QQQM trading strategies, backtested

Invesco NASDAQ 100 ETF: a lower-cost share class of the Nasdaq-100 trade, built for holding. Every DeployQuant template run on QQQM over 5.7 years of minute data, same engine, same window, sorted by return.

QQQM holds the same Nasdaq-100 basket as QQQ, in a share class with a lower expense ratio. In this test the two behave as one fund. The correlation between them is 1.00, and the 200-day regime filter returned 13.0% a year on QQQM and 12.9% on QQQ, with 10 round trips on each. We ran all 12 DeployQuant templates on QQQM from 2021-01-04 to 2026-10-02, with $10,000 per run, minute-bar fills, no margin, and no fees or slippage in the headline numbers.

Buy-and-hold returned 16.7% a year with a maximum drawdown of 34.4% and a final value of $24,315. None of the 12 templates beat it on CAGR. The closest was the weekly 7% target at 16.6%, a gap of 0.1 percentage points, and it did so with a 19.2% maximum drawdown. Eleven of the 12 had a shallower drawdown than holding. The exception was the monthly cycle, whose maximum drawdown of 36.6% was deeper than the 34.4% for holding.

The window matters for how to read this. QQQM gained 29.2% in 2021, lost 32.5% in 2022, gained 54.9% in 2023, then added 25.7%, 20.8%, and 22.5% (part year) in the following years. One deep bear year sits inside five strong ones. A rule that sits out part of the market has one year to earn its keep and five to give back return.

The QQQ page covers the same index over the same window, and its numbers can be read next to these.

Ranking the 12 templates on QQQM

The ranking runs from 16.6% down to 5.3%. The top five all finish near holding: weekly 7% target at 16.6%, RSI(2) snapback at 15.6%, monthly cycle at 14.9%, golden cross at 13.8%, and the 200-day filter at 13.0%. The bottom four, from the 20-day momentum template at 7.9% to the dip buyer at 5.3%, left a large share of the Nasdaq-100's gain on the table.

The weekly 7% target reached its result with 108 round trips, a 31% win rate, and an average win of 7.07% against an average loss of 1.79%. It was invested 83.7% of the time. Its drawdown of 19.2% started at the 2021-12-27 peak and bottomed on 2022-05-24, well before the fund's own low on 2022-11-03. The Thursday exit rule for losing trades cut the weekly positions that were underwater, and that is the mechanism behind the shallower fall. The same rule cost it in the rebounds: it trailed holding by 11.6 points in 2023 and by 12.3 points in 2021. Its two years ahead of holding were 2022, by 16.4 points, and 2024, by 4.2 points.

RSI(2) snapback had the best Sharpe ratio, 1.06, and the shallowest drawdown, 14.3%, with only 36.0% of time in the market. It made 164 round trips and won 70% of them. Its year of 2023 stands out as the one where it trailed most, at 12.8% against 54.9% for the fund, a gap of 40.6 points. A rule that buys short-term weakness and sells after a bounce is flat during a strong straight-line rally. The RSI(2) page for QQQM lists every trade.

Golden cross made 2 round trips in the whole window. The first was a loss of 6.28% over 135 days, entered on 2021-10-19 and closed on 2022-03-03. The second ran 764 days from 2023-03-13 to 2025-04-15 for a gain of 62.2%, and a third position opened on 2025-06-24 was still held at the end with a gain of 40.56%. Its CAGR of 13.8% rests almost entirely on that one long hold. The cost runs barely moved it because there were so few trades. Two completed round trips cannot say whether the rule has an edge on this fund. The golden cross page shows the dates.

The 200-day filter and the trailing-stop version of it share the same entry rule. The filter returned 13.0% with a 19.8% drawdown. The version with a 15% trailing stop returned 11.4% with a 26.8% drawdown. The trailing stop added a second way to exit, and in 2022 it exited late: its two 2022 trades lost 7.41% and 14.23%, and its 2022 return was negative 24.7% against negative 16.2% for the plain filter.

The three slower trend rules, SMA 10/50 at 9.1%, 3-month momentum at 8.0%, and EMA 12/26 at 7.8%, show the cost of a faster signal on a fund that did not whipsaw much. SMA 10/50 made 18 round trips and EMA 12/26 made 22, with win rates of 39% and 41%. Both had average wins of 13.52% and 11.29% against average losses near 3 to 4%, which gives a positive profit factor of 2.50 and 1.92, but not enough to catch the long uptrends. The comparison pages cover these pairs, for example SMA 10/50 against the 200-day filter.

The dip buyer made 4 round trips and won all 4, each at about 8%, and still finished last at 5.3%. It was in the market 30.5% of the time. The 2022-01-21 entry ran 543 days before reaching its target on 2023-07-18, and the 2022 drawdown of 28.6% came from holding through that stretch. A perfect win rate on 4 trades says little about the rule.

Where QQQM and QQQ differ in the results

The two funds hold one basket, so the differences between their pages are small and worth listing. The weekly target made 16.6% on QQQM, a hair above its figure on QQQ. The RSI(2) snapback made 15.6% here, with a 2023 gap to holding of negative 40.6 points. The 200-day filter's worst drawdown was 19.82%. The same rules lead and lag in the same order on both pages, and the gaps between the two funds are a few hundredths of a point on most lines.

The share prices differ by a wide margin, and that shows up in the trade lists. The weekly target and the 3-month momentum switch both entered on 2026-09-28 at an adjusted 304.83 on QQQM, a share price a fraction of QQQ's on the same date, and both are up 1.24% here. The trailing stop strategy has been in since 2025-05-13 at 207.86 and is up 48.47%. The golden cross has held since 2025-06-24 at 219.57 and is up 40.56%, and the 200-day filter since 2026-04-09 at 248.89, up 24%. Return percentages are comparable across the two funds. Dollar prices are not.

Month by month, April 2026 was the best month for holding at 15.44% and for the golden cross at 15.3%, the trailing stop strategy at 15.25%, the monthly cycle at 13.82% and the weekly target at 13.29%. April 2022 was the worst month for holding at negative 13.16% and for the dip buyer at negative 13.03%. January 2022 was the worst month for the monthly cycle at negative 11.34%, the trailing stop strategy at negative 12.72%, the 200-day filter at negative 9.36% and the golden cross at negative 8.39%. In May 2026 the SMA 10/50 trend, the 3-month momentum switch, the EMA 12/26 trend and the momentum breakout each had their best month, between 10.21% and 10.3%.

The longest streaks were 9 wins for the RSI(2) snapback, 7 losses for the weekly target and the 200-day filter, 6 wins for the monthly cycle and 4 wins for the dip buyer, which never lost. Holding's second drawdown, 22.43% from 2025-02-19 to 2025-04-08, was recovered on 2025-06-24. The golden cross had a drawdown of 22.4% over the same fall and did not recover until 2025-10-28, four months after the holder.

Overnight gaps account for 78.83% of QQQM's log return and intraday trading for 21.17%, so a rule that sits in cash overnight gives up most of the fund's return even on days when it is right.

How each strategy traded QQQM

Time in market, hold lengths, and the cost runs

The trading table shows two things that the CAGR column hides: how long each template stayed invested, and how much trading cost it can absorb.

The monthly cycle was invested 95.2% of the time, with an average hold of 28 days. Its worst trade was a loss of 11.61% in January 2022 and its best was a gain of 14.23% in April 2026. That is a rule with almost the same exposure as holding and almost the same drawdown, and it lost 1.79 percentage points of CAGR against holding for no reduction in risk. Its worst year was 2022 at negative 34.4%, slightly worse than the fund's negative 31.8% on the buy-and-hold run.

Golden cross was in the market 65.1% of the time with an average hold of 450 days, and its profit factor of 9.34 comes from two trades. The 200-day filter had 64.8% exposure, an average hold of 118 days, a best trade of 68.0%, and a worst trade of negative 3.9%. Its profit factor was 4.91 with a win rate of only 20%. Eight of its 10 round trips lost money, and the losses were small: the average loss was 2.08% and the average win was 42.29%. This is a typical shape for a trend filter. It pays a series of small losses in choppy stretches, as in the clusters of one-day trades in February to April 2022, and recovers them with two long holds of 724 and 314 days.

The weekly 7% target was invested 83.7% of the time. Its median trade lost 0.77% and its median hold was 9 days. The profit factor of 1.79 comes from a few winners reaching 7% against many small losses.

On the cost runs, the templates divide into three groups.

The low-turnover rules barely moved. Golden cross went from 13.8% to 13.8% at 10 basis points of slippage. The dip buyer stayed at 5.4%. The trailing-stop trend rule went from 11.4% to 11.3%. The 200-day filter went from 13.0% to 12.7%.

The weekly 7% target dropped from 16.6% to 14.1%, with its drawdown widening from 19.2% to 21.8%. At 10 basis points it fell behind golden cross and the 200-day filter in CAGR. The monthly cycle dropped from 14.9% to 12.3%.

The short-term rules took the largest hit. RSI(2) snapback went from 15.6% to 12.4% at 5 basis points and to 9.2% at 10, with its Sharpe falling from 1.06 to 0.68. It made 328 fills, and each one pays the spread. The 3-month momentum template, SMA 10/50, and EMA 12/26 all landed between 7.1% and 8.4% at 10 basis points.

QQQM is a liquid fund, with an average daily dollar volume of $366,984,628 and a median minute volume of 2,083 shares. The 5 and 10 basis point runs are a rough guide to cost, not a measured fill quality. The headline numbers have no costs at all, so the cost runs are the more realistic comparison for the high-turnover templates.

How QQQM behaved

Calendar years

Biggest single days

Average return by calendar month

Most and least correlated funds

What the fund's own behaviour explains

QQQM's annualized volatility over the window was 22.3%. Its deepest drawdown was 35.1%, from the peak on 2021-12-27 to the trough on 2022-11-03, and it took until 2023-12-12 to recover, 492 sessions from peak to recovery. That single episode explains why the template results look so similar in 2022. Every rule lost money that year except where it was out of the market, and the best any of them did was RSI(2) snapback at negative 5.5%.

The fund closed above its 200-day average on 75.2% of sessions and crossed that line 20 times. A fund that spends three quarters of its time on one side of the line gives a trend filter few chances to be right by exiting. The filter made 10 round trips against 20 crossings, and the trade list shows all 5 of its 2022 round trips lost money. The one-year gaps to holding were 2022 at plus 15.6 points and 2023 at negative 18.6 points: the filter was ahead while the fund fell and behind when it recovered, because the average took time to turn and the filter re-entered after the first leg of the rebound.

The fund moved up on 54.9% of days, with an average up day of 1.01% and an average down day of negative 1.06%. The first-order autocorrelation of daily returns was negative 0.03, which is close to zero. There was no day-to-day persistence for a short-horizon rule to use. RSI(2) snapback and the RSI mean reversion template are the two that rely on a pullback being followed by a bounce, so the data on oversold readings applies directly to them.

There were 148 sessions with RSI(2) below 10. The median 5-day forward return after those sessions was 0.77% against a baseline of 0.56%, and the median 20-day forward return was 2.18% against 1.73%. Both are above the baseline, though by small margins. For RSI(14) below 30 there were only 16 sessions, with a median 5-day forward return of 3.21% and a median 20-day forward return of 1.54% against a baseline of 1.73%. The 20-day figure is below the baseline, which fits the RSI(14) template's result: 6.5% CAGR, in the market 28.9% of the time, with a 73% win rate and a long wait between signals. The sample of 16 oversold sessions is small.

The fund also had 21 falls of 10% or more from a 20-day high, spread over 81 days. The dip buyer enters on that condition and made only 4 round trips, since it waits for the drawdown condition to appear and then sits in the trade until it reaches 8%.

Overnight gaps account for most of the fund's return in the data: the average overnight gap was 0.60%, the average intraday range was 1.55%, and the overnight share of the log return was 78.83% against 21.17% for the intraday session. A rule that exits at the open and re-enters at the open captures the overnight part only while invested. That is relevant to the weekly template, which buys at the first open of the week, and to every template whose orders fill at the next open after a signal.

The calendar data covers 5 or 6 observations per month, so it describes this window and nothing wider. May averaged 5.2% and November 4.4%, while February averaged negative 0.7% and September negative 1.7%. The biggest single days were 2025-04-09 at 11.7% and 2022-11-10 at 7.4%, and the worst were 2025-04-04 at negative 6.1% and 2022-09-13 at negative 5.5%. The April 2025 pair of a large fall followed within days by the fund's largest rise shows up in the RSI(2) template, which recorded its best trade of 11.42% from 2025-04-07 to 2025-04-10.

QQQM's beta to SPY was 1.28 and its correlation to SPY was 0.94. Its correlation to TLT was 0.09. Among the other funds in the test, the closest were QQQ, TQQQ, and QLD at 1.00, and VOOG at 0.98. Other broad funds in the group are SPY, VOO, and QQQE. On QQQE, the equal-weighted Nasdaq-100 fund, buy-and-hold returned 9.83% and the best template, the weekly 7% target, returned 11.21%, so the rules did better there than on QQQM.

Over this window, the data says a Nasdaq-100 fund with a long strong uptrend gave simple exit rules little to improve on. The backtest is one window of 5.74 years, and none of the numbers are a forecast.

Year by year against holding

The yearly gaps to holding line up with the QQQ results. In 2022, when QQQM lost 32.47%, the RSI(2) snapback beat holding by 26.3 points, the golden cross by 20.1, the 200-day filter by 15.6 and the weekly target by 16.4. In 2023, when the fund gained 54.92%, the same rules trailed by 40.6, 11, 18.6 and 11.6 points. The monthly cycle stayed within a few points of holding every year and beat it in 2024 and 2025 by 3.2 and 1.7 points. The momentum breakout trailed by 27.7 points in 2023 and 25.4 in 2026 and beat holding in 2022 and 2025. A rule that exits in a bear year and re-enters in the recovery is behind in the recovery year, and the table of gaps is the clearest record of that on this page.

Other broad index 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.