DeployQuantRSI(2) Dip Snapback on SPY
SPDR S&P 500 ETF Trust: the most-traded S&P 500 ETF and a common benchmark for US large caps. Backtest 2021-01-04 to 2026-10-02, $10,000 starting capital, computed by the same engine that runs live DeployQuant strategies.
The RSI(2) snapback buys SPY at the open after the 2-day RSI closes under 10 and sells the whole position at the open after the 2-day RSI closes above 70. It uses 98% of the sleeve, takes no margin, and has no stop. The SPY run covers 2021-01-04 to 2026-10-02 with $10,000. The result was 13.17% a year, a final value of $20,342, a max drawdown of 12.69% and a Sharpe of 1.21. Buy-and-hold SPY made 14.56% a year with a 23.99% max drawdown and a Sharpe of 0.95, and ended at $21,820.
The rule made 172 round trips and won 122 of them, a win rate of 71%. It was invested 36.2% of the time and held for 4.4 days on average. Its profit factor was 2.18. It trailed holding by 1.39 points of CAGR, finished $1,479 behind in dollars, and had a drawdown 11.3 points shallower. On the SPY page the rule ranks second of the 12 templates by CAGR, behind the weekly 7% target at 13.88%, which carries a 24.1% drawdown against this rule's 12.69%.
The headline run has no fees or slippage. With 5 basis points per trade the CAGR falls to 9.98%, and with 10 it falls to 6.75%. That is the main limit on this result and the costs section covers it.
The test is one backtest on one fund and one window. SPY rose in five of the six calendar years and fell in 2022. The rule's behaviour in a window with a different mix of years could differ.
Year by year
| Year | RSI(2) snapback | buy & hold |
|---|---|---|
| 2021 | 26.6% | 29.6% |
| 2022 | −2.5% | −17.8% |
| 2023 | 8.7% | 25.5% |
| 2024 | 22.1% | 24.3% |
| 2025 | 14.6% | 17.4% |
| 2026 | 8.3% | 13.6% |
Why the rule lost ground in the strong years and gained in 2022
The rule returned 26.6% in 2021, negative 2.5% in 2022, 8.7% in 2023, 22.1% in 2024, 14.6% in 2025 and 8.3% in 2026 to date. Buy-and-hold returned 29.6%, negative 17.8%, 25.5%, 24.3%, 17.4% and 13.6% in the same years. The rule beat holding in one year only, 2022, by 15.3 points. It trailed in the other five, by 3 points in 2021, 16.8 in 2023, 2.2 in 2024, 2.8 in 2025 and 5.3 in 2026.
The 2022 result comes from the rule's structure. It buys after a sharp multi-day fall and exits after a rebound, so it is invested only about a third of the time and sits in cash through the slow part of a decline. A fall of 17.8% for the fund became a loss of 2.5% for the rule. The rule made 34 round trips in 2022 and won 18, a win rate far below the 71% for the whole window. Of all six years it was the hardest on the entries. The losing trades cluster in the decline, and the sections on trades show the three worst.
The 2023 gap of 16.8 points is the largest shortfall. The fund returned 25.5% and the rule 8.7%. The rule made 30 round trips and won 19 that year. A steady rise gives a dip rule few oversold readings, and each one captures a few days of a long climb. Holding keeps the entire climb.
In 2021, 2024 and 2025 the rule captured most of the fund's return. It made 26.6% against 29.6% in 2021 with 29 round trips and 23 wins, 22.1% against 24.3% in 2024 with 29 round trips and 24 wins, and 14.6% against 17.4% in 2025 with 24 round trips and 20 wins. In those years most trades won and the rule gave up only a few points. The year that cost the most was the one with a steady climb.
In 2026 it made 8.3% against 13.6% with 26 round trips and 18 wins. The year is partial and runs to 2026-10-02, and the final trade, entered on 2026-09-30, was still open on the last day with a return of 0.41%.
Month by month
| Year | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2021 | 0.2% | 2.5% | 4.9% | 2.5% | 4.3% | 0.8% | 1.6% | 2.2% | −1.0% | 4.3% | −2.1% | 4.1% |
| 2022 | −3.0% | −0.3% | −0.1% | −0.9% | −1.2% | −6.4% | 5.5% | 1.9% | −6.1% | 8.2% | 2.6% | −1.7% |
| 2023 | 3.3% | −1.6% | −0.2% | 2.6% | 2.5% | 2.3% | 2.1% | −0.8% | −1.7% | −2.4% | 2.4% | 0.2% |
| 2024 | 2.3% | 5.2% | 3.0% | 1.4% | 1.6% | 0.2% | 0.3% | 4.1% | −1.1% | 0.7% | 2.8% | −0.1% |
| 2025 | 2.8% | −0.7% | −6.0% | 9.6% | 2.7% | 1.3% | −0.3% | −0.4% | 2.5% | 2.5% | 0.1% | 0.3% |
| 2026 | 0.4% | −1.1% | 1.6% | 1.4% | 1.9% | 0.0% | 0.8% | 0.8% | 1.2% | 0.9% | – | – |
What the monthly returns show
The monthly table shows the rule's worst months were 2022-06 at negative 6.41%, 2022-09 at negative 6.13% and 2025-03 at negative 5.95%. Its best were 2025-04 at 9.56%, 2022-10 at 8.23% and 2022-07 at 5.53%. In buy-and-hold the worst month was 2022-09 at negative 9% and the best was 2026-04 at 10.33%.
The pairing of 2025-03 and 2025-04 is the clearest example in the data. The rule lost 5.95% in March 2025, bought the washout, and made 9.56% in April, its best month. That is what the rule is built to do, and it is also why the worst month and the best month sit next to each other. A rule that buys under RSI 10 will often buy during a fall that continues for a while before it snaps back.
The same pattern appears in 2022. June and September were deep losing months, at negative 6.41% and negative 6.13%, and each was followed by a strong month, with July at 5.53% and October at 8.23%. Between them, August returned 1.93%. Eight of the twelve months of 2022 were negative, with the smaller losses in January through May at negative 2.99%, negative 0.3%, negative 0.15%, negative 0.92% and negative 1.23%, and December at negative 1.66%. That is a year in which entries kept meeting new lows.
In the rising years the monthly table is mostly positive. 2021 had ten positive months, and the only two losses were 2021-09 at negative 0.98% and 2021-11 at negative 2.09%. The strongest run was 2021-02 to 2021-05 with 2.46%, 4.87%, 2.45% and 4.32%. In 2024 the only losses were 2024-09 at negative 1.15% and 2024-12 at negative 0.05%, and the strongest months were 2024-02 at 5.21% and 2024-08 at 4.06%.
Monthly figures are for the rule's account, which was in cash for most days. A month with little signal shows a return near zero, as in 2026-06 at 0.01% and 2024-06 at 0.21%. The partial month 2026-10 shows 0.89% through 2026-10-02.
Every trade
RSI(2) snapback on SPY made 172 closed round trips and one position still open at the end of the test, an average hold of 4 days, an average winner of 1.16%, an average loser of −1.30%, a profit factor of 2.18, a longest losing streak of 3. It held a position at the close on 36.2% of trading days.
Best 10 round trips
| Entry | Entry price | Exit | Exit price | Return | Days held |
|---|---|---|---|---|---|
| 2025-04-07 | $480.27 | 2025-04-10 | $524.28 | 9.2% | 3 |
| 2024-08-05 | $497.80 | 2024-08-09 | $516.16 | 3.7% | 4 |
| 2022-10-03 | $342.88 | 2022-10-05 | $354.82 | 3.5% | 2 |
| 2022-07-27 | $373.12 | 2022-07-29 | $385.39 | 3.3% | 2 |
| 2022-10-21 | $345.70 | 2022-10-24 | $356.69 | 3.2% | 3 |
| 2022-11-29 | $375.92 | 2022-12-01 | $387.70 | 3.1% | 2 |
| 2023-03-02 | $374.79 | 2023-03-06 | $386.28 | 3.1% | 4 |
| 2024-11-01 | $558.47 | 2024-11-06 | $574.66 | 2.9% | 5 |
| 2024-02-21 | $479.84 | 2024-02-23 | $493.20 | 2.8% | 2 |
| 2021-05-19 | $377.75 | 2021-05-21 | $387.61 | 2.6% | 2 |
Worst 10 round trips
| Entry | Entry price | Exit | Exit price | Return | Days held |
|---|---|---|---|---|---|
| 2025-02-24 | $590.51 | 2025-03-18 | $553.27 | −6.3% | 22 |
| 2022-09-15 | $371.68 | 2022-09-29 | $348.37 | −6.3% | 14 |
| 2022-06-10 | $371.77 | 2022-06-16 | $348.80 | −6.2% | 6 |
| 2022-01-18 | $431.05 | 2022-01-31 | $414.04 | −4.0% | 13 |
| 2023-03-08 | $380.36 | 2023-03-15 | $368.32 | −3.2% | 7 |
| 2022-04-22 | $411.13 | 2022-04-29 | $398.70 | −3.0% | 7 |
| 2023-09-19 | $426.85 | 2023-09-29 | $415.83 | −2.6% | 10 |
| 2022-08-30 | $381.86 | 2022-09-08 | $373.11 | −2.3% | 9 |
| 2023-10-19 | $415.81 | 2023-10-25 | $406.60 | −2.2% | 6 |
| 2022-12-05 | $383.60 | 2022-12-09 | $375.28 | −2.2% | 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.
What the individual trades looked like
The median trade returned 0.42% and the median hold was 4 days. The average win was 1.16% and the average loss was 1.3%, so the win rate of 71% does most of the work. The profit factor of 2.18 says gross wins were more than twice gross losses over the window. The longest winning streak was 16 trades and the longest losing streak was 3. The shortest hold was 1 day and the longest was 22.
The first trade entered on 2021-01-13 at an adjusted price of 351.06 and exited the next day at 352.54 for 0.42%. That is typical: a one- or two-day bounce worth under half a percent. The prices in the trade list are adjusted for splits and dividends, so they are lower than the quotes printed at the time.
The best trade entered on 2025-04-07 at 480.27 and exited on 2025-04-10 at 524.28, a gain of 9.16% in 3 days. It sits at the centre of the April 2025 selloff and rebound, and it is the one trade where the dip rule caught a large part of a sharp reversal. The next best were 3.69% from 2024-08-05 to 2024-08-09, 3.48% from 2022-10-03 to 2022-10-05, 3.29% from 2022-07-27 to 2022-07-29 and 3.18% from 2022-10-21 to 2022-10-24. All five best trades came from rebounds after sharp lows in 2022, 2024 or 2025, and each lasted 2 to 4 days.
The worst trade entered on 2025-02-24 at 590.51 and exited on 2025-03-18 at 553.27, a loss of 6.31% over 22 days. It is also the longest hold in the sample. The rule bought after a sharp drop in February 2025 and the fund kept falling until the March low, and the rule only sold after RSI(2) recovered above 70. Two more of the worst five came in 2022: 2022-09-15 to 2022-09-29 for negative 6.27% over 14 days and 2022-06-10 to 2022-06-16 for negative 6.18% over 6 days. The fourth was 2022-01-18 to 2022-01-31 at negative 3.95% and the fifth was 2023-03-08 to 2023-03-15 at negative 3.17%.
All of the five worst trades were held longer than the typical 4 days. The rule has no stop, so a trade that does not rebound stays open until RSI(2) crosses 70, and the losing trades are the ones where that took the longest. The best trades were 2 to 4 days. The worst were 6 to 22. That asymmetry in time is the main risk of the rule on a fund that can fall for weeks.
Trade counts by year were 29 in 2021, 34 in 2022, 30 in 2023, 29 in 2024, 24 in 2025 and 26 in 2026, with wins of 23, 18, 19, 24, 20 and 18. The count of 173 trades in the list includes the open trade from 2026-09-30.
Largest drawdowns
| Peak | Low point | Depth | Days to low | Recovered | Days to recover |
|---|---|---|---|---|---|
| 2022-01-14 | 2022-07-14 | −12.7% | 181 | 2023-02-07 | 208 |
| 2025-02-21 | 2025-03-13 | −8.2% | 20 | 2025-04-09 | 27 |
| 2023-09-11 | 2023-10-27 | −6.2% | 46 | 2024-02-01 | 97 |
Buy-and-hold's deepest drawdown ran from 2022-01-03 to 2022-10-12 and reached −24.0%.
The three drawdowns
The deepest drawdown was 12.69%, from the peak on 2022-01-14 to the trough on 2022-07-14, and it recovered on 2023-02-07. It took 181 days to reach the trough and 208 more to recover. The buy-and-hold drawdown over the same period was 23.99%, from 2022-01-03 to 2022-10-12, and it recovered on 2023-12-13. The rule's trough came three months before the fund's, and its recovery came ten months before the fund's.
The rule's second drawdown was 8.21%, from 2025-02-21 to 2025-03-13, recovering on 2025-04-09. It took 20 days down and 27 to recover. This one matches the worst trade, the entry on 2025-02-24. Buy-and-hold fell 18.41% in the same episode, from 2025-02-19 to 2025-04-08. The third was 6.22%, from 2023-09-11 to 2023-10-27, with recovery on 2024-02-01.
All three drawdowns are smaller than the fund's own. The rule's drawdown was 11.3 points shallower than holding over the window, because it held cash for about two-thirds of the time and exited each rebound. The price it paid was the CAGR shortfall of 1.39 points. The Sharpe of 1.21 against 0.95 reflects the same exchange.
The test uses a fixed $10,000 sleeve and 98% of it per trade, and orders fill on minute bars.
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) | 13.2% | −12.7% | $20,342 | 1.21 |
| 5 basis points | 10.0% | −14.4% | $17,264 | 0.94 |
| 10 basis points | 6.8% | −16.2% | $14,547 | 0.66 |
How much the rule gives up to costs
The headline CAGR of 13.17% assumes no fees or slippage. At 5 basis points per trade the CAGR is 9.98%, the max drawdown is 14.39%, and the end equity is $17,264. At 10 basis points the CAGR is 6.75%, the max drawdown is 16.19%, and the end equity is $14,547. The Sharpe falls from 1.21 to 0.937 and then to 0.66.
The cost arrives because the rule makes 345 fills in the window. Each round trip pays the charge twice, on entry and on exit. With an average win of 1.16%, a charge of 5 or 10 basis points on each side takes a large share of each trade. At 10 basis points the rule is below the 14.56% for holding by a wide margin, and its Sharpe of 0.66 is below the 0.95 for holding.
SPY traded an average of $30,453,859,411 a day and had a median minute volume of 107,870 shares, so a small order should meet little market impact. The test cannot say how large the real cost would be, because it uses a flat charge and does not model spread or queue position.
The RSI(2) rule is the highest-turnover template in the library, and its cost runs show it. The monthly cycle at 12.84% and the EMA 12/26 trend rule at 10.2% with an 11.84% drawdown are the nearest templates on this fund by CAGR.
Changing the parameters
| Version | CAGR | Max drawdown | Round trips | Win rate | Final value |
|---|---|---|---|---|---|
| Published rules | 13.2% | −12.7% | 172 | 71% | $20,342 |
| RSI(2) < 5 / > 70 | 13.7% | −12.2% | 165 | 72% | $20,907 |
| RSI(2) < 15 / > 70 | 12.4% | −12.1% | 179 | 70% | $19,583 |
| RSI(2) < 10 / > 60 | 12.0% | −13.1% | 180 | 70% | $19,204 |
| RSI(2) < 10 / > 80 | 14.7% | −12.3% | 164 | 72% | $21,919 |
What changing the thresholds does
Four variants change one threshold each. Buying under RSI(2) of 5 instead of 10 gives a CAGR of 13.71%, a drawdown of 12.24%, a Sharpe of 1.271, and 165 trades with 118 wins. Buying under 15 gives 12.42%, 12.14%, a Sharpe of 1.132 and 179 trades with 125 wins. Selling above 60 instead of 70 gives 12.04%, 13.12%, 1.151 and 180 trades with 126 wins. Selling above 80 gives 14.65%, 12.29%, 1.303 and 164 trades with 118 wins.
The base rule with 10 and 70 made 13.17% with 172 round trips and a 12.69% drawdown. The table above shows three variants above it and one below on CAGR. The stricter entry at 5 made slightly more with fewer trades, since a deeper washout is a better entry. The looser entry at 15 made less, with more trades. The earlier exit at 60 made less, since it cut each rebound short. The later exit at 80 made the most, at 14.65%, which is above the 14.56% for holding.
The variant with an exit at 80 is the only one in the facts that beats buy-and-hold on CAGR, and it does so by a very small margin. It also has the highest Sharpe at 1.303 and a drawdown of 12.29%. The difference from the base rule comes from holding each winning trade slightly longer. It shows the direction of the sensitivity and says little about whether 80 is better than 70 in a different window. Choosing the best of four variants on the same data it was measured on overstates how well it would do on new data.
The variants span 12.04% to 14.65% on CAGR, and every variant has a drawdown between 12.14% and 13.12%. Trade counts range from 164 to 180. The result does not depend on a precise threshold. It does depend on the rule making many trades, which links it back to the cost runs. Each variant has about the same turnover and would face about the same cost drag.
The RSI mean reversion template on the same fund uses the 14-day RSI and made 4.37% a year with a 17.09% drawdown, which shows how much the shorter lookback matters on SPY.
How SPY behaved
| Measure | SPY |
|---|---|
| Data in this test | 2021-01-04 to 2026-10-02 (1444 sessions) |
| Total return, buy and hold | 125.3% |
| Annualized volatility | 16.4% |
| Deepest drawdown | −24.5% (2022-01-03 to 2022-10-12) |
| Up days | 54.4% |
| Average daily range | 1.15% |
| Average overnight gap | 0.44% |
| Correlation to QQQ | 0.94 |
| Correlation to TLT | 0.08 |
| Sessions above the 200-day average | 78.0% |
| Crossings of the 200-day average | 30 |
| Falls of 10% or more from a 20-day high | 11 |
Why SPY suited a short-horizon dip rule
SPY returned 125.29% over the window on its own price series, 15.2% a year, with annualized volatility of 16.41% and a max drawdown of 24.51% from 2022-01-03 to 2022-10-12. It recovered on 2023-12-13 and spent 488 sessions under a prior high. Calendar years were 30.46% in 2021, negative 18.16% in 2022, 26.18% in 2023, 24.86% in 2024, 17.72% in 2025 and 13.77% in 2026.
The fund closed above its 200-day average on 77.99% of sessions, so most of the window was an uptrend. A dip rule works best in an uptrend with brief pullbacks. The 2-day RSI closed under 10 on 146 sessions. After those sessions the median 5-day forward return was 0.89% against a baseline of 0.42% for any 5-day period, and the median 20-day return was 2.29% against 1.73%. The edge per signal is under half a percent over 5 days. The rule's 172 round trips are the same small edge taken many times, and the 71% win rate matches a fund that rises more often than it falls, with 54.4% up days.
The 14-day RSI closed under 30 on only 18 sessions, with a median 5-day forward return of 3.74% against 0.42%. The slower signal was a stronger bounce but arrived rarely. The 2-day signal came far more often, and the rule traded on it.
First-order autocorrelation was negative 0.02, nearly zero. The average up day was 0.74% and the average down day was negative 0.74%, with an average intraday range of 1.15%. Over the whole window the overnight share of log return was 62.46% and the intraday share was 37.54%. SPY had 11 drawdown events of 10% or more inside 20 days, spread over 23 sessions. Those are the sharp falls where the rule gets its entries and where it can also take a loss.
The largest days were 9.39% on 2025-04-09 and 5.48% on 2022-11-10. The worst were negative 5.98% on 2025-04-04 and negative 4.78% on 2025-04-03. The rule's best trade straddles those days: it entered on 2025-04-07 and exited on 2025-04-10 for 9.16%, catching the rebound that began on 2025-04-09.
How SPY compares with the other broad funds
Across the broad index group, the same rule produced a median CAGR of 13.17%, and SPY sits at exactly that figure. It ranks 16th of 59 funds for this strategy. The median across all 59 is 4.74%, so the rule worked much better on large liquid index funds than across the whole universe.
Within the group, VOOG made 15.89% with a 16.59% drawdown and 161 round trips, QQQM made 15.56% with a 14.25% drawdown, and QQQ made 14.17% with a 14.27% drawdown. VOO made 14.08% with a 12.64% drawdown and IOO made 14.04% with an 11.7% drawdown. SPY and VOO are nearly the same fund, with a correlation of 1, and each made 172 round trips, with results of 13.17% and 14.08%.
The weaker results are on funds with less trend or more noise. IWM made 11.38% with a 22.81% drawdown, QQQE made 9.49%, VOOV made 8.58%, VTV made 7.29%, and EEM made 3.87%. The growth and large-cap funds led, and the value, equal-weight and emerging-market funds trailed.
The drawdowns across the group are in a narrow band between about 11.7% and 22.81%, and SPY's 12.69% is near the low end. Round trips ranged from 153 to 173. The rule makes about the same number of trades on every fund in the group, so the cost drag shown for SPY would apply across it.
The rules
Buy extreme 2-day RSI washouts under 10, exit as soon as RSI(2) recovers above 70.
- WHEN the market opens · IF not invested AND RSI(2) < 10 · THEN buy with 98% of the sleeve
- 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.
How the two rules interact with SPY
The entry rule fires only when the strategy is flat, at the open after RSI(2) closes under 10. The exit rule fires only when invested, at the open after RSI(2) closes above 70. Because RSI(2) reacts to the last two closes, a two-day pullback is enough to trigger it, and a two-day bounce is enough to exit.
The median hold of 4 days and the median trade of 0.42% show that the typical trade is a quick bounce. The exit threshold of 70 is high for a 2-day RSI, which means the rule waits for a strong recovery and does not take a small one. That is why the longest hold was 22 days: a trade that does not rebound strongly stays open. A tighter exit, as in the variant that sells above 60, shortened trades and lowered the CAGR to 12.04%.
There is no stop. The fund's worst falls inside the window came in 2022 and early 2025, and the rule's worst trades came from the same periods. Its largest loss of 6.31% is far smaller than the 23.99% drawdown of holding, because the rule only holds about a third of the time. A rule that sells at the open after a signal also takes the gap between the signal close and the next open, and SPY's average overnight gap was 0.44%.
The dip buyer and the golden cross are two other templates on SPY, and they ranked twelfth and sixth by CAGR. The RSI(2) rule is the one on SPY that combines a high win rate with a shallow drawdown, at the cost of a lag to holding.
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 SPY?
Over 2021-01-04 to 2026-10-02, RSI(2) snapback on SPY returned 13.2% annualized vs 14.6% for buy-and-hold: it trailed buy-and-hold by 1.4% per year, with a maximum drawdown 11.3 points shallower than holding (12.7% vs 24.0%).
How many trades did it make?
172 completed round trips over 5.7 years (345 fills), with 71% 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.
Does the RSI(2) snapback beat buy-and-hold on SPY?
Not on return in this test. It made 13.17% a year against 14.56% for holding, and ended at $20,342 against $21,820. It had a 12.69% max drawdown against 23.99%, and a Sharpe of 1.21 against 0.95.
How often does the strategy win on SPY?
It won 122 of 172 round trips, a win rate of 71%. The average win was 1.16% and the average loss was 1.3%. The longest winning streak was 16 trades and the longest losing streak was 3.
How do trading costs affect the RSI(2) snapback on SPY?
At 5 basis points per trade the CAGR falls from 13.17% to 9.98%, and at 10 basis points to 6.75%. The rule makes 345 fills in the window, so costs accumulate. Its max drawdown also deepens, to 14.39% and 16.19%.
What was the worst trade?
The worst trade entered on 2025-02-24 at an adjusted price of 590.51 and exited on 2025-03-18 at 553.27, a loss of 6.31% over 22 days. The best trade returned 9.16% in 3 days, from 2025-04-07 to 2025-04-10.
What happens if I change the RSI thresholds?
Buying under 5 gave 13.71% and buying under 15 gave 12.42%. Selling above 60 gave 12.04% and selling above 80 gave 14.65%, which is above the 14.56% for holding. All four variants had drawdowns between 12.14% and 13.12%.
Why did the strategy do so well in 2022?
The fund fell 17.8% in the buy-and-hold record that year, while the rule lost 2.5%. It was invested about a third of the time and sat in cash through much of the decline. It made 34 round trips in 2022 and won 18.
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.