DeployQuantGolden Cross (SMA 50/200) on QQQ
Invesco QQQ Trust: tracks the Nasdaq-100, a tech-heavy growth index. Backtest 2021-01-04 to 2026-10-02, $10,000 starting capital, computed by the same engine that runs live DeployQuant strategies.
Golden cross on QQQ was a two-trade test. The rule buys when the 50-day average is above the 200-day average and sells when the 50-day falls below it. From 2021-01-04 to 2026-10-02 it closed 2 round trips, held a third position open at the end, and turned $10,000 into $20,793. Buy-and-hold turned the same amount into $24,265. The CAGR was 13.6% against 16.7%, and the maximum drawdown was 22.2% against 34.2%.
Among the 12 templates on QQQ, golden cross ranked fourth by CAGR. Across the 59 funds in the test, QQQ was its sixth best result. The template's median CAGR across all 59 funds was 2.05%, and its median across the broad index funds was 8.71%, so QQQ is well above both. That position says more about the fund than about the rule. QQQ rose strongly in the window, and a rule that held it for 65.1% of trading days kept most of that gain.
The whole result rests on one trade. The second round trip held for 764 days and gained 61.96%. The first lost 6.27%. The position that was open at the end had gained 40.48%. A profit factor of 9.10 sounds strong, but it comes from dividing one large win by one small loss.
The test is a single window of 5.74 years. A rule that trades about once every two years cannot be judged on two closed trades.
Year by year
| Year | golden cross | buy & hold |
|---|---|---|
| 2021 | 6.4% | 28.4% |
| 2022 | −11.8% | −31.7% |
| 2023 | 41.7% | 53.0% |
| 2024 | 24.8% | 25.1% |
| 2025 | 2.8% | 20.4% |
| 2026 | 21.7% | 22.1% |
Where the gap to buy-and-hold opened
The year table shows a pattern that repeats across trend rules. In 2021 golden cross returned 6.4% against 28.4% for holding. The rule did not buy until 2021-10-19, so it missed the first nine months of the year. The months from January to September 2021 all read 0.0%, and the last three months of the year, October at 3.3%, November at 1.9%, and December at 1.1%, are all the rule earned. The gap that year was 22.0 points.
In 2022 the rule did its job. It returned negative 11.8% against negative 31.7% for holding, a gain of 19.9 points. It was the only year of the six in which golden cross beat holding. The reason is in the monthly table: the position was open through January and February 2022, which cost 8.4% and 4.3%, and the sell order went in on 2022-03-03. From April to December the rule held cash. QQQ's deepest drawdown ran from 2021-11-19 to 2022-11-03, so the rule sat out most of it, but it still gave back the first leg.
In 2023 the rule returned 41.7% against 53.0%. It was in cash for January and February and bought on 2023-03-13. March returned 11.4%, and by that point holding had already made part of the year's gain. The gap that year was 11.3 points.
In 2024 the result was 24.8% against 25.1%, a gap of 0.3 points. The rule was invested the whole year, so it tracked the fund closely. The 98% position size is one reason for the small gap.
In 2025 the rule returned 2.8% against 20.4%, a gap of 17.6 points. That is the second largest gap in the table and it came from a single event: the position was held through the drop from 2025-02-19 to 2025-04-08, sold on 2025-04-15 after the low, and re-bought on 2025-06-24. February, March, and April of 2025 returned negative 2.6%, negative 7.4%, and negative 2.2%. May shows 0.0%, a month in cash, and the rule returned to the market in June at 2.7%. It took the loss of the decline and then missed the first part of the recovery.
In 2026 the rule returned 21.7% against 22.1%, a gap of 0.4 points. It held a position the whole year, from the June 2025 entry. Three of the six years trailed holding by more than 10 points (2021, 2023, and 2025), and each came from time spent in cash while the fund was rising.
Month by month
| Year | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2021 | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 3.3% | 1.9% | 1.1% |
| 2022 | −8.4% | −4.3% | 0.7% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| 2023 | 0.0% | 0.0% | 11.4% | 0.5% | 7.5% | 6.2% | 3.8% | −1.5% | −4.9% | −2.0% | 10.4% | 5.4% |
| 2024 | 1.8% | 5.2% | 1.2% | −4.3% | 6.2% | 6.0% | −1.6% | 1.1% | 2.5% | −0.8% | 5.2% | 0.5% |
| 2025 | 2.1% | −2.6% | −7.4% | −2.2% | 0.0% | 2.7% | 2.3% | 0.9% | 5.2% | 4.7% | −1.5% | −0.6% |
| 2026 | 1.2% | −2.2% | −4.7% | 15.1% | 10.3% | −0.2% | −6.3% | 4.1% | 3.2% | 1.3% | – | – |
The months that mattered
A strategy with two trades has a monthly table made mostly of zeros and long runs of fund returns. A handful of months decided the outcome.
The best month was April 2026 at 15.1%, which compares with 15.4% for buy-and-hold in the same month, so the rule had the full move. The worst was January 2022 at negative 8.4%. Buy-and-hold's worst month was April 2022 at negative 13.1%, a month in which golden cross had already moved to cash. That is the clearest single comparison in the table: the exit on 2022-03-03 came before the fund's worst month, and the filter avoided it.
The zero stretches show what the rule did not do. The months from April 2022 to February 2023 all read 0.0%. That is 11 months in cash while QQQ went through its drawdown low on 2022-11-03 and started to recover. The rule re-entered on 2023-03-13, about four months after the low, and March 2023 returned 11.4%. May, June, and July 2023 returned 7.5%, 6.2%, and 3.8%, and November 2023 returned 10.4%. A lag of that length is the cost of waiting for the 50-day average to rise above the 200-day.
The losing months after re-entry are scattered. August, September, and October 2023 returned negative 1.5%, negative 4.9%, and negative 2.0%. April 2024 returned negative 4.3%. The period from February to April 2025 returned negative 2.6%, negative 7.4%, and negative 2.2%. March 2026 returned negative 4.7% and July 2026 negative 6.3%. In none of those months did the rule step aside, because a crossover rule does not react to a one-month fall. It reacts when the 50-day average has fallen far enough to cross.
The month of May 2025 is the one month of cash in the rule's later life. It came after the exit on 2025-04-15 and before the re-entry on 2025-06-24. The fund had already recovered most of its drawdown by then. Holding recovered from the 2025 low by 2025-06-24, according to the drawdown table, so the rule was out of the market during the sharpest part of the rebound.
Every trade
golden cross on QQQ made 2 closed round trips and one position still open at the end of the test, an average hold of 450 days, an average winner of 61.96%, an average loser of −6.27%, a profit factor of 9.10, a longest losing streak of 1. It held a position at the close on 65.1% of trading days.
| Entry | Entry price | Exit | Exit price | Return | Days held |
|---|---|---|---|---|---|
| 2021-10-19 | $361.85 | 2022-03-03 | $339.16 | −6.3% | 135 |
| 2023-03-13 | $281.15 | 2025-04-15 | $455.35 | 62.0% | 764 |
| 2025-06-24 | $533.57 | open | – | 40.5% | – |
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 three trades, one by one
The first trade was entered on 2021-10-19 and closed on 2022-03-03 for a loss of 6.27% after 135 days. QQQ had peaked on 2021-11-19, a month after the entry. The 50-day average stayed above the 200-day for nearly four months after the peak, since both averages lag, and by the time the sell signal came the position had fallen from $361.85 to $339.16. Those are adjusted prices, below the quotes printed at the time. This is the typical failure of a crossover on a sharp fall: the exit comes after most of the early damage.
The second trade was entered on 2023-03-13 at $281.15 and closed on 2025-04-15 at $455.35 for a gain of 61.96%, after 764 days. It is the only complete trade that made money, and it is the one that carries the template's CAGR. It was open through the 2024 drawdown of 13.2% and the 2025 drawdown of 22.2%, and the exit came on 2025-04-15, a week after the low of 2025-04-08. The rule gave back part of the gain and sold after the low.
The third trade was entered on 2025-06-24 at $533.57 and was still open at the end of the data, up 40.48%. It is marked at the last close and has not been tested by a sell signal. If the position closes later, the result will be different from the number shown.
The summary statistics follow from these three numbers. The average winner was 61.96% and the average loser was 6.27%. The average hold was 450 days, and the win rate was 50% because one of two closed trades won. The longest losing streak was 1. The rule held a position at the close on 65.1% of trading days, which is the exposure figure.
There is a useful comparison with the QQQM page for golden cross. QQQM tracks the same index, and the rule made 2 round trips there as well. The two funds returned 13.8% and 13.6%, so the result tracks the index and not the fund wrapper. The RSI(2) snapback page for QQQ shows the opposite end of the trade-count range for the same fund.
Largest drawdowns
| Peak | Low point | Depth | Days to low | Recovered | Days to recover |
|---|---|---|---|---|---|
| 2025-02-19 | 2025-04-08 | −22.2% | 48 | 2025-10-28 | 203 |
| 2021-11-19 | 2022-02-23 | −18.0% | 96 | 2023-05-18 | 449 |
| 2024-07-10 | 2024-08-07 | −13.2% | 28 | 2024-11-06 | 91 |
Buy-and-hold's deepest drawdown ran from 2021-11-19 to 2022-11-03 and reached −34.2%.
Three drawdowns, and why the deepest came last
The deepest drawdown for the rule was 22.2%, from 2025-02-19 to 2025-04-08, and it took 203 days to recover, ending on 2025-10-28. That is deeper than the 18.0% drawdown in 2021 and 2022, and it happened after the rule had made its large gain. The position was open through the whole decline. The 50-day average had not crossed below the 200-day, and it did not cross until after the low.
The second drawdown was 18.0%, from 2021-11-19 to 2022-02-23. The low point came 96 days after the peak, and the recovery to the old peak took until 2023-05-18, 449 days after the low. The rule exited on 2022-03-03, just after the low, which means the position took the decline and avoided the part of the fall that came afterward. Buy-and-hold's low came on 2022-11-03 at negative 34.2%.
The third was 13.2%, from 2024-07-10 to 2024-08-07, recovered by 2024-11-06 after 91 days. The 28 days to the low and 91 days to recover match what holding went through in the same period, with a depth of 13.31% on the holding side.
Buy-and-hold's three largest drawdowns were 34.2%, 22.4%, and 13.3%. The rule's were 22.2%, 18.0%, and 13.2%. The gain is in the first: the rule's 2022 exit left the biggest drawdown shallower than holding. The other two are almost the same size as holding, since the rule was invested in both. This is how a slow crossover behaves. It protects against long declines that take months to develop and gives no protection against short ones that reverse in weeks.
The Sharpe ratio of 0.88 against 0.83 for holding reflects that. Return was lower, drawdown was shallower, and risk-adjusted return came out slightly ahead. With two closed trades that margin is within the noise of the sample.
With trading costs
The headline run fills at the bar price. These runs charge slippage on every fill.
| Slippage per fill | CAGR | Max drawdown | Final value | Sharpe |
|---|---|---|---|---|
| None (headline) | 13.6% | −22.2% | $20,793 | 0.88 |
| 5 basis points | 13.6% | −22.3% | $20,765 | 0.88 |
| 10 basis points | 13.5% | −22.3% | $20,736 | 0.88 |
Why slippage barely registers
The cost runs charge slippage on every fill. Golden cross made 5 fills in the window, so a flat charge lands on very few orders. At 5 basis points the CAGR was 13.57% and at 10 basis points 13.54%, against 13.6% in the headline. The final value went from $20,793 to $20,765 to $20,736. The maximum drawdown moved from 22.23% to 22.25% to 22.27%, and the Sharpe ratio stayed at 0.88.
That result does not mean costs are irrelevant to the strategy. It means that the cost of a trade that happens once every two years is negligible, and a rule that trades hundreds of times in the same window can lose much of its return to the same charge. The RSI(2) snapback template is the useful contrast. The headline numbers have no costs, and the cost runs here are flat offsets that do not model the spread on the specific days of the fills. QQQ trades an average of $18,772,024,579 a day, with a median minute volume of 77,976 shares.
Changing the parameters
| Version | CAGR | Max drawdown | Round trips | Win rate | Final value |
|---|---|---|---|---|---|
| Published rules | 13.6% | −22.2% | 2 | 50% | $20,793 |
| SMA 40/200 | 9.8% | −23.0% | 2 | 50% | $17,136 |
| SMA 50/150 | 8.3% | −21.0% | 3 | 67% | $15,804 |
| SMA 60/250 | 11.3% | −22.1% | 2 | 50% | $18,473 |
Moving the averages: three variants
The published rule uses a 50-day average over a 200-day average. We ran three alternatives: 40/200, 50/150, and 60/250.
The published rule returned 13.6% with a 22.2% drawdown and 2 round trips. The 60/250 variant returned 11.3% with a 22.1% drawdown and 2 round trips. The 40/200 variant returned 9.8% with a 23.0% drawdown, also 2 round trips. The 50/150 variant returned 8.3% with a 21.0% drawdown and 3 round trips, of which 2 won. Final values were $20,793, $18,473, $17,136, and $15,804.
The published setting came first on both CAGR and final value. That reads as a good result for 50/200 on this fund and window, and it also reads as sensitivity. Moving the fast average from 50 to 40 days lowered the CAGR even though the trade count did not change. With only two trades, the timing of one entry or exit is a large part of the result. The 50/150 version added a third round trip.
The variants share the same property as the main run: a trade count of 2 or 3. We do not read the ranking as evidence that 50/200 is the best choice. The most that can be said is that four reasonable settings gave results between 8.3% and 13.6% on one fund, and the spread comes from where a handful of crossings landed. For the same rule across all 59 funds, see the golden cross hub page.
How QQQ behaved
| Measure | QQQ |
|---|---|
| Data in this test | 2021-01-04 to 2026-10-02 (1444 sessions) |
| Total return, buy and hold | 151.1% |
| Annualized volatility | 22.4% |
| Deepest drawdown | −35.0% (2021-11-19 to 2022-11-03) |
| Up days | 54.8% |
| Average daily range | 1.58% |
| Average overnight gap | 0.59% |
| Correlation to SPY | 0.94 |
| Correlation to TLT | 0.09 |
| Sessions above the 200-day average | 75.2% |
| Crossings of the 200-day average | 20 |
| Falls of 10% or more from a 20-day high | 21 |
What QQQ's price history did to a crossover rule
QQQ's annualized volatility was 22.4% and its total return in the window was 151.1%. It was above its 200-day average on 75.2% of sessions and crossed the line 20 times, so the fund spent most of the window on the side a trend rule wants. That is the favourable case for golden cross, and the favourable case still lost 3.1 percentage points a year to holding. The cost is the lag: each of the two exits came after most of a decline, and the entry after the low came about four months later.
The fund's calendar-year returns were 29.2% in 2021, negative 32.4% in 2022, 54.8% in 2023, 25.6% in 2024, 20.8% in 2025, and 22.4% in 2026 to date. One bad year and five good ones is a poor mix for a rule that exits slowly and re-enters slowly. The rule needs a long, deep decline to pay for the months it spends out of the fund. QQQ's drawdown lasted 516 sessions, and the exit on 2022-03-03 saved the rule from most of the fall, but the 2025 drop was too short to trigger an exit in time.
The fund had 21 falls of 10% or more from a 20-day high, spread over 82 days. Most of them reversed quickly, which a 200-day average does not react to. The best days were 2025-04-09 at 11.75% and 2022-11-10 at 7.37%, and the worst were 2025-04-04 at negative 6.10% and 2022-09-13 at negative 5.52%. A rule that exited on 2025-04-15 sold after the largest one-day gain of the window.
Among the other broad index funds, golden cross returned 13.82% on QQQM, 12.39% on IOO, 8.99% on VOOG, and 8.71% on SPY. IWM returned 1.05% and was the weakest of the group, with 4 round trips. For the same fund under different rules, the 200-day filter on QQQ returned 12.9%, and the weekly 7% target returned 16.52%, close to holding.
The rules
Hold while the 50-day average is above the 200-day; step aside when it crosses below (the death cross).
- WHEN the market opens · IF not invested AND SMA(50) > SMA(200) · THEN buy with 98% of the sleeve
- WHEN the market opens · IF invested AND SMA(50) < SMA(200) · THEN sell the whole position
A widely used trend filter. When the 50-day simple moving average is above the 200-day, the asset is in a long-term uptrend and the strategy holds. When it crosses below, the strategy moves to cash. It trades rarely, with a handful of signals per decade on an index. Its use is skipping the deepest bear markets, and it will lag some rallies.
Good for: long-horizon investors who want to hold trends but sidestep multi-year bear markets.
Watch out: crosses lag at turning points: the strategy gives back the first leg of a crash and misses the first leg of a recovery, and choppy sideways markets whipsaw it.
How the two rules interact with a slow index
The entry rule asks whether the 50-day average is above the 200-day. The exit rule asks the opposite. Both use the open after the signal, so each order fills at the open after the averages cross. The 98% sizing leaves a small cash residue.
On QQQ, the first entry on 2021-10-19 came late in a rise that peaked a month later. The first exit on 2022-03-03 came after the position had lost value. The second entry on 2023-03-13 came after the fund was already up from its November 2022 low. The second exit on 2025-04-15 came after the April low. The third entry on 2025-06-24 came after the rebound had started. Each of the five fills was after the move it responded to, and that is how a two-average crossover is built: the signal needs both averages to move, and averages move slowly.
The other 11 templates on QQQ trade more and respond faster, or use different signals. The monthly cycle on QQQ was invested for most of every month and returned 14.73%. The SMA 10/50 trend uses faster averages and returned 9.03%. Golden cross sits between them in CAGR and below most in the number of decisions it makes. The result on QQQ is one test of that choice over one window.
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 golden cross beat buy-and-hold on QQQ?
Over 2021-01-04 to 2026-10-02, golden cross on QQQ returned 13.6% annualized vs 16.7% for buy-and-hold: it trailed buy-and-hold by 3.1% per year, with a maximum drawdown 12.0 points shallower than holding (22.2% vs 34.2%).
How many trades did it make?
2 completed round trips over 5.7 years (5 fills), with 50% of round trips closing profitably.
How often does a golden cross happen?
On a broad index, roughly every couple of years. The backtest pages show the exact trade count for each ETF over the 2021 to 2026 window.
Golden cross vs buy and hold: which does better?
It depends on the asset and the window. Every backtest page here shows the same-window buy-and-hold comparison.
Does golden cross beat buy-and-hold on QQQ?
Not in this test. From 2021-01-04 to 2026-10-02 golden cross returned 13.6% a year against 16.7% for buy-and-hold. The maximum drawdown was 22.2% against 34.2%. It beat holding only in 2022.
How many trades did golden cross make on QQQ?
It made 2 closed round trips and had a third position open at the end of the data. The first lost 6.27% over 135 days and the second gained 61.96% over 764 days. The open position was up 40.48%.
When did golden cross buy and sell QQQ?
It bought on 2021-10-19 and sold on 2022-03-03. It bought again on 2023-03-13 and sold on 2025-04-15. The third entry was on 2025-06-24 and the position was still open on 2026-10-02.
Does a 50/200 crossover protect against bear markets?
In 2022 it did, in part. The rule returned negative 11.8% against negative 31.7% for holding. It gave no protection in early 2025, when it held the position through a drawdown of 22.2% and sold after the low.
Is 50/200 the best setting for golden cross on QQQ?
Of four settings tested, 50/200 had the highest CAGR at 13.6%. The 60/250 version returned 11.3%, 40/200 returned 9.8%, and 50/150 returned 8.3%. With 2 or 3 trades each, the differences reflect timing on a few crossings.
How much do trading costs affect golden cross on QQQ?
Very little. With 5 fills, the CAGR went from 13.6% to 13.54% at 10 basis points of slippage. The final value went from $20,793 to $20,736.
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.