The Test Bench › VWAP › Study 011
The long-only VWAP rule makes about 3.9 round trips a day on QQQ, each a buy and a sale. We measured how much each buy and sale can cost before the rule falls behind buying and holding.
STUDY 011 · EXPLORATORY COST ANALYSISOn QQQ from May 2018 to September 2023, with no trading costs, the long-only VWAP rule from Study 010 finished $4,761 ahead of buying and holding. That lead came from 5,284 round trips: about $0.90 per round trip.
The lead disappeared once each buy and each sale cost about 0.2 cents per QQQ share (0.0008% of the trade value). At that cost, the charges on all 5,284 round trips add up to the lead. It is less than half of the narrowest quote QQQ could have in those years.
On 149 other large US stocks in 2026, the rule beat buying and holding on 46 with no costs, and on only 14 at 0.02% on each buy and each sale.
This is an exploratory analysis: its cost levels were chosen after Study 010's results were known.
| Cost on each buy and sale (QQQ, May 2018 to September 2023) | $25,000 with the VWAP rule | Buying and holding |
|---|---|---|
| None | $60,475 | $55,714 |
| 0.0002% (the study’s assumption) | $59,306 | $55,714 |
| 0.0018% (half the narrowest quote) | $49,752 | $55,712 |
| 0.02% (assumed in Study 010) | $7,305 | $55,692 |
These are historical analyses. They are neither advice nor forecasts. A backtest describes what a rule would have done over a past window. It does not say what the rule will do next.
Exploratory cost analysis. We re-priced the trades of the VWAP strategy tested in Study 010. The cost levels and the break-even were chosen after Study 010's results were known, which makes this an exploratory sensitivity analysis.
The rule is Study 010's, unchanged: the long side of the published VWAP strategy, on 5-minute candles, buying at the close of a candle that closes above VWAP and selling at the close of one that closes below it or at 16:00, with the whole account in every trade. How VWAP and Anchored VWAP work explains the indicator. On QQQ from May 2018 to September 2023 it made about 3.9 round trips a day.
Trading costs can include the bid-ask spread, commissions, regulatory fees, slippage and currency conversion. This study models them as one percentage of each trade's value, charged on every buy and every sale (“per side”). The fills are Study 010's: each order fills at the close of the candle that signals it, an idealised assumption, so the break-evens below are the rule's best case.
The same percentage is charged on every buy and every sale. Because the rule moves the whole account in and out of the position thousands of times, the cost compounds across the trades. On QQQ over May 2018 to September 2023, the rule's final balance equals buying and holding's at 0.0008% per side, about 0.2 cents a share: a cumulative break-even, where the charges on all 5,284 round trips use up the $4,761 lead. Against holding each session the break-even is 0.0057%. From September 2023 to September 2026 the rule was behind buying and holding QQQ with no costs, so no cost level puts it ahead.
Method. With the whole account in every trade, a cost c per side scales each round trip by (1 − c) ÷ (1 + c), so a run's final value at any cost follows from its result with no cost and its number of round trips. This reproduces Study 010's own engine runs at 0.02% and 0.04% per side for QQQ and every stock, with the same trade counts. Buying and holding pays the cost on one purchase and one sale.
The figures below are reference points from published sources, listed from smallest to largest. A trader's actual costs depend on the broker and the account. The spread figures are the minimum possible quoted spreads; real spreads, slippage, commissions and market impact can be larger. The costs add up: one trade can pay several of them.
| Cost | Per buy and per sale | $25,000, VWAP rule | Holding each session | Buying and holding | Stocks beating buying and holding, of 149 |
|---|---|---|---|---|---|
| The study’s assumption | 0.0002% | $59,306 | $38,471 | $55,714 | 44 |
| Half a $0.005 quote, 2026 prices | 0.0004% | $58,135 | $38,276 | $55,714 | 43 |
| US regulatory fee on sales | 0.001% | $54,238 | $37,603 | $55,713 | 43 |
| Half a $0.01 quote, 2018–2023 prices | 0.0018% | $49,752 | $36,783 | $55,712 | 40 |
| Assumed in Study 010 | 0.02% | $7,305 | $22,531 | $55,692 | 14 |
| Currency conversion, lowest published | 0.03% | $2,539 | $17,200 | $55,681 | 9 |
| Currency conversion, highest published | 1.5% | $0 | $0 | $54,068 | 0 |
With no costs the rule beat buying and holding on 46 of 149 stocks over April 2026 to September 2026: 40 remained at half a 1-cent quote and 14 at 0.02% per side. For the 46 stocks ahead with no costs, the median break-even was 0.0129% per side. Study 010 describes the limits of this six-month, selected list. Study 005 reports an opening range breakout on the same stocks and dates at four cost levels.
An account held in pounds that converts to dollars on each purchase, and back on each sale, pays a conversion fee each time. Published fees for UK accounts run from 0.03% to 1.5% per conversion. At 0.03%, 9 of 149 stocks still beat buying and holding; at 1.5%, none did. An account that holds dollars avoids the fee.
The rule's lead over buying and holding QQQ in the original study's years was small relative to the number of trades: $4,761 over 5,284 round trips. A cost of only 0.2 cents a share on each buy and each sale (0.0008% of the trade) was enough to remove the entire lead, less than half the minimum quoted spread. Across the stocks tested, 0.02% per side left 14 of 149 ahead of buying and holding.
Fee figures are as their sources published them.
Using this research. The analysis, tables and figures on this page are GU Analyser Ltd’s own work, published under a Creative Commons BY-NC 4.0 licence. You are welcome to quote it, cite it and build on it, with credit and a link. Suggested credit: GU Analyser, The Test Bench, Study 011, linking to guanalyser.com/test-bench/vwap-strategy-trading-costs/. For commercial use, ask us at hello@guanalyser.com.
Rebuild the test
Backtest the VWAP rule on 5-minute candles and set your own cost on each buy and each sale. Running the same strategy at several costs shows how much of its result the costs decide.
Create a free account and build your own strategies → or read how VWAP and the VWAP rule work →The method: how a backtest is built