of 10% or more from a 20-day high, with 56 days spent in them.

Calendar months, longer RSI readings and the rule

QQQE's average return by calendar month was 4.2% in November, 3.37% in May, 2.18% in July and 2.14% in June, against -2.45% in September and -0.43% in April. Each month has 5 or 6 observations. The rule's own May was positive in all six years: 4.79% in 2021, 0.25% in 2022, 5.15% in 2023, 1.91% in 2024, 4% in 2025 and 0.05% in 2026. September was negative for the rule in 2021, 2022, 2023, 2024 and 2026, with 2022 the worst at -4.82%, and positive only in 2025 at 2.17%. November was mixed: -4.43% in 2021, 0.66% in 2022, 3.48% in 2023 and -1.38% in 2024. The fund's strong November average did not carry over to the rule, which sells strength and buys weakness.

The slower oscillator behaved the same way as the two-day one. RSI(14) closed below 30 on 18 sessions and above 70 on 74. After the oversold readings the median 5-day return was 3.82% and the median 20-day return was 3.53%, against baselines of 0.41% and 1.22%. That is a larger edge than the RSI(2) readings gave, at 0.74% and 2.23%, but 18 observations is a small sample, and the rule never trades on RSI(14).

Day-of-week averages were close to zero. Monday was the highest at 0.14%, Tuesday was 0%, and Wednesday and Friday were 0.04%. The fund's beta was 1.15 to SPY and 0.86 to QQQ. Equal weighting lowered the exposure to the largest Nasdaq stocks, and the beta to QQQ below one shows it moved less than the cap-weighted fund on the same days. QQQE had 166 sessions with RSI(2) below 10 and gave the rule 156 round trips, while QQQ and QQQM each gave it 164 round trips at higher CAGRs of 14.17% and 15.56%.

The rules

Buy extreme 2-day RSI washouts under 10, exit as soon as RSI(2) recovers above 70.

  1. WHEN the market opens · IF not invested AND RSI(2) < 10 · THEN buy with 98% of the sleeve
  2. WHEN the market opens · IF invested AND RSI(2) > 70 · THEN sell the whole position

A short-horizon mean-reversion template popularized by Larry Connors' RSI-2 research. A 2-period RSI under 10 flags a sharp multi-day selloff. In assets with a persistent upward drift, those selloffs have tended to snap back within days. Trades are frequent and short. This is the highest-turnover template in the library.

Good for: liquid index ETFs with strong long-term drift; turnover is high so per-trade edges are small.
Watch out: high trade counts make results sensitive to execution quality; a crash that keeps crashing will hand this template several losing entries in a row.

The rule on this fund and on the other broad funds

Equal weighting spreads the index across all its stocks, so QQQE has less mega-cap concentration than the cap-weighted Nasdaq funds. For this rule the broad group ranks as follows: VOOG 15.89%, QQQM 15.56%, QQQ 14.17%, VOO 14.08%, IOO 14.04%, SPY 13.17% and VV 12.55%. IWM made 11.38%, QQQE 9.49%, VOOV 8.58%, VTV 7.29% and EEM 3.87%. QQQE is ninth of the twelve, and the median across the 59 funds was 4.74% and across the broad group 13.17%.

On QQQE itself the weekly 7% target was ahead at 11.21% with a 19.33% drawdown. Behind this rule came the monthly cycle at 7.95%, RSI mean reversion at 7.22% and the golden cross at 6.98%. Trend and momentum templates were far behind: the EMA 12/26 trend 1.87%, the SMA 10/50 trend 1.16%, and the 3-month momentum -2.86%. In this window a rule that bought pullbacks beat the rules that followed trends on this fund.

The test stops at 5.74 years, one fund, daily decisions and no margin. The headline run has no costs, which the cost runs show matters here. Prices are adjusted for splits and dividends.

Run RSI(2) snapback on QQQE yourself, free →

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

Did RSI(2) snapback beat buy-and-hold on QQQE?

Over 2021-01-04 to 2026-10-02, RSI(2) snapback on QQQE returned 9.5% annualized vs 9.8% for buy-and-hold: it trailed buy-and-hold by 0.3% per year, with a maximum drawdown 18.3 points shallower than holding (13.2% vs 31.4%).

How many trades did it make?

156 completed round trips over 5.7 years (312 fills), with 62% of round trips closing profitably.

How often does RSI(2) trade?

Far more than RSI(14), with dozens of round trips per year on a volatile ETF. The backtest table on each page shows the exact count over the test window.

Is RSI(2) too fast for daily bars?

It is designed for daily bars. The 2-day window catches short, sharp washouts rather than long regimes.

Did RSI(2) snapback beat buy-and-hold on QQQE?

Slightly behind on return and well ahead on risk. It returned 9.49% a year against 9.83%, ending at $16,825 against $17,128. Its maximum drawdown was 13.16% against 31.44%, and its Sharpe ratio 0.73 against 0.59.

How many trades did it make on QQQE?

156 closed round trips from 312 fills, with 96 winners, a win rate of 62%. The average hold was 5.4 days and the longest 25 days. No position was open at the end of the data.

How do trading costs affect it?

A great deal. With 5 basis points of slippage per fill the CAGR was 6.62%, and with 10 basis points 3.81%. The headline run, with no costs, was 9.49%. The rule trades often and each trade gains or loses about 1.5% on average.

What was the worst trade?

An entry on 2025-02-21 at $96.60 sold on 2025-03-18 at $88.25, a loss of 8.64% after 25 days. The rule has no stop, so it held while the fund fell.

Which RSI thresholds worked best on QQQE?

Selling above 80 returned 11.21% with a 12.05% drawdown, against 9.49% for the published rule that sells above 70. Selling above 60 returned 7.3%. Buying below 15 returned 10.06% and below 5 returned 8.93%. Each is a single run on one fund.

Does the rule work better on other funds?

On broad index funds the median CAGR was 13.17%. VOOG returned 15.89%, QQQM 15.56% and SPY 13.17%. QQQE at 9.49% was ninth of 12 in that group.

Related

RSI(2) Dip Snapback on all 59 ETFsfull results table All strategies on QQQE12 templates compared RSI(14) Mean Reversion on QQQEsame ETF, different rulesGolden Cross (SMA 50/200) on QQQEsame ETF, different rules

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.