min_rr = 2.0 was hand-set in Admin (2026-06-24) and never swept — the gate
ablation only tested the floor on-vs-off, never its level. It was the last
un-swept knob in the live gate.
Swept against portfolio Sharpe under the real exit, with a parity self-check
(reproduces_production_gate: the row at the live floor must rebuild production's
exact 1,089-setup qualified set — it does).
min_rr qualified in-sample Sh/CAGR OOS Sh/CAGR (entries >= 2024-07)
0.0 6636 1.98 / 58.5% 2.02 / 66.2%
1.2 3897 1.34 / 33.9% 1.12 / 28.8%
1.5 3127 1.20 / 29.6% 1.12 / 28.8%
1.75 1974 1.64 / 44.5% 1.15 / 27.4%
2.0 (live) 1089 2.04 / 50.4% 2.78 / 73.3%
2.25 577 1.64 / 31.8% 1.71 / 31.9%
2.5 286 1.67 / 29.0% 0.68 / 8.7%
KEEP 2.0. It is the optimum in both windows, and a peak that reproduces in data
it was never fitted to is real evidence. But treat it as fragile: unlike the ATR
trail (a plateau), this is a spike with a trough beside it — +/-0.25 costs ~0.4
Sharpe in-sample and ~1.6 out-of-sample — and the curve is bimodal (floor-off is
good, 1.2-1.75 is bad, 2.0 is good). The hand-set value landed on the peak by
luck, not by tuning. Do not nudge it.
Worth knowing: turning the floor OFF entirely is the second-best row in both
windows, with substantially higher CAGR (58.5% / 66.2%) and more trades. If CAGR
ever outranks Sharpe here, "no R:R floor" is a live option — and it would sever
the gate's last dependency on the weak S/R detector.
Also fixes a metric artifact in the holdout harness. The train book's equity curve
ran to the end of the data while its entries stopped at the split, so it sat in
flat cash for two years and deflated its own CAGR/Sharpe (reported 0.95 / 14.6%;
actually 1.31 / 29.6%). _simulate_portfolio now truncates the calendar to
hold_days after the last entry when end_date is set — it only triggers on the
holdout train window, so no other number moves. The clear-air OOS verdict is
unaffected: it rests on the test row, whose entries and curve both start at the
split and were always clean. Both holdout reports regenerated.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Investigated whether our support/resistance detection follows best practice
and whether we actually use it that way. Three findings, all backed by runs
against the prod snapshot and written up in docs/research/sr-levels-and-exits.md:
- The S/R target must NOT become an exit. Honoring it as a take-profit on top
of the 3x ATR trail drops Sharpe 2.04 -> 1.47 and halves CAGR. Win rate rises
(37.5% -> 40.0%), which is the tell: it truncates the right tail where
momentum's edge lives.
- The clear-air fallback (synthesize a 3xATR target where no resistance exists,
so 52-week-high breakouts stop being vetoed) looked strictly better in-sample
(Sharpe 2.04 -> 2.07, CAGR 50.4% -> 62.3%, DD 21.4% -> 20.1%) but FAILED a
real out-of-sample holdout: on entries after 2024-07-01 it is worse on Sharpe
(2.78 -> 2.45) and Calmar, better only on raw CAGR. Not shipped.
- The detector itself is weak vs best practice (POC/VAH/VAL computed then
discarded, HVN = any above-mean bin, 1.48x volume double-counting, "touch"
counts pass-throughs, no round numbers), but its only causal path to P&L is
the entry gate. Fix it for the displayed levels, not for returns.
Method note: nested lookback windows are NOT out-of-sample. The in-sample result
was clean, large, and consistent across five windows, and still did not survive
a proper entry-date split.
All research paths are off by default and the default report is unchanged:
BACKTEST_RESEARCH_EXITS=1 take-profit exit rows
BACKTEST_ATR_TARGET_FALLBACK=k synthetic k*ATR target when S/R offers none
BACKTEST_FALLBACK_CLEAR_AIR_ONLY=1 restrict that to genuinely clear air
BACKTEST_HOLDOUT_SPLIT=YYYY-MM-DD train/test split by entry date
Also fixes two reproducibility holes found while reconciling our local baseline
against the live report:
- create_backtest_snapshot.py now copies paper_% settings. The production
monitor row replays the runtime exit policy via get_exit_policy(); without
those keys a snapshot silently falls back to code defaults, so a live-tuned
exit would never be reflected.
- Migration 020 drops activation_min_expected_value and
activation_min_target_probability. Both are orphans of the June EV-gate
redesign, read by no code path, but prod carries min_target_probability = 50.0
which implies a probability floor that is not enforced (the real floor is the
20% constant in qualification.py).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A scoring error no longer skips setup detection for the ticker: the
rollback already restores a clean transaction, and qualification
re-gates on live scores at alert time, so a stale score is recoverable
but a skipped scan is not.
Iterating symbol strings instead of Ticker instances fixes a latent
crash the new regression test caught: rollback() expires ORM objects
regardless of expire_on_commit, so touching ticker.symbol in the except
handler triggered sync lazy-loading, which raises on an AsyncSession
and killed the whole scan on the first per-ticker error.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Three follow-ups to the gate probability floor (8f41143):
- Signals table shows the starred primary target (shared primaryTarget
helper) instead of an independently computed max-probability best,
so Overview, Signals and ticker details agree by construction.
- Targets pinned at the 3% probability clamp floor collapse to the
nearest one (enhance_trade_setup + backtest candidates in parity):
floor-pinned levels are indistinguishable to the model, so farther
ones were duplicate 3% rows inviting lottery headlines.
- get_trade_setups only returns setups re-emitted within
LIVE_SETUP_MAX_AGE_DAYS (3): an older latest row means the daily
scan no longer confirms the setup, and such rows otherwise surface
forever on Overview/Signals/ticker/alerts. History endpoints keep
full history.
Backtest on the Jul-3 snapshot is metric-identical to the gate-floor
run on all qualified stats (1089 qualified, Sharpe 2.02, CAGR +49.6%,
DD -15.8%): the prune only removes noise the gate already rejected.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
A qualified setup's primary target must now clear MIN_TARGET_PROBABILITY
(20%), shared with the primary-selection floor in recommendation_service
and mirrored in the frontend gate. Closes the read-time hole where a
stale pre-c7a198b row starring a far lottery target (probability pinned
at the 3% clamp floor, R:R inflated by the same distance) qualified
forever: the scanner emits no replacement row and live R:R never decays.
A/B backtest vs c7a198b baseline (same July-3 snapshot): 7 of 1096
qualified setups removed; qualified net avg R 0.202 -> 0.207, hold
Sharpe 2.00 -> 2.02, CAGR +48.8% -> +49.6%, max DD unchanged at -15.8%.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Layout regrouped by relationship, not size: the setup-in-focus card
and the radar sit side by side (they are one decision surface), the
four account ribbons move directly above the open positions they
describe, and a new performance chart closes the page.
- Radar rows are selectable: clicking one swaps the focus card to that
setup - including below-gate rows, whose card shows a muted
"rank N / below gate" badge and the disqualify reason in the footer,
with a "back to top pick" reset. The row currently in focus is
highlighted; ticker links still deep-link without selecting.
- Performance chart (the mockup's missing piece): new
GET /paper-trades/equity-curve computes, per benchmark trading day
since the first paper trade, the book's cumulative P&L (realized +
mark-to-market from stored OHLCV) vs the same cost basis riding SPY
over each trade's window (benchmark_prices). Pure curve math in
paper_trade_service with unit tests; hidden until there are 2+
points of data. Frontend renders both lines with crosshair readout,
zero baseline, and direct end labels.
Backend unit suite: 501 passed.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
A setup's primary target could carry a ~3% probability: the picker
chose the most likely target among those with R:R >= 1.5, and after a
run-up that pool can contain only far "lottery" levels (the near,
likely levels fail the R:R floor). The lottery target's inflated R:R
then became the setup's headline and passed the activation gate's
min_rr floor - the gate's probability check only requires a value to
exist.
Fix, no new tuning knobs: the primary must clear BOTH floors
(R:R >= 1.5 AND probability >= 20%). When nothing does, fall back to
the most likely target overall, so the headline carries an honest low
R:R and the gate rejects the setup on real numbers instead of being
gamed by an unreachable target.
Deliberately NOT pure EV-maximization (p*RR): the probability model
adds strength/alignment bonuses as flat percentage points, so EV
arithmetic would scale those bonuses by (RR+1) and systematically
favor far targets on aligned setups - the same lottery bias through
the back door.
Shared by production (enhance_trade_setup) and the backtest simulator,
so backtest comparisons stay apples-to-apples. Unit tests pin the
degenerate case; full unit suite green (497 passed).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The portfolio monitor's Production row now replays the live qualification
flag and the Admin exit policy (mode/ATR multiplier/hold days) instead of a
frozen research-variant gate, so Admin tuning is reflected in the next run.
Single-source the 80/20 strategy_rank weights in momentum_service and pin
every dual-defined constant with a parity test. Behavior-preserving today:
the production sim reproduces the README baseline exactly.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
A local backtest (offline prod snapshot, 506 tickers) evaluated blue-sky
projected targets under the PRODUCTION exit (3x ATR trailing + 30d max hold,
paper_trade_service DEFAULT_EXIT_MODE="atr_trailing"). Blue-sky setups are
dilutive: the qualified book scored 328% return / Sharpe 1.84 / DD -21.0%
WITHOUT them vs 300% / 1.58 / -18.7% WITH them. They rank high on momentum by
construction, so they grab slots from S/R setups that catch bigger runs under
a trailing-stop exit (only ~2pp worse drawdown doesn't justify the lost return
and Sharpe).
Reverts the scanner/TargetGenerator measured-move projection, the stricter
projected activation gate, the frontend qualification mirror, the `projected`
type field, and the projected tests -- all backend files are now byte-identical
to the pre-blue-sky commit.
Keeps the played-out "No current setup" UX (RecommendationPanel): when price
has run past the target (played out) or through the stop (invalidated), the
panel shows a plain no-setup state instead of a stale actionable card. This is
frontend-only (reads last close + existing setup fields) and is what actually
fixes the reported stale-below-price bug -- no backend change or rescan needed.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Fixes stale below-price setups showing as current recommendations. Three
distinct causes share the symptom (get_trade_setups returns the latest stored
setup per direction and never expires it):
- Genuine blue-sky (no overhead S/R): scanner + TargetGenerator now project a
measured-move target (entry +/- 3*ATR, ~2:1 R:R), flagged projected with a
low sr_strength probability haircut. Overhead check keys on level tag OR price
so it never projects through a straddling resistance cluster.
- Projected targets clear a stricter activation bar (long-only, momentum >= 90,
confidence >= min+10), independent of the general momentum gate. Mirrored in
frontend qualification.ts.
- Played-out UX (fixes the reported TTWO case, which is R:R-starved under a
resistance cluster, not blue-sky): when price is at/past target or through the
stop, RecommendationPanel shows a "No current setup" state and softens the
stale ticker-level header/reasoning, instead of a stale actionable card.
No migration: the projected flag rides in existing targets_json. 504 backend
unit tests pass; frontend typechecks.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Both _atr_trailing_close (scheduled) and _atr_trailing_level (dashboard
read path) recomputed ATR from scratch on every post-entry bar via
compute_atr(rows[:idx+1]) — O(n*k) per trade. Replace with a single O(n)
Wilder pass, _atr_series_from_rows, that stores round(running, 4) at each
index. compute_atr keeps its running ATR unrounded through the recurrence
and rounds only at return, so this reproduces its per-prefix value exactly
(no behavior change; live-vs-backtest atr_trail3 parity still byte-identical).
Remove the now-unused _atr_from_rows and its compute_atr import. Add a
per-index parity test against compute_atr; existing ATR tests now mock
_atr_series_from_rows (same effect as the old fixed-ATR mock).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Fixes ruff E741 in the lookbacks comprehension of _portfolio_monitor,
which failed the CI lint step and blocked the deploy.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Stored TradeSetup rows are point-in-time snapshots from the RR scan, so
the ticker page could show stale confidence/reasoning/composite (e.g.
sentiment=neutral in the setup card while the sentiment panel showed
bullish). Overlay current score/sentiment context onto the API payload
for GET /trades and GET /trades/{symbol}, gate and format Telegram
qualified-setup alerts on the same live values, and apply the
min_confidence/recommended_action filters after the overlay so they
judge what the caller actually sees. Stored setups stay frozen for
outcome analysis and backtests.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Backtest report now includes research-only hold-to-horizon portfolio variants comparing raw vs residual 12-1 momentum, cutoff 80 vs 90, max 10 vs 15 positions, and SPY-200 risk scaling. A dynamic research recommendation panel flags residual momentum, cutoff 90, or regime scaling only when transparent promotion rules pass.
Adds signal_context_snapshots with migration 016 and captures one point-in-time context row per newly generated TradeSetup: setup fields, composite/dimensions, latest sentiment, latest fundamentals, and strategy_version=momentum_12_1_rr_time_v1. This is forward-only; no historical sentiment/fundamental backfill is attempted.
No live gate, paper-trade exit, or production ranking behavior changes.
Verification: 458 backend tests pass, ruff check app/ clean, frontend npm run build clean.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Adds a research-only 12-1 residual momentum signal to the cross-sectional signal-evaluation harness. The signal estimates benchmark beta over the 12-1 formation window and ranks cumulative stock return minus beta-adjusted benchmark return; it only appears when benchmark closes are available.
No production qualification behavior changes. The Backtest signal table labels the new row as 12-1 residual momentum. Tests cover benchmark-gated emission and beta removal while keeping stock-specific drift.
Verification: 453 backend tests pass, ruff check app/ clean, frontend npm run build clean.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Production strategy change based on the July 2026 backtest: paper trades now default to a 30-trading-day hold with the initial stop (classic momentum hold-and-rerank), while target and trailing exits remain available in Admin. The exit policy API/UI now carries hold_days and close_reason can be 'time'.
The activation confidence floor default is now 0/off because the gate ablation showed it added no per-trade edge while filtering out usable setups. Migration 015 clears stored activation_min_confidence and paper_exit_mode so the new defaults take effect; this intentionally resets Track Record comparability from this deploy.
Verification: 451 backend tests pass, ruff check app/ clean, frontend npm run build clean.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The robustness warning was computed on the target-model distribution
while the same panel recommends the hold exit — internally inconsistent.
_robustness_stats (median, profit factor, ex-top-5% expectancy) is now
shared by _bucket_stats and _time_exit_bucket, the time-exit table shows
Median Net R and Ex-Top-5% per hold length, and _build_recommendation
reads the trimmed expectancy from the recommended exit's bucket (falling
back to the target model when no hold is recommended).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Robustness (answers 'is the edge just outliers?'):
- _bucket_stats gains median_net_r, profit_factor, and net_avg_r_ex_top5
(expectancy with the top 5% of winners removed); shown as stat tiles.
- Portfolio sim gains per-calendar-year returns, shown in the sim table.
Dynamic recommendation ('What this backtest recommends' panel):
- _build_recommendation derives advice from the report's own numbers on
every run — exit policy (target vs best hold, with sim CAGRs), which
gate floors earn their keep (ablation Hold column), best momentum
cutoff, book-vs-SPY verdict, and an outlier-dependence warning when
the trimmed expectancy goes non-positive.
Retired (conclusions reached, tables removed from report + UI):
- Take-profit sweep (no interior optimum — fixed TP is the wrong tool
for momentum), trailing sweep (converged to the hold-to-horizon exit),
probability calibration (model is display-only by decision).
- _tp_primitives slimmed to _risk_and_stop_day; trailing machinery gone.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Per-trade additions to the report:
- Gap-through-stop fills: stops now fill at the worse of the stop or the
bar's open across every exit model (target, TP, trailing, time), so a
loss can exceed -1R; targets never fill better than their level.
- best_r / worst_r, avg holding days, and net R per day of capital
deployed on the summary buckets and the time-exit sweep.
Portfolio simulation (the stats a per-setup replay cannot give):
- One capital-constrained book over the qualified setups: 10k start, max
10 concurrent positions (one per ticker, best momentum first), 1%
fixed-fractional risk with a 20% no-leverage notional cap, entries at
the detection close, 0.1%/side costs, daily mark-to-market.
- Two exit policies compared: S/R target race vs hold-to-horizon.
- Equity-curve stats: final equity, total return, CAGR, max drawdown,
annualized daily Sharpe, win rate, avg P&L, best/worst trade, avg
hold, entries skipped on a full book, and SPY price return over the
same window (benchmark history refreshed to cover the replay span).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The ablation judged floors under the target/stop model, but the exit
sweeps point at replacing that exit with a fixed hold — under which the
R:R floor's rationale (bigger payoff at the target) may not apply. Each
ablation row now also carries hold_avg_r / hold_net_avg_r / hold_total_r
(30d hold, initial stop only), so the Phase 3 gate decision can be read
under the exit policy that would actually be used.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
_window_setups computed them but _replay_ticker dropped them, so the
ablation's NEUTRAL/tightener checks saw None for every candidate and the
'without confidence floor' / 'without R:R floor' rows collapsed to 0
setups (impossible — removing a floor can only add setups). Regression
test now goes through the real _replay_ticker path.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Phase 1 of the strategy-measurement plan — report-only, no production
trading behavior changes:
- Cost haircut: every bucket/sweep now reports net_avg_r/net_total_r
alongside gross (COST_PER_SIDE=0.1% of notional, converted to R via
each setup's stop distance); params carry cost_per_side_pct.
- Gate ablation table: re-qualifies candidates at the current momentum
cutoff with one floor removed per row (confidence / R:R / NEUTRAL /
momentum-only) to show which floors earn their keep.
- Time-based exit sweep: hold 5/10/21/30 days with the initial ATR stop,
exit at the day-N close — the classic momentum implementation, to
disambiguate the wide-trailing result.
- TP sweep extended to +40/+50%, trailing to 25/30% so the optima are
interior instead of starred at the sweep edge.
- BacktestPanel: Net Avg R columns everywhere, gate-ablation and
time-exit tables, stars now mark best net avg R; stale cached reports
still render (all new fields optional/guarded).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Store an optional company name on Ticker (migration 014) and backfill it from
Alpaca's asset list in a single Trading-API call for the whole universe — no
per-ticker fetch. Runs automatically at the end of universe bootstrap and via a
manual "Backfill Names" button (admin) / POST /admin/tickers/backfill-names.
The name ships on /tickers; a shared symbol→name map (useTickerNames) lets any view
show it without its own request. Displayed subtly next to the symbol — in the global
search, the ticker header, and as a small muted line under the symbol in Top Setups
and Open Trades (no extra column, truncated so it never widens the table).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
get_score computed base_score / sentiment_score / sentiment_adjustment /
max_sentiment_adjustment, but the router's _map_composite_breakdown built the
response model from only the five original keys and silently dropped the rest — so
the API always returned null for them. That's why the ticker page showed neither the
"Composite = Base + Sentiment" caption nor the ± marker on the sentiment row despite
the frontend and scoring service both supporting it. Pass the fields through, with a
guard test so they can't be dropped again.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Going from no sentiment to a bullish read used to be able to *lower* the composite:
sentiment was blended into the weighted average as an absolute level, so a bullish
75 diluted a ticker already scoring 78. That's backwards for a directional signal.
Now the non-sentiment dimensions form a re-normalized weighted-average base, and
sentiment is applied as a signed adjustment around neutral (50):
composite = clamp(base + MAX_ADJ * (sentiment - 50) / 50)
MAX_ADJ = sentiment weight * 100 (default weight 0.10 → ±10)
Neutral leaves the base unchanged, bullish adds and bearish subtracts (scaled by
confidence, since a 50%-confidence call maps to 50 → no effect), and no sentiment
never penalises. Default sentiment weight 0.15 → 0.10; the weight now means "max ±
points." Composite breakdown exposes base_score/sentiment_score/sentiment_adjustment,
and the ScoreCard shows "Base 78 · sentiment +5.0" plus the per-dimension adjustment.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Applies the backtest-validated trailing stop to live paper trading, and surfaces
it transparently.
Exit (A):
- New paper-trade exit policy (paper_exit_mode=trailing, paper_trailing_pct=12),
tunable in Admin → Paper-Trade Exit. resolve_open_trades runs a trailing stop
(initial stop as floor, ratchets up from the peak; target ignored — the
validated rule) and records close_reason (trailing|stop|target|manual; +migration
013).
- list_trades enriches open trades with the live trailing-stop level + distance %.
Open Trades panel shows the active tactic and a Trail Stop column.
Alerts (B):
- Daily digest now lists open trades with unrealized gain, trailing stop, and how
far away it is.
- New "trade closed" alert: one summary per auto-close (trailing/target/stop, not
manual) — direction, reason, days held, P&L abs+%/R — covering wins AND
stop-loss losses. Deduped by trade id; toggle in Admin alerts.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Third exit model alongside target-vs-stop and the fixed take-profit. The TP sweep
showed the edge lives in the fat tail (avg R keeps rising as you let winners run),
but a fixed wide target is win-rate-brutal and gives everything back on a reversal.
A trailing stop harvests the tail while protecting gains.
Per setup the replay computes the realized R for several trail widths (3/5/7/10/
15/20%) in a single conservative pass — stop ratchets up via max(initial_stop,
peak*(1-trail)), exit on the pullback or at the horizon close, R vs the initial
risk. Aggregated into a trailing sweep (win rate = share closed in profit, avg R,
total R) over the qualified set and shown as a new table in the Backtest panel.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Avg R was still rising at the previous top level (+15%), so the optimum was off
the table. Extend TP_LEVELS to 20/25/30% to reveal where letting winners run
stops paying (it plateaus toward "just hold to the horizon close").
Also clarify in the panel that the take-profit model deliberately does NOT use
the setup's S/R target — it's a standalone fixed-% exit; exiting at the target is
the target-vs-stop model above. The two are complementary ends, not in conflict.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The target-vs-stop model counts a near-miss of a far S/R target as a full loss
and ignores the partial gains you actually bank — so it measures a different
strategy than "scalp the early pop, take +8%". Add a realistic take-profit exit
model next to it (original untouched).
Per setup the replay now also records risk%, whether the stop was hit, the
favourable excursion reachable before the stop (MFE), and the horizon-close move.
From those a fixed-take-profit sweep (4/6/8/10/12/15%) is scored in R: bank +X%
if reached before the stop, else -1R, else the horizon close. Hit rate = how
often +X% was banked (the MFE CDF), so you can pick the EV-optimal TP without
top-ticking fantasy. Shown as a new table in the Backtest panel; the IC,
calibration and momentum sweep are unchanged.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
