okx-codex-trader/reports/eth-exploration/eth-bb-squeeze-t-gate-robustness-report.md

ETH BB squeeze T-gated robustness validation

Scope: adjacent variants around the current best T-gated BB squeeze candidate. No live executor changes. Cost model: maker_taker.

Output files:

• reports/eth-exploration/eth-bb-squeeze-t-gate-robustness-detail.csv
• reports/eth-exploration/eth-bb-squeeze-t-gate-robustness-score.csv
• reports/eth-exploration/eth-bb-squeeze-t-gate-robustness-report.md

Candidate score: | reentry_bars | middle_exit_buffer_pct | middle_exit_confirm_bars | 10y_return | 10y_calmar | 180d_return | 180d_mdd | 90d_return | 90d_mdd | positive_recent_windows | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | 96 | 0 | 1 | 4.33177 | 1.1984 | 0.237966 | 0.267407 | 0.21666 | 0.189711 | 2 | | 192 | 0 | 1 | 4.55563 | 1.23198 | 0.235549 | 0.267407 | 0.213983 | 0.189711 | 2 | | 48 | 0 | 1 | 7.66686 | 1.5778 | -0.175767 | 0.243728 | 0.221521 | 0.186473 | 1 | | 48 | 0.001 | 1 | 3.66416 | 1.10985 | -0.181163 | 0.247998 | 0.266474 | 0.164032 | 1 |

Recent horizon detail: | horizon | reentry_bars | middle_exit_buffer_pct | trades | reentry_entries | net_total_return | net_max_drawdown | net_calmar | worst_month | worst_month_return | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | 180d | 96 | 0 | 57 | 5 | 0.237966 | 0.267407 | 2.02566 | 2026-05 | -0.110546 | | 180d | 192 | 0 | 62 | 10 | 0.235549 | 0.267407 | 2.00285 | 2026-05 | -0.110546 | | 180d | 48 | 0 | 20 | 0 | -0.175767 | 0.243728 | -1.33049 | 2025-12 | -0.0980951 | | 180d | 48 | 0.001 | 19 | 0 | -0.181163 | 0.247998 | -1.34363 | 2025-12 | -0.0967073 | | 90d | 48 | 0.001 | 25 | 0 | 0.266474 | 0.164032 | 9.7948 | 2026-04 | -0.0853657 | | 90d | 48 | 0 | 26 | 0 | 0.221521 | 0.186473 | 6.7103 | 2026-05 | -0.110546 | | 90d | 96 | 0 | 27 | 1 | 0.21666 | 0.189711 | 6.40539 | 2026-05 | -0.110546 | | 90d | 192 | 0 | 31 | 5 | 0.213983 | 0.189711 | 6.30157 | 2026-05 | -0.110546 |

Verdict:

• Most stable adjacent candidate: reentry_bars=96, middle_exit_buffer_pct=0, middle_exit_confirm_bars=1.
• Recent performance check: 180d 23.80%, 90d 21.67%; max recent MDD 26.74%.
• Small live executor readiness should require positive 90d and 180d behavior across adjacent variants, not only the 10y top row.