VV trading strategies, backtested

Vanguard Large-Cap ETF: broad US large-cap exposure beyond just the S&P 500. Every DeployQuant template run on VV over 5.7 years of minute data, same engine, same window, sorted by return.

VV is the Vanguard Large-Cap ETF, a broad US large-cap fund that reaches beyond the S&P 500. Between 2021-01-04 and 2026-10-02 it gained 121.23% in total return terms on its own price series, a CAGR of 14.83% with annualized volatility of 16.69%. Buy-and-hold in the strategy engine, starting with $10,000 and no fees, ended at $21,500 with a CAGR of 14.26% and a max drawdown of 25.21%.

Twelve rule sets were run on VV over the same window. One of them finished ahead of buy-and-hold on CAGR: the weekly 7% target at 14.94% against 14.26%. Eleven of the twelve had a shallower max drawdown than holding, and all twelve ended with a positive CAGR. The spread between the best and the worst is wide. The weekly target ended at $22,238, and the dip buyer ended at $11,622 after only 2 round trips.

The pattern on this fund is that the rules which stay invested most of the time keep up with holding, and the rules that wait for a signal fall behind. A fund that rose in five of the six calendar years and fell in one gives a waiting rule little to catch.

The method is the same for every strategy: daily decisions, orders on minute bars, no margin, and a headline run with no fees or slippage. Cost runs at 5 and 10 basis points are shown in the tables and discussed below.

How the twelve rule sets ranked on VV

The weekly 7% target buys at the first open of each week with 98% of the sleeve, rests a limit order 7% above the entry, and sells on Thursday afternoon if the trade is losing. On VV it made 83 round trips with a 30% win rate. That win rate looks poor, but the average win was 7.14% and the average loss was 1.56%, which gave a profit factor of 2.06. Exposure was 87.5%, so the rule was nearly always in the fund. It beat holding in 2022 by 6.5 points, in 2024 by 6.1 and in 2025 by 9.7, and trailed it in 2021, 2023 and 2026. The CAGR gap to buy-and-hold is small, and the drawdown of 21.53% is a few points shallower than the fund's 25.21%.

The RSI(2) snapback came second on a CAGR basis at 12.55%, with the best Sharpe ratio of the group at 1.12 and a max drawdown of 13.12%. It made 173 round trips, won 69% of them, held for 4.4 days on average and was invested 36.5% of the time. Its worst calendar year was 2022 at 2.5% negative, against 19.4% negative for buy-and-hold. The cost of that protection shows in 2023, when it returned 7.0% and the fund returned 26.8%. A short-horizon dip buyer has limited room in a year when the fund keeps rising.

The monthly cycle buys on the first session of a month and sells on the last. It was invested 95.2% of the time, made 69 round trips and ended at a CAGR of 12.44% with a 23.69% drawdown. It behaves like a slightly leaky version of holding, with a 2022 return of 21.6% negative.

The trend rules sit in the middle. The EMA 12/26 returned 9.84% with a 13.43% drawdown and the SMA 10/50 returned 8.18% with a 12.58% drawdown. The golden cross returned 8.43% on only 2 round trips. The 20-day momentum breakout returned 9.34% on 3 round trips. These rules had profit factors between 3.51 and 5.42 because they win rarely and win large, but they were out of the fund through part of every rally.

The bottom three are the 200-day regime filter at 7.86%, the trend plus trailing stop at 6.22% with a 29.58% drawdown, and the RSI mean reversion at 5.02% with a 79% win rate and a 17.66% drawdown. The trend plus trailing stop is the one rule whose drawdown was deeper than buy-and-hold, and the only one that beat it in none of the six years. The trailing-stop templates are backtest-only for now, so these two results describe simulated history only.

How each strategy traded VV

What the trade lists show

The trade counts split the group into three styles. The fast styles are the weekly target at 83 round trips, the RSI(2) snapback at 173 and the monthly cycle at 69. The slow styles are the golden cross at 2, the dip buyer at 2, the trend plus trailing stop at 3 and the momentum breakout at 3. The trend followers in between made between 13 and 20.

The slow styles depend on a handful of trades. The momentum breakout on VV had one loss of 8.22% from 2022-08-11 to 2022-09-06, a gain of 5.8% from 2022-11-14 to 2023-10-26, and a gain of 24.17% from 2023-11-27 to 2025-03-11. A fourth position was still open at the end of the window with a return of 40.32% since 2025-05-06. A result built on three closed trades says little about repeatability, and a different start date would change the picture.

The golden cross holds the longest trade in the set. Its entry on 2023-02-01 ran to 2025-04-17 for a gain of 35.67%, a hold of 806 days. Its only closed loss, 6.25%, came from the entry on 2021-10-19. The same date appears as the first entry of the 200-day regime filter and the trend plus trailing stop, which both lost on that trade: 5.3% and 9.88%. All three bought near the late-2021 high of 196.65 in adjusted prices and were stopped out in early 2022.

The 200-day regime filter shows what a long-horizon filter does in a choppy decline. In 2022 it made 6 round trips and won none of them. Its median trade was a loss of 1.02% with a median hold of 7 days, yet its best trade was a gain of 33.81% from 2023-11-02 to 2025-03-11. A rule that loses repeatedly in sideways markets and then holds one long winner has a low win rate of 31% and a profit factor of 3.10.

The RSI mean reversion tells the opposite story. It won 11 of 14 trades, and its worst trade was a loss of 8.69% from 2022-04-25 to 2022-07-11, a position held for months while the fund kept falling. The average loss of 5.53% was larger than the average win of 4.32%, so a high win rate still produced the lowest CAGR but one in the group.

The weekly target's worst trades cluster in 2022: 7.04% from 2022-05-23 to 2022-06-16 and 6.72% from 2022-05-02 to 2022-05-12. Its best trade, 8.01% from 2023-10-30 to 2023-11-14, is capped near the 7% target plus the gap at the open. The table in this section lists the dip buyer's worst trade as 8.0%, but both of its closed trades were gains of about 8%, so that cell is the exit target and not a loss.

Long holds, short bursts and the open trades

The templates at the top and bottom of the table differ in how long they held. Golden cross averaged 474.5 days a trade and made 2 round trips: a loss of 6.25% over 143 days from 2021-10-19, and a gain of 35.67% over 806 days from 2023-02-01 to 2025-04-17. The 200-day filter averaged 93.3 days, won 31% of 13 trades, and still had a profit factor of 3.1 because its average win of 13.59% was far larger than its average loss of 1.98%. Its best trade was 33.81% over 495 days from 2023-11-02. The RSI(2) snapback averaged 4.4 days a trade across 173 round trips, with an average win of 1.18% and an average loss of 1.27%, and it won 69%.

The weekly target had a profit factor of 2.06 with a 30% win rate. Its average win was 7.14% and its average loss 1.56%, and it held 20.8 days on average. Its best trade ran 15 days from 2023-10-30 for 8.01%, and its worst ran 24 days from 2022-05-23 for -7.04%. The result of 14.94% a year is the highest in the table, ahead of the RSI(2) snapback at 12.55% and the monthly cycle at 12.44%, but it came with a 21.53% drawdown from 2022-02-09 to 2022-06-16 that took until 2023-07-18 to recover.

The trend plus trailing stop rule shows the cost of a wide stop. It lost 9.88% over 128 days from 2021-10-19, then lost 12% over 47 days from 2022-03-23, and then held 855 days from 2022-12-01 for 33.46%. Its calendar 2022 was -28.1% against -19.71% for the fund, and its drawdown, 29.58%, is the deepest in the table. Its profit factor was 1.3 on closed trades. The rule is still long from 2025-05-13 with a gain of 34.33%.

Four templates carry an open trade at the end that is not in their win rates: golden cross from 2025-06-30 at 26.3%, the 200-day filter from 2026-04-09 at 15.06%, the trailing stop at 34.33%, and the weekly target from 2026-09-14 at 1.71%. The final values include these gains. A trade table that only lists closed trades understates how much of the return was still unrealised when the window stopped.

The fund's two largest days, 9.39% on 2025-04-09 and 5.57% on 2022-11-10, both came inside drawdowns. The worst day was -5.99% on 2025-04-04. A rule that exited on 2025-04-04 and re-entered the next week would have missed the rebound, and a rule that was already out of the fund before the 2025-04-09 jump would have missed the largest single day of the window.

How VV behaved

Calendar years

Biggest single days

Average return by calendar month

Most and least correlated funds

Why the fund's behaviour suits some rules and not others

VV closed up on 53.98% of sessions. The average up day was 0.75% and the average down day was 0.75% negative, so the edge comes from the frequency of up days and not from their size. Lag-1 autocorrelation of daily returns is 0.01 negative, which is effectively zero. A fund with no day-to-day persistence gives little to rules that expect a move to continue the next day, and little to rules that expect it to reverse.

The fund was above its 200-day average on 76.79% of sessions and crossed it 26 times. That share explains why trend rules were invested 64% to 73% of the time and still lagged: they sat out the sessions below the average, and some of those sessions fell inside rallies. The crossings are the whipsaws, and each one is a chance for a trend rule to buy late or sell early. The two trend templates with shorter averages, the EMA 12/26 and the SMA 10/50, made 20 and 15 round trips, and their profit factors of 4.46 and 4.74 show that the winners were several times larger than the losers even though only 55% and 53% of trades won. The 200-day regime filter made 13 round trips and the golden cross made 2, which shows the same filter family can produce very different trade counts depending on the signal.

The largest calendar-year gains were 28.86% in 2021, 27.31% in 2023 and 25.11% in 2024. The one losing year was 2022 at 19.71% negative. In the maximum drawdown, the fund fell 25.63% on its own price series from the 2021-12-27 peak to the 2022-10-12 trough and took until 2023-12-13 to recover, a span of 493 sessions. Every rule that beat holding in 2022 did so by being out of the fund for part of that fall. The RSI(2) snapback lost 2.5% that year, the EMA 12/26 lost 8.9% and the SMA 10/50 lost 10.2%.

The best day was 2025-04-09 at 9.39% and the worst was 2025-04-04 at 5.99% negative, with 2025-04-03 close behind at 4.87% negative. April 2025 therefore holds both tails. The RSI(2) snapback booked its best trade, 9.19%, from 2025-04-07 to 2025-04-10, which straddles the best day. The 2022 days dominate the rest of the worst list: 2022-09-13 at 4.39% negative, 2022-05-18 at 4% negative and 2022-06-13 at 3.94% negative.

The overnight gap averaged 0.44%, and the fund's overnight log return was 60.67% against 16.99% for the intraday portion, an overnight share of 78.12%. Rules that buy at the open and sell at the open collect both pieces. Rules that exit on Thursday at 2:00pm, as the weekly target does for losing trades, hand some overnight return to the days they sit out.

On oversold readings, RSI(14) fell below 30 on only 19 sessions in the window, and the median forward 5-day return after those was 3.66%, against a baseline median of 0.46%. The RSI(2) below 10 reading occurred on 148 sessions, with a median forward 5-day return of 0.91% and a 20-day median of 2.4%. The 20-day baseline median is 1.75%. These are the facts behind the RSI results: the short-horizon reading fires often enough to build 173 trades, while the 14-day reading fires rarely and produced 14 trades with a long hold.

VV correlates at 1.00 with SPY and with VOO, and at 0.95 with QQQ. Its beta to SPY is 1.02. Results here should therefore read as results for large-cap US equities as a group. Other funds in the category show how the same rules did elsewhere: SPY's best rule was the weekly 7% target at 13.88%, VOO's was the RSI(2) snapback at 14.08%, and IWM showed the largest lift from a rule over holding, at 11.38% against 7.5%.

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.