All issues

Search results

walk-forward fold_actual carries no trade count — the 'fewer positions vs better picks' question is unanswerable from a surface

Status: Open.

#2412 In dayfine/trading;

· dayfine opened on Aug 19, 2026

[backlog] Candidate archetype taxonomy — track safer vs bigger-swing candidates separately

Status: Open.

#2409 In dayfine/trading;

· dayfine opened on Aug 19, 2026

[backlog] Surface stop_anchor_at_entry_base x initial_stop_buffer

Status: Open.

#2408 In dayfine/trading;

· dayfine opened on Aug 19, 2026

Base-broken cancel: the structural test that age and stage-flicker both miss

Status: Open.

#2407 In dayfine/trading;

· dayfine opened on Aug 19, 2026

H1: the 808s per-run floor is noise-dominated — profile the two stable costs instead

Status: Open.

#2406 In dayfine/trading;

· dayfine opened on Aug 19, 2026

Re-flip entry_order_max_rest_weeks to 26 once goldens are re-pinned to live

Status: Open.

#2405 In dayfine/trading;

· dayfine opened on Aug 19, 2026

entry_extension_max_pct conflates two different rules — and the live 15% has no derivation

Status: Open.

#2404 In dayfine/trading;

· dayfine opened on Aug 19, 2026

Goldens should track the live config by default, and declare their deviations

Status: Open.

#2403 In dayfine/trading;

· dayfine opened on Aug 19, 2026

Long cascade diagnostics have no RS phase, so an armed min_rs_normalized shows up as grade drops

Status: Open.

#2394 In dayfine/trading;

· dayfine opened on Aug 19, 2026

Goldens that gate behaviour run post-merge only, so a config-default PR gets no signal from them

Status: Open.

#2393 In dayfine/trading;

· dayfine opened on Aug 19, 2026

weinstein_strategy.mli documents stale_exit_after_days default as None; shipped default is Some 5

Status: Open.

#2388 In dayfine/trading;

· dayfine opened on Aug 19, 2026

QC agents build the PARENT tree, not the PR — two project rules contradict each other

Status: Open.

#2386 In dayfine/trading;

· dayfine opened on Aug 19, 2026