DeployQuant200-Day SMA Regime Filter on QAI
NYLI Hedge Multi-Strategy Tracker ETF: replicates a basket of hedge-fund strategies. Backtest 2021-01-04 to 2026-10-02, $10,000 starting capital, computed by the same engine that runs live DeployQuant strategies.
The 200-day regime filter keeps QAI in the portfolio only while yesterday's close is over its 200-day simple average, and otherwise holds cash. QAI, the NYLI Hedge Multi-Strategy Tracker ETF, tries to replicate a basket of hedge-fund strategies. From 2021-01-04 to 2026-10-02 the rule turned $10,000 into $12,704.41, a 4.26% CAGR, with a 6.02% max drawdown and a Sharpe ratio of 0.94. Holding QAI turned $10,000 into $12,450.50, a 3.89% CAGR, with a 14.65% drawdown and a Sharpe of 0.64. The rule beat holding on return and on drawdown, and by a small margin on return.
The result shows what a regime filter does when it works. The rule was in cash through the 2022 decline, so it avoided most of a fall that cost holding 8.6% for the year. It then held the fund through the long rise from early 2023. The cost of that choice was a 29% win rate. The rule made 7 closed round trips and only 2 won. The average winner gained 7.55% and the average loser lost 1.18%, which gave a profit factor of 2.41. The five losers were small, and the two winners carried the result.
The rule ranks second of the 12 templates on QAI and 26th of 59 funds for this template. The median CAGR for the template across all 59 funds was 1.84%, so QAI's 4.26% sits well above the median. The 200-day regime filter is the simplest rule in the library, with one number and no oscillators, and QAI is a fund with low volatility, which keeps the whipsaw losses small.
This is a single window of about 5.74 years, and the headline figures include no fees or slippage. QAI has only 7 closed trades in it, and the result rests on two of them.
Year by year
| Year | 200-day regime filter | buy & hold |
|---|---|---|
| 2021 | −1.2% | −0.2% |
| 2022 | −1.8% | −8.6% |
| 2023 | 9.0% | 10.0% |
| 2024 | 6.6% | 6.6% |
| 2025 | 4.4% | 8.0% |
| 2026 | 8.0% | 8.0% |
Year by year
The rule and holding by calendar year: 2021, -1.2% against -0.2%; 2022, -1.8% against -8.6%; 2023, 9% against 10%; 2024, 6.6% against 6.6%; 2025, 4.4% against 8%; and 2026 so far, 8% against 8%. The rule beat holding only in 2022. The two were level in 2024 and 2026, and the rule lagged in 2021, 2023 and 2025.
The window begins on 2021-01-04, and a 200-day average needs 200 sessions of history. The monthly table reads 0 from January to September 2021, so the rule was waiting. Its first entry came on 2021-10-20 at $29.15 and it sold on 2021-10-28 at $29.06, a 0.31% loss. A second trade from 2021-11-02 at $29.12 to 2021-11-23 at $28.86 lost 0.89%. Both were short holds that ended when the close fell back under the average.
2022 is the year the rule paid for itself. QAI lost 8.6% in 2022 and the rule lost 1.8%. The monthly table shows 0 for every month of 2022 except December. The fund's own drawdown ran from 2021-02-17 to 2022-10-14, and the rule was out of the fund for that stretch. In December 2022 it bought on 2022-12-02 at $26.62 and sold on 2022-12-06 at $26.5, then bought on 2022-12-13 at $26.75 and sold on 2022-12-16 at $26.37. Those two trades lost 0.45% and 1.42%, and they produced the 1.83% loss in the monthly table. The price was crossing the average back and forth near the bottom, which is the whipsaw the strategy page warns about.
2023 was the turn. The rule bought on 2023-01-05 at $26.48 and held to 2023-10-27, a 295-day trade that gained 3.1%. It sold at $27.3 when the close fell under the average, and it bought again on 2023-10-31 at $27.42. That second trade ran 497 days to 2025-03-11 and gained 12%, the best of the test. The rule made 9% for 2023 against 10% for holding, because it entered in January and not at the start of the year.
In 2024 the rule and the fund both made 6.6%, since the rule held the same position throughout. In 2025 the rule made 4.4% against 8%. The 497-day trade ended on 2025-03-11, and the rule bought back on 2025-03-12 at $30.93 and lost 2.85% by 2025-04-04, and the rule was out of the fund until it bought again on 2025-05-05 at $30.88, missing the sharp recovery from the April low. That last trade was open at the end of the data and up 17.36%. For 2026 so far the rule made 8%, the same as holding.
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% | −0.3% | −0.9% | 0.0% |
| 2022 | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | −1.8% |
| 2023 | 2.7% | −1.1% | 0.6% | 0.5% | −0.7% | 2.2% | 1.9% | −0.4% | −0.7% | −1.8% | 3.2% | 2.3% |
| 2024 | −0.5% | 1.6% | 1.3% | −1.0% | 1.3% | 0.3% | 1.1% | 0.8% | 1.4% | −0.6% | 2.2% | −1.4% |
| 2025 | 1.5% | −0.5% | −2.0% | −2.7% | 1.6% | 1.6% | 0.6% | 1.4% | 1.7% | 0.7% | −0.1% | 0.5% |
| 2026 | 2.3% | 1.7% | −2.2% | 4.5% | 1.9% | 0.6% | −1.4% | 1.0% | −0.9% | 0.4% | – | – |
The monthly table
The early cells of the table are almost all zeros, which is the 200-day warm-up and the 2022 decline. After that the pattern changes. From January 2023 the rule held the fund nearly every month, and the table reads like a record of the fund itself.
The best month was April 2026 at 4.51%, followed by November 2023 at 3.21% and January 2023 at 2.7%. The worst was April 2025 at -2.69%, then March 2025 at -1.97% and December 2022 at -1.83%. Holding had its best month in April 2026 at 4.52% and its worst in September 2022 at -3.91%. The two tables differ in September 2022: the rule was in cash and held 0, while the fund lost 3.91%.
The April 2025 loss deserves a comment. The rule bought on 2025-03-12 and sold on 2025-04-04 at $30.05, a 2.85% loss. The fund's worst day of the window was 2025-04-04 at -2.72%, the same date as the rule's exit and its drawdown trough. The best day was 2025-04-09 at 3.17%, and the rule was not in the fund for it. The next entry on 2025-05-05 came a month later. That is how the rule gives up some of a V-shaped recovery: it needs a close back above the average before it buys.
The fund's calendar-month averages show November at 1.58% and March at -0.46%, with September at -0.63%. Each has 5 or 6 observations. These describe the window, and the rule does not use them.
Every trade
200-day regime filter on QAI made 7 closed round trips and one position still open at the end of the test, an average hold of 122 days, an average winner of 7.55%, an average loser of −1.18%, a profit factor of 2.41, a longest losing streak of 4. It held a position at the close on 65.2% of trading days.
| Entry | Entry price | Exit | Exit price | Return | Days held |
|---|---|---|---|---|---|
| 2021-10-20 | $29.15 | 2021-10-28 | $29.06 | −0.3% | 8 |
| 2021-11-02 | $29.12 | 2021-11-23 | $28.86 | −0.9% | 21 |
| 2022-12-02 | $26.62 | 2022-12-06 | $26.50 | −0.5% | 4 |
| 2022-12-13 | $26.75 | 2022-12-16 | $26.37 | −1.4% | 3 |
| 2023-01-05 | $26.48 | 2023-10-27 | $27.30 | 3.1% | 295 |
| 2023-10-31 | $27.42 | 2025-03-11 | $30.71 | 12.0% | 497 |
| 2025-03-12 | $30.93 | 2025-04-04 | $30.05 | −2.9% | 23 |
| 2025-05-05 | $30.88 | open | – | 17.4% | – |
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 trades
The list has 8 entries, 7 closed and 1 open. The median closed trade lost 0.45% and the median hold was 21 days. The shortest hold was 3 days and the longest was 497. The open trade began on 2025-05-05 at an adjusted $30.88 and was up 17.36% at the end of the data.
- 2021-10-20 to 2021-10-28 at $29.15 to $29.06: -0.31% in 8 days.
- 2021-11-02 to 2021-11-23 at $29.12 to $28.86: -0.89% in 21 days.
- 2022-12-02 to 2022-12-06 at $26.62 to $26.5: -0.45% in 4 days.
- 2022-12-13 to 2022-12-16 at $26.75 to $26.37: -1.42% in 3 days.
- 2023-01-05 to 2023-10-27 at $26.48 to $27.3: +3.1% in 295 days.
- 2023-10-31 to 2025-03-11 at $27.42 to $30.71: +12% in 497 days.
- 2025-03-12 to 2025-04-04 at $30.93 to $30.05: -2.85% in 23 days.
- 2025-05-05 at $30.88, open: +17.36%.
Five of the seven closed trades were losses, and all five were under 3%. The two winners held for 295 and 497 days. The shortest holds were in the two whipsaw clusters, October to November 2021 and December 2022, and each cluster had 2 losing trades. No more than 2 winners came in a row, and as many as 4 losers did.
Round trips by exit year were 2 in 2021 with 0 wins, 2 in 2022 with 0, 1 in 2023 with 1, none in 2024 and 2 in 2025 with 1. There was no exit in 2024 because the 497-day trade was held through the whole year. The rule's average hold of 121.6 days is long because of those two trades. Listed prices are adjusted for distributions, which puts them under the quotes shown at the time.
The distribution of the 8 trades shows why a win rate of 29% is a poor summary. The rule gives away small amounts in chop, and it keeps long trends. The rule's two long holds captured the rise from 2023.
Largest drawdowns
| Peak | Low point | Depth | Days to low | Recovered | Days to recover |
|---|---|---|---|---|---|
| 2025-02-18 | 2025-04-04 | −6.0% | 45 | 2025-09-11 | 160 |
| 2026-02-25 | 2026-03-30 | −3.6% | 33 | 2026-04-13 | 14 |
| 2023-02-02 | 2023-03-15 | −3.3% | 41 | 2023-04-04 | 20 |
Buy-and-hold's deepest drawdown ran from 2021-02-17 to 2022-10-14 and reached −14.7%.
Drawdowns
The rule's deepest drawdown was 6.02%, from a peak on 2025-02-18 to a trough on 2025-04-04, 45 days later, recovered on 2025-09-11 after another 160 days. The second was 3.64%, from 2026-02-25 to 2026-03-30, recovered on 2026-04-13, and the third was 3.32%, from 2023-02-02 to 2023-03-15, recovered on 2023-04-04.
Holding QAI had a 14.65% drawdown from 2021-02-17 to 2022-10-14, 604 days to the trough, recovered on 2024-03-12 after 515 days. It also had 7.63% from 2025-02-18 to 2025-04-08, recovered on 2025-06-26, and 3.65% from 2026-02-25 to 2026-03-30. The rule's maximum drawdown was 8.63 points shallower. The regime filter avoided the long 2021 to 2022 slide completely, and it took the 2025 drop in full.
The 2025 drawdown is the rule's worst, and it shows a limit of the filter. The rule sold on 2025-03-11 and bought back the next day, then held the new position down to its exit on 2025-04-04, near the low. A slow filter reacts after a decline has already happened, and a quick re-entry can land inside it. In 2022 the same slowness did not matter, because the decline was long enough for the close to fall under the average and stay there.
The Sharpe ratio was 0.94 for the rule against 0.64 for holding. QAI's annualized volatility was 6.52%, low for a fund, and the rule reduced it further by sitting out the one bad stretch. The drawdown gap and the Sharpe gap are the real advantage of the rule here, since the return gap is under half a point.
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) | 4.3% | −6.0% | $12,704 | 0.94 |
| 5 basis points | 4.1% | −6.2% | $12,613 | 0.91 |
| 10 basis points | 4.0% | −6.3% | $12,518 | 0.88 |
Cost runs
With 5 basis points per trade the CAGR was 4.13%, the max drawdown 6.16% and the end equity $12,613.04, with a Sharpe of 0.91. With 10 basis points the CAGR was 3.99%, the drawdown 6.3% and the end equity $12,518.06, with a Sharpe of 0.88.
The rule made 15 fills in about 5.74 years, so the cost was small. At 10 basis points the CAGR of 3.99% is still a little above holding's 3.89%. Average dollar volume is about $3,152,202 a day and the median minute holds 317 shares, which is thin. A $10,000 account is small against that volume. A larger account would face a wider effective spread, and the flat charge in these runs does not model it.
Changing the parameters
| Version | CAGR | Max drawdown | Round trips | Win rate | Final value |
|---|---|---|---|---|---|
| Published rules | 4.3% | −6.0% | 7 | 29% | $12,704 |
| 100-day SMA | 3.5% | −4.5% | 24 | 42% | $12,208 |
| 150-day SMA | 4.4% | −4.0% | 14 | 36% | $12,778 |
| 250-day SMA | 5.0% | −5.4% | 2 | 50% | $13,205 |
Parameter variants
Three variants changed the window of the average. A 100-day average returned 3.54% with a 4.46% drawdown, 24 trades and 10 winners. A 150-day average returned 4.36% with a 4% drawdown, 14 trades and 5 winners. A 250-day average returned 4.96% with a 5.38% drawdown, 2 trades and 1 winner. The standard 200-day average returned 4.26% with a 6.02% drawdown and 7 round trips.
The returns are close together, from 3.54% to 4.96%, and each of the three variants ended with a shallower drawdown than the standard rule. The 150-day average had the shallowest at 4%. The 250-day average had the highest CAGR and the highest Sharpe ratio, at 1.098, with 2 trades. The 100-day average made 24 trades and had the lowest CAGR, which fits a shorter average crossing the price more often.
The 250-day result has to be read with care. With 2 trades, its CAGR comes from one long hold, since a 250-day average needs 250 sessions before its first signal, and the window starts in 2021. A result built on a single position is a measure of that position, and the Sharpe of 1.098 follows from the same one. The spread across all four settings is small compared with the gap to a bad outcome, so the rule does not depend on a lucky window length. It also shows that on QAI the benefit came from staying out of 2022, which every variant did.
How QAI behaved
| Measure | QAI |
|---|---|
| Data in this test | 2021-01-04 to 2026-10-02 (1444 sessions) |
| Total return, buy and hold | 25.3% |
| Annualized volatility | 6.5% |
| Deepest drawdown | −14.9% (2021-02-17 to 2022-10-14) |
| Up days | 51.4% |
| Average daily range | 0.53% |
| Average overnight gap | 0.27% |
| Correlation to SPY | 0.82 |
| Correlation to QQQ | 0.80 |
| Correlation to TLT | 0.18 |
| Sessions above the 200-day average | 75.7% |
| Crossings of the 200-day average | 15 |
| Falls of 10% or more from a 20-day high | 0 |
What QAI is and how it behaved
QAI replicates a basket of hedge-fund strategies. Its total return over the window was 25.27%, a 4% CAGR, with a 14.94% max drawdown from 2021-02-17 to 2022-10-14, recovered on 2024-03-12. The fund's returns by calendar year were -0.24% for 2021, -8.77% for 2022, 10.25% for 2023, 6.75% for 2024, 8.13% for 2025 and 8.15% for 2026 up to now. Annualized volatility was 6.52%, the average intraday range was 0.53%, and the fund rose on 51.35% of days, with an average up day of 0.31% and an average down day of -0.32%.
The fund moves with stocks. Its beta to SPY was 0.33 and its correlation to SPY was 0.82, and its beta to QQQ was 0.23 with a correlation of 0.8. It correlated most with QQQE at 0.84 and least with SDS at -0.82. Its correlation to TLT was 0.18. A fund that behaves like a low-beta stock fund fell when stocks fell, as it did in 2022, and a rule that watches its own average can leave during that fall.
QAI closed over its 200-day average on 75.74% of sessions, and the price went through that line 15 times. That is a low number of crossings, and it fits a fund that trends slowly. The fund had 0 events of a 10% drop from its 20-day high. Its longest drawdown lasted 770 sessions.
The RSI figures show a mild bounce effect. RSI(14) fell under 30 on 24 sessions with a median 20-day return of 0.86% against a baseline of 0.45%. On the 154 sessions where RSI(2) sank below 10, the next 5 days had a median gain of 0.29%, with 0.16% as the baseline.
The split between overnight and intraday return is unusual. The overnight share of the log return was -216.57%, and the intraday share was 316.57%. Overnight log returns summed to -48.34% and intraday ones to 70.66%. The fund lost value between the close and the next open and gained it during the day. The rule trades at the open, so it buys after the overnight change and sells after it.
The best sessions were +3.17% on 2025-04-09 and +1.95% on 2022-11-10. The worst were -2.72% on 2025-04-04 and -2.27% on 2022-06-13.
The rules
Own the asset when price closes above its 200-day average; hold cash when it closes below.
- WHEN the market opens · IF not invested AND yesterday's close > SMA(200) · THEN buy with 98% of the sleeve
- WHEN the market opens · IF invested AND yesterday's close < SMA(200) · THEN sell the whole position
One rule and one number. Price above the 200-day moving average has historically coincided with better returns and lower volatility than price below it. This template uses no crossovers and no oscillators, only which side of the long-term average the price is on.
Good for: a first systematic strategy, simple enough to audit every trade.
Watch out: price whips around the 200-day line during volatile bottoms, generating clusters of buy-sell pairs. Some traders add a small buffer band to reduce churn.
How the rule fits this fund
The logic fits in two lines. At the open, buy if yesterday closed over the 200-day average. At the open, sell if yesterday closed under it. It needs no indicator other than the average and trades only when the long trend changes. For a fund with 6.52% volatility, the average is close to the price, and a cross can happen after a small move. That produced the two whipsaw clusters, with 4 losses between 0.31% and 1.42%.
The cost of those clusters was small because QAI moves little each day. A 0.45% loss on a 4-day trade is a small price for a rule that avoided a fall of 8.6% in 2022. Volatility works in both directions here. Low volatility keeps whipsaws cheap, and it also means the gains from a trend are modest, which is why the CAGR is 4.26% and not higher.
Among the 12 templates on QAI, the golden cross made 4.28% with a 7.63% drawdown and the regime filter made 4.26% with 6.02%. Behind it, the weekly 7% target earned 4.21% at a 15.8% drawdown. The RSI(2) snapback earned 3.43% at 6.84%, and the monthly cycle 3.28% at 16.07%. The EMA 12/26 trend made 2.76%, the trend with trailing stop 2.61%, the RSI mean reversion 2.15%, the SMA 10/50 trend 2.12% and the 3-month momentum 2.11%. The dip buyer and the momentum breakout made no trades, with a CAGR of 0.
The regime filter's 6.02% drawdown is the second shallowest of the 10 templates that traded, behind the 3-month momentum's 5%. The golden cross, the closest relative, made almost the same CAGR with a deeper drawdown. Both are slow trend rules.
Within the alternative-strategy group the spread was wide. CLSE returned 20.38%, QAI 4.26%, ALTY 1.84% and KMLM 0.24%, while RINF lost 2.11% and CTA lost 3.02%. QAI had 7 round trips, the fewest in the group. KMLM, RINF and CTA made 21, 37 and 25, and those chop trades are where the funds lost. The QAI fund page lists every template on the fund.
This is one low-volatility fund over one window with one decline in it. The test cannot say how the rule behaves in a faster or deeper decline.
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 200-day regime filter beat buy-and-hold on QAI?
Over 2021-01-04 to 2026-10-02, 200-day regime filter on QAI returned 4.3% annualized vs 3.9% for buy-and-hold: it beat buy-and-hold by 0.4% per year, with a maximum drawdown 8.6 points shallower than holding (6.0% vs 14.7%).
How many trades did it make?
7 completed round trips over 5.7 years (15 fills), with 29% of round trips closing profitably.
Why the 200-day average specifically?
It approximates a year of trading days and has been studied across decades of data. It is not the best window for every asset. The per-ETF backtests here show where it helped and where it didn't.
How did the 200-day regime filter do on QAI?
It turned $10,000 into $12,704.41, a 4.26% CAGR, with a 6.02% max drawdown and a Sharpe of 0.94 from 2021-01-04 to 2026-10-02. Holding QAI returned 3.89% a year with a 14.65% drawdown.
Why is the win rate only 29%?
The rule made 7 closed round trips and 2 won. The five losses were between 0.31% and 2.85%, while the two winners gained 3.1% and 12%, over 295 and 497 days. The profit factor was 2.41.
Did the filter avoid the 2022 decline?
Mostly. QAI lost 8.6% in 2022 and the rule lost 1.8%. The monthly table shows 0 for every month of 2022 except December, when two short trades lost a combined 1.83%.
What was the worst drawdown?
The rule's worst was 6.02%, from 2025-02-18 to 2025-04-04, recovered on 2025-09-11. Holding had 14.65% from 2021-02-17 to 2022-10-14.
Do costs change the result?
Slightly. At 5 basis points the CAGR was 4.13% and at 10 it was 3.99%, against 4.26% with none. The rule made 15 fills.
Does a different average length help?
The 100, 150 and 250-day averages returned 3.54%, 4.36% and 4.96%. The 250-day version made only 2 trades, so its result comes from one long hold. See the [200-day regime filter page](/learn/strategies/sma200-regime/) for other funds.
Is this a recommendation for QAI?
No. It is one backtest window with 7 closed trades, and the headline run has no fees. The result would differ in another period.
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.