Two tests at the VWAP band
FailedNQ 9 · ES 9 /100VWAP from 18:00 with its 2-sigma bands, on 5-minute bars. The first bar of the day that touches the lower band is the primary test. A later bar with its low within 10 points of the primary test's low, on volume under each of the prior two bars, is the secondary test. If the next bar closes back toward VWAP, past the signal candle's high by a tick or more, on volume over 1.1 times the signal candle's, buy at the next bar's open. Stop a tick past the lowest low of the two tests; skip it if the stop is bigger than the distance to VWAP. Target 2R; stop to breakeven when VWAP trades. Shorts the mirror at the upper band. Tests and trigger between 9:30 and 15:00, flat at 15:30, one trade a day, every weekday.
A rule sent by a reader of the shelf's data post and written down line by line with him before it ran; one rule, two markets, two records. The Wyckoff reading it sits inside is his and is not in the test.
Real, or luck?
NQ. Not real, or not worth its costs. 1,047 trades over 18.6 years. Net after costs: -$50,282. On the 15 years it never saw: -$31,691, and 4,945 of 5,000 coin flips did as well or better; 2 of those 15 years made money.
ES. Not real, or not worth its costs. 1,510 trades over 18.6 years. Net after costs: -$71,892. On the 15 years it never saw: -$54,605, and 4,934 of 5,000 coin flips did as well or better; 0 of those 15 years made money.
Where the money was
From the two records, not from memory.
NQ. The stop is the rule: 668 of 1047 exits were the stop (-$129,100.60), and the trades that were not stopped made $78,818.20. Longest wait for a new equity high: 224 months, median year -$2,494.20. Win rate 26.6% (278 of 1,047 trades closed above zero after costs); -$48.03 a trade.
ES. The stop is the rule: 942 of 1510 exits were the stop (-$194,306.40), and the trades that were not stopped made $122,414.40. Longest wait for a new equity high: 224 months, median year -$3,678.00. Win rate 25.2% (380 of 1,510 trades closed above zero after costs); -$47.61 a trade.
| exit | trades | net |
| the stop | 668 | -$129,101 |
| the 2R target | 221 | $75,397 |
| breakeven after VWAP traded | 84 | -$773 |
| flat on the clock | 74 | $4,194 |
| exit | trades | net |
| the stop | 942 | -$194,306 |
| the 2R target | 299 | $116,594 |
| breakeven after VWAP traded | 135 | -$2,280 |
| flat on the clock | 133 | $8,304 |
| stop and target in one bar | 1 | -$204 |
The breakeven, the stop moved to entry once VWAP traded, ended 84 trades on NQ and 135 on ES, beside the clock's 74 and 133: most trades are stopped before VWAP is reached.
The years from 2023 on, the seen era on every page here, are its worst per trade: NQ -$177.06 a trade from 2023 on (105 trades) against -$33.64 on the years before (942 trades); ES -$73.56 a trade from 2023 on (235 trades) against -$42.83 on the years before (1,275 trades).
Can you live on it?
Passing and being able to hold it are different things. These numbers are the difference.
NQ. Longest wait for a new high: 224 months, ending 2026-08. Worst twelve months: -$10,818, ending 2026-03. A typical year: -$2,494. One contract needs about $55,088 of room to get through its worst stretch (that stretch, with the record's own safety margin).
ES. Longest wait for a new high: 224 months, ending 2026-08. Worst twelve months: -$9,156, ending 2021-02. A typical year: -$3,678. One contract needs about $76,201 of room to get through its worst stretch (that stretch, with the record's own safety margin).
What it costs
Every trade above already paid one tick each way plus commission.
NQ. With costs doubled: -$59,167. Tripled: -$68,432. It stops making money at 0.0 ticks a side.
ES. With costs doubled: -$107,430. Tripled: -$147,017. It stops making money at 0.0 ticks a side.
The ten points
One setting was never settled, and it is the reader's.
The secondary test counts when its low is within 10 points of the primary test's. The number is the reader's, in his words: "the secondary test doesn't have to be at the same exact price level it could be a few point before or after... let's test with +/- 10 points of the primary test and see what comes back." Whether NQ, at its higher price, needs a wider one was asked ("10 points on nq too or wider there") and not answered, so 10 ran on both markets. It is the one setting on this page that is the reader's to change; a wider test on NQ is a new run with its own record.
If you trade this by eye
The level did not hold: 64% of trades on NQ and 62% on ES ended at the stop, which sits where the setup says price should not go; the setup's premise failed more often than it held.
Against 5,000 coin flips the written rule is a coin (4,945 of 5,000 matched or beat it on the unseen years on NQ, 4,934 on ES); whatever you do on top of the written steps is where any edge is, and your fills are how to find out which part.
This is the VWAP-band version of the setup on its own. The Wyckoff half, the Phase C structure and the higher-timeframe wall, is read by eye and was not in the test; the exchange said so before it ran.
That is what the record says about the rule as written. It is not what next year says.
The score, in six parts
NQ 9 · ES 9 of 100. Each part is a question; the number is how well each record answered it.
NQ 5 · ES 5
NQ 0 · ES 0
NQ's own line: Sharpe retention 0.0 (per-trade) -> 0; absolute OOS -> 0
ES's own line: Sharpe retention 0.0 (per-trade) -> 0; absolute OOS -> 0
NQ 0 · ES 0
NQ's own line: windows profitable 0.13 -> 0; efficiency 1.14 -> 0; window concentration -> 0
ES's own line: windows profitable 0.0 -> 0; efficiency 0.95 -> 0; window concentration -> 0
NQ 4 · ES 4
NQ's own line: null p 0.989 (5000 seeds) -> 0; DD95/BT 1.1 -> 4; top trade None -> 0
ES's own line: null p 0.9868 (5000 seeds) -> 0; DD95/BT 1.06 -> 4; top trade None -> 0
NQ 0 · ES 0
NQ's own line: retention@1x 0.0 -> 0; PF@2x 0.55 -> 0
ES's own line: retention@1x 0.0 -> 0; PF@2x 0.51 -> 0
NQ 0 · ES 0
NQ's own line: plateau 0.0 -> 0; parsimony 8 -> 0; sign flip True -> 0
ES's own line: plateau 0.0 -> 0; parsimony 8 -> 0; sign flip True -> 0
The code
The code is withheld until every gate is cleared; the ones it missed are marked.
- ✗ it passed
- ✗ fewer than 1 in 20 coin flips matched it on the years it never saw
- ✗ the money on those years is clearly more than zero
- ✗ it still makes money with at least four ticks of cost a side
- ✗ no single trade is more than a quarter of the money
How it was scored: Scored under Protocol v1.1 on FirstRate 1-min unadjusted via Thinridge engine. Intake testimony: NOT received (rule tested as locked). Contamination map sworn: no (seen era as declared or by archive date). Human review of reconciliation residue: NO.