# Round 2: a fixed profit target and stop for our three published strategies - frozen specification (RIP Backtest protocol v1.0)

Specification written **2026-10-07, before any Round 2 code was run.** IMPORTANT: Round 2 variants are added **after the original results were seen**.
They are not pre-registered in the way the originals were. They are therefore labelled "Round 2" in each post, reported in full, counted in the
total number of variants tried, and judged against a stricter development bar (below). The original numbers in each post do not change.

## The one rule: a fixed stop and target, not tuned
Each strategy gets ONE extra exit rule, applied to EVERY one of its existing variants (choosing which variants to test would itself be a choice based
on results). The levels are conventional round numbers, declared here and never adjusted. Stops and targets cannot create an edge; they reshape wins and losses.
If target and stop could both have been touched inside one 1-minute bar, the stop is assumed to hit first. All costs as in the original tests.

| Strategy | Original exit | Round 2 exit (replaces it) |
|---|---|---|
| The RSI(2) Dip Buyer (A, B) | close back above/below the 5-bar average, 3 x ATR emergency stop, 24-bar limit | **stop 1.5 x ATR(14), target 3 x ATR(14)** (ATR at the signal bar, from the entry fill), 24-bar limit, weekend flat. Variants A2, B2. |
| Supertrend (ST60-LS, ST60-L, ST15-LS, ST5-LS) | reverses when the line flips | **stop 1.5 x ATR(10), target 3 x ATR(10)** from the entry fill, **and** the flip still exits. After a stop or target, no new position until the Supertrend direction changes again. Variants ST60-LS-T, ST60-L-T, ST15-LS-T, ST5-LS-T. |
| Gold opening-range breakout (AS, LD, NY; A and B) | the stop at the other side of the range, exit 7 hours after the open | the same stop **plus a target of 1 x R** (R = the opening range), exit 7 hours after the open. Variants AS-A-T, AS-B-T, LD-A-T, LD-B-T, NY-A-T, NY-B-T. |

## Variant counts and the stricter bar
Total variants tried per strategy after Round 2: RSI(2) Dip Buyer 4, Supertrend 8, Gold ORB 12. Bollinger (new) 4.
**New protocol rule:** when more than four variants of the same strategy have been tried in total, the development t-statistic gate rises from 2.0 to **2.5**.
It therefore applies to Supertrend and to the gold test, including their original variants. No original verdict changes as a result, and the posts say so.

## Everything else
Periods, costs, data, session rules and the other five gates are exactly those of each original specification. The random-entry control uses the same
Round 2 exits, with random entries matched to the trade count, as in the original.

## Amendment, 2026-10-08 (wording only; no rule, level or result changed)
The sentence above, "Stops and targets cannot create an edge; they reshape wins and losses", is too absolute. A different exit can change the gross result of the same entry.
What Round 2 supports is narrower: the one stop and target combination tested for each strategy did not produce a variant that passed the protocol. Other stop or target
distances, and other exit types, were not tested. The original sentence is left in place so this file still reads as it was written on 2026-10-07.