DeployQuantVOX trading strategies, backtested
Vanguard Communication Services ETF: communication services: media, internet platforms and telecom. Every DeployQuant template run on VOX over 5.7 years of minute data, same engine, same window, sorted by return.
VOX holds communication services stocks, and its history in this window is one deep fall and a long climb out. Buy-and-hold returned 9.08% a year from 2021-01-04 to 2026-10-02, ended $10,000 at $16,467 and had a maximum drawdown of 45.59% with a Sharpe of 0.55. One of the 12 templates beat holding on CAGR. That was the golden cross at 11.2%, and it did so with 3 trades and a drawdown of 20.91%.
The other 11 templates finished behind holding on return, and 11 of the 12 had a shallower drawdown than the fund. The spread between templates was wide. After golden cross came 3-month momentum at 8.92%, the SMA-200 plus trailing stop at 8.57%, and the weekly 7% target at 8.12%. At the bottom were RSI mean reversion at 2.39% and the 20-day momentum breakout at 2.19%. All 12 had a positive CAGR.
What decided the order was how each template handled 2022. VOX fell 38.81% that year in the fund's own record and rose 44.73% in 2023, which is a V shape that punishes any template that sold low and bought back late. Templates that sat out the decline entirely and re-entered early in the recovery did well, and those that traded through the decline did badly.
The numbers are one window of 5.7 years, with no fees or slippage in the headline run and a $10,000 starting balance. The sample sizes for the best results are very small.
| Strategy | CAGR | max DD | Sharpe | trades | win rate | final value |
|---|---|---|---|---|---|---|
| Golden Cross (SMA 50/200) | 11.2% | −20.9% | 0.84 | 3 | 67% | $18,392 |
| 3-Month Momentum Switch | 8.9% | −13.0% | 0.77 | 11 | 64% (+1 open) | $16,336 |
| SMA-200 Trend + 15% Trailing Stop | 8.6% | −22.1% | 0.65 | 3 | 33% (+1 open) | $16,033 |
| Weekly Entry + 7% Target | 8.1% | −36.8% | 0.53 | 114 | 25% | $15,652 |
| 200-Day SMA Regime Filter | 6.5% | −20.1% | 0.55 | 17 | 24% | $14,370 |
| RSI(2) Dip Snapback | 6.1% | −30.1% | 0.49 | 157 | 67% (+1 open) | $14,039 |
| SMA 10/50 Trend | 6.1% | −19.6% | 0.52 | 18 | 56% (+1 open) | $14,018 |
| First-to-Last Day of Month | 6.0% | −46.2% | 0.41 | 69 | 61% (+1 open) | $13,953 |
| EMA 12/26 Trend | 5.8% | −19.4% | 0.50 | 24 | 46% (+1 open) | $13,825 |
| Drawdown Dip Buyer + 8% Target | 4.1% | −40.9% | 0.32 | 3 | 100% | $12,568 |
| RSI(14) Mean Reversion | 2.4% | −35.1% | 0.23 | 14 | 71% | $11,455 |
| 20-Day Momentum + Trailing Stop | 2.2% | −27.7% | 0.23 | 8 | 63% | $11,324 |
| Buy & hold VOX | 9.1% | −45.6% | 0.55 | – | – | $16,467 |
Golden cross buys when the 50-day average crosses above the 200-day and sells on the reverse. On VOX it made 3 round trips. The first, from 2021-10-19 to 2021-12-17, lost 7.77%. The second, from 2023-03-13 to 2025-05-06, gained 77.05% over 785 days. The third, from 2025-06-18 to 2026-07-29, gained 13.99%. The middle trade carries the result. It also explains the yearly pattern: golden cross returned 0% in 2022, with no position, against −37.7% for holding, then returned 36.2% in 2023 and 32.5% in 2024. In 2021 it lost 7.5% against a gain of 15.2% for holding, and in 2025 it returned 14.7% against 25.6%. The profit factor of 12.13 reflects one large win next to one small loss.
The drawdown of 20.91% ran from 2025-02-14 to 2025-04-08, a fall the position rode through because the crossover is slow, since the 785-day trade was still open on those dates. A reader should treat the 2.12 point CAGR edge as the result of one well-timed 785-day hold and not as a rate that 3 trades can establish.
The 3-month momentum switch made 11 round trips with a 64% win rate and a profit factor of 6.09. It returned 8.92% with a 12.96% drawdown, the shallowest of all 12 templates. Its best trades were the 264-day hold from 2023-11-16 to 2024-08-06 at 17.84% and the 191-day hold from 2025-06-10 to 2025-12-18 at 17.24%. It beat holding in 2022 by 36.4 points of return, yet gave back ground in 2023, when it returned 21.9% against 42.8%. A template that is flat for much of a V-shaped recovery trades its drawdown for upside, and the CAGR gap to holding was only 0.16 points.
The SMA-200 trend plus 15% trailing stop returned 8.57%. It lost 14.9% on the trade from 2021-10-19 to 2022-01-24, lost 8.05% on a 45-day trade in early 2023, then held from 2023-03-17 to 2025-04-03 for 59.93% over 748 days. The stop fired on 2025-04-03, and the position was re-entered on 2025-05-05 at an adjusted price of 147.62. At the end of the window that open trade was up 28.78%. The trade list has 3 closed trades and 1 open one, with a win rate of 33%.
The weekly 7% target made 114 round trips, a win rate of 25%, an average win of 7.17% and an average loss of 1.8%. It was invested 82.6% of the time and returned 8.12%, but its drawdown was 36.79%, almost as deep as the fund's own. The longest losing streak was 15 trades. The 200-day regime filter returned 6.52% with 17 round trips and a win rate of 24%. Its profit factor of 2.09 comes from three long trades: 43.97% over 517 days, 23.63% over 322 days and 11.57% over 227 days. Eleven of its 17 round trips came in 2026, when it flipped in and out around the average.
The RSI(2) snapback made 157 round trips, returned 6.09% and had a 30.05% drawdown, with a worst trade of −9.41% from 2025-02-20 to 2025-03-18. SMA 10/50 and EMA 12/26 returned 6.06% and 5.8%. Monthly cycle returned 5.97% with a drawdown of 46.22%, slightly deeper than holding, because it is invested 95.2% of sessions. The dip buyer returned 4.06% with a win rate of 100% on 3 trades, one of which was held for 916 days. RSI mean reversion returned 2.39% and its worst trade, 189 days from 2021-09-21 to 2022-03-29, lost 14.78%. The 20-day momentum breakout returned 2.19%. Details for each are on the template pages, such as 3-month momentum and the trailing stop version.
The later years separated the templates in a different way. In 2025 holding returned 25.6%. SMA 10/50, the weekly 7% target and monthly cycle returned 26.2%, 26% and 26%, and the momentum breakout returned 29.4%. Golden cross returned 14.7% that year because its position was only entered on 2025-06-18 after the previous trade ended on 2025-05-06. In 2026 holding was down 1.1% through 2026-10-02. The dip buyer returned 7.9% and RSI mean reversion returned 8%, while the 200-day regime filter lost 16.9% after flipping in and out of the fund 11 times and the momentum breakout lost 13.7%. The leader changed from one calendar year to the next.
How each strategy traded VOX
| Strategy | Time in market | Avg hold (days) | Best trade | Worst trade | Profit factor | With 10 bps slippage |
|---|---|---|---|---|---|---|
| golden cross | 59.5% | 417 | 77.0% | −7.8% | 12.13 | 11.1% |
| 3-month momentum | 56.0% | 106 | 17.8% | −4.0% | 6.09 | 8.5% |
| trend + trailing stop | 66.8% | 297 | 59.9% | −14.9% | 2.19 | 8.4% |
| weekly 7% target | 82.6% | 14 | 9.3% | −9.6% | 1.35 | 3.3% |
| 200-day regime filter | 59.0% | 73 | 44.0% | −6.2% | 2.09 | 5.9% |
| RSI(2) snapback | 40.6% | 5 | 4.2% | −9.4% | 1.36 | 0.5% |
| SMA 10/50 trend | 60.5% | 69 | 22.1% | −6.6% | 2.01 | 5.4% |
| monthly cycle | 95.2% | 28 | 12.0% | −12.9% | 1.33 | 3.5% |
| EMA 12/26 trend | 62.3% | 53 | 21.8% | −7.3% | 1.82 | 4.9% |
| dip buyer | 48.1% | 337 | 8.3% | 8.0% | – | 4.0% |
| RSI mean reversion | 36.8% | 55 | 9.7% | −14.8% | 1.39 | 1.9% |
| momentum breakout | 56.7% | 148 | 21.4% | −9.5% | 1.48 | 1.9% |
The dip buyer's trade list shows a limit of the design. Its first trade entered on 2021-12-02 and exited on 2024-06-05 at an 8.01% gain after 916 days. The template has no stop, so the position sat through the whole 2022 decline and waited for the 8% target. The drawdown for that run was 40.93% from 2021-12-27 to 2022-11-09, almost as deep as holding's 45.59%. The two later trades closed in 86 and 9 days at 8.26% and 8.17%. The win rate is 100% and the drawdown is among the deeper ones on the page.
Time in market splits the templates into three groups. Monthly cycle and the weekly 7% target were invested 95.2% and 82.6%. The trend group sat between 59.5% for golden cross and 66.8% for the trailing stop. The RSI templates and the breakout sat at 36.8% to 56.7%. The groups with the highest exposure had the deepest drawdowns, which is what a fund with a 46.74% fall would produce.
Slippage cost the active templates the most. RSI(2) snapback went from 6.09% to 3.26% at 5 basis points and 0.52% at 10 basis points, and its ending value fell from $14,039 to $10,302. The weekly 7% target went from 8.12% to 4.81% and 3.32%, with the drawdown widening from 36.79% to 42.24%. Monthly cycle fell from 5.97% to 3.52%. Golden cross barely moved, from 11.2% to 11.11%, and the trailing stop template went from 8.57% to 8.38%.
VOX's liquidity makes these cost runs worth reading carefully. The fund's average daily dollar volume in the data was $26,455,650 and the median minute volume was 339 shares. A minute that thin means a market order of a few hundred shares can move the price noticeably. The test fills at the minute bar and applies the stated basis points. A thin fund with 157 or 114 round trips pays more in practice than the 10 basis point run shows, if the spread is wider than that.
The overnight share of log return was 56.86% and the intraday share was 43.14%, with an average overnight gap of 0.55% and an average daily range of 1.45%. Day-of-week averages were small, between 0% on Thursday and 0.14% on Monday.
What the winning trades were
Golden cross is the one template that beat holding, and three trades explain it. The first, from 2021-10-19 to 2021-12-17, lost 7.77% over 59 days. The second ran 785 days from 2023-03-13 to 2025-05-06 and gained 77.05%. The third ran 406 days from 2025-06-18 to 2026-07-29 and gained 13.99%. With a 67% win rate on 3 trades, one trade accounts for the result, and the test has no way to separate the rule from the entry date. The trailing stop template has the same structure: a 59.93% gain over 748 days from 2023-03-17 against losses of 14.9% and 8.05% on its two other trades.
The 200-day regime filter won 24% of its 17 trades and still returned 6.52% a year. Its best trades were 43.97% over 517 days from 2023-10-31 and 23.63% over 322 days from 2025-05-05. In 2022 it returned 0%, since it was out of the fund. In 2026 it lost 16.9%, and its drawdown of 20.15% from 2026-01-29 to 2026-10-01 had not recovered by the last day, so the filter finished the window in a loss that started in January.
The weekly 7% target made 114 round trips and won 25%. Its longest losing streak was 15. It returned 44.8% in 2023 and lost 29.2% in 2022, with a drawdown of 36.79% from 2021-09-07 to 2022-12-28, which recovered on 2024-01-29. The cost runs hurt it: 4.81% a year at 5 basis points and 3.32% at 10, with the drawdown reaching 42.24%. The RSI(2) snapback made 157 round trips, won 67%, and went from 6.09% to 3.26% to 0.52% across the same cost runs. Its worst trade, a loss of 9.41% over 26 days from 2025-02-20, came before the April 2025 low.
The monthly cycle was the template with the deepest drawdown, 46.22%, a little more than holding's 45.59%. Its worst months were April 2022 at -12.88% and September 2022 at -11.15%, and its best was January 2023 at 12.03%. A rule that is invested 95.2% of the time follows the fund almost exactly, so it inherits the fall and the late recovery without any of the protection the other templates had.
The buy-and-hold record shows how large the swing was. The fund's best month was 2023-01 at 14.39% and its worst was 2022-04 at -14.44%. The month-of-year averages put January at 4.65%.
How VOX behaved
| Measure | VOX |
|---|---|
| Data in this test | 2021-01-04 to 2026-10-02 (1444 sessions) |
| Total return, buy and hold | 70.2% |
| Annualized volatility | 21.0% |
| Deepest drawdown | −46.7% (2021-09-01 to 2022-11-09) |
| Up days | 52.3% |
| Average daily range | 1.45% |
| Average overnight gap | 0.55% |
| Correlation to SPY | 0.84 |
| Correlation to QQQ | 0.83 |
| Correlation to TLT | 0.07 |
| Sessions above the 200-day average | 68.5% |
| Crossings of the 200-day average | 34 |
| Falls of 10% or more from a 20-day high | 26 |
Calendar years
| Year | Return |
|---|---|
| 2021 | 15.6% |
| 2022 | −38.8% |
| 2023 | 44.7% |
| 2024 | 33.1% |
| 2025 | 26.2% |
| 2026 (part) | −1.1% |
Biggest single days
| Best day | Move |
|---|---|
| 2025-04-09 | 9.4% |
| 2022-11-10 | 6.0% |
| 2023-02-02 | 5.5% |
| 2023-04-27 | 5.0% |
| 2022-11-30 | 4.3% |
| Worst day | Move |
|---|---|
| 2022-02-03 | −6.1% |
| 2025-04-04 | −5.4% |
| 2022-09-13 | −5.1% |
| 2025-04-03 | −5.0% |
| 2022-06-13 | −4.7% |
Average return by calendar month
| Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 4.7% | −0.9% | −0.3% | −0.3% | 3.1% | 0.8% | 1.6% | 0.8% | −1.6% | 0.6% | 2.9% | 0.5% |
Most and least correlated funds
| Most correlated | Least correlated | ||
|---|---|---|---|
| VV | 0.85 | SH | -0.84 |
| SPY | 0.84 | SDS | -0.84 |
| VOO | 0.84 | SPDN | -0.84 |
| SSO | 0.84 | SQQQ | -0.83 |
| SPUU | 0.84 | QID | -0.83 |
The fund's own record explains why drawdown protection counted. Annualized volatility was 20.99%. The maximum drawdown was 46.74%, from 2021-09-01 to 2022-11-09, and the fund did not recover until 2024-10-04. The longest drawdown lasted 776 sessions. Calendar returns were 15.64% in 2021, −38.81% in 2022, 44.73% in 2023, 33.15% in 2024, 26.22% in 2025 and −1.15% in 2026 so far. A second pullback of 20.73% ran from 2025-02-14 to 2025-04-08 and recovered in 80 days, and a third of 13.32% began on 2026-01-29 and had not recovered by the end of the data.
The fund was above its 200-day average on 68.51% of sessions and crossed it 34 times. The fund spent nearly a third of sessions below the average, so the trend templates had real periods to sit out. There were 26 falls of 10% or more from a 20-day high, covering 78 days, so the dip buyer had plenty of signals. It took 3, because it holds a position until the target is hit.
Mean reversion shows a pattern in this fund. After 24 sessions with RSI(14) below 30, the median move over the next 5 days was 2.14% and over 20 days 2.34%, against baselines of 0.34% and 1.31%. After 162 sessions with RSI(2) below 10, the medians were 1.13% over 5 days and 2.21% over 20 days. Lag 1 autocorrelation was −0.01, so a down day did not predict the next day. The medians show above-baseline forward moves, yet the RSI(14) template returned 2.39% and its drawdown was 35.15%, because the oversold readings in 2022 arrived while the fund kept falling. Its three completed trades in 2022 all lost, at −14.78%, −9.42% and −8.63%.
The biggest days were 9.38% on 2025-04-09 and 5.97% on 2022-11-10, against worst days of −6.13% on 2022-02-03 and −5.39% on 2025-04-04. The fund's beta to SPY was 1.08 and to QQQ 0.78, with a correlation of 0.84 to SPY and 0.83 to QQQ. It correlated 0.85 with VV and 0.07 with TLT. The inverse funds SH, SDS and SPDN were at −0.84. Monthly averages were highest in January at 4.65%, May at 3.12% and November at 2.86%, and lowest in September at −1.65% and February at −0.95%, from 5 or 6 observations per month.
Among other sector funds, SOXX held 31.11% a year with weekly 7% target at 30.54%, and XLK held 22.07% with monthly cycle at 20.86%. XLF held 12.58% and RSI mean reversion reached 11.82%. XLP held 5.8% and RSI(2) snapback returned 8.22%, while XLY held 6.27% and the weekly 7% target returned 10.19%. VOX is the one fund on this list where a slow trend rule beat holding, and the fund's drawdown was a single long episode, which is the shape a 200-day average handles best.
The 2026 drawdown deserves a separate note because it is still open. Holding fell 13.32% from 2026-01-29 to 2026-03-27 and had not made a new high by the end of the data. The 200-day regime filter, SMA 10/50 and EMA 12/26 all show an open drawdown in the same period, with depths of 20.15%, 14.4% and 15.41%. The golden cross shows 13.23% over the same dates. The templates with gains in 2026 were those that bought the March dip: the dip buyer entered on 2026-03-30 and exited on 2026-04-08 at 8.17%, and RSI mean reversion returned 8% for the year. Whether those trades repeat is outside what one window can show, and each rests on a single entry.
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 VOX?
Of the 12 templates tested on VOX over 2021-01-04 to 2026-10-02, the strongest by CAGR was golden cross at 11.2% (max drawdown 20.9%), versus 9.1% 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 VOX?
1 of 12 templates beat VOX buy-and-hold (9.1% CAGR) on this window; 11 of 12 had a shallower maximum drawdown than holding (45.6%).
Which strategy worked best on VOX?
Golden cross returned 11.2% a year against 9.08% for holding, with a 20.91% maximum drawdown. It made 3 round trips, and one of them gained 77.05% over 785 days. Treat it as one good trade more than a repeatable rate.
How many strategies beat buy-and-hold on VOX?
One of 12, the golden cross. Eleven of 12 had a shallower maximum drawdown than the fund's 45.59%. Every template had a positive CAGR.
Why did trend strategies do well on VOX in 2022?
VOX fell 38.81% that year. Golden cross and the 200-day regime filter returned 0% because they held no position, and holding lost 37.7%. The SMA-200 trailing stop lost 9.7%.
Is VOX liquid enough for high-frequency templates?
Average daily dollar volume in the data was $26,455,650 and median minute volume was 339 shares. RSI(2) snapback fell from 6.09% to 0.52% at 10 basis points of slippage. A thin fund with 157 trades can cost more than that run shows.
Does RSI mean reversion work on VOX?
It returned 2.39% a year with a 35.15% drawdown, the second weakest result. After RSI(14) fell below 30, the median 20-day move was 2.34% against a 1.31% baseline, but the oversold signals in 2022 came before the fund's low on 2022-11-09.
How deep was the VOX drawdown?
Buy-and-hold lost 45.59% from 2021-09-01 to 2022-11-09 and recovered on 2024-10-04. The fund's own close to close figure was 46.74%, and the longest drawdown lasted 776 sessions.
Other sector 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.