Replay any date
Research tool · not investment advice. Full disclaimer →
Ask a past date; see the clock and limits of each answer.
Descriptive record only — past data and published method.
Pick a date
any past trading daySource: controls.
Try ALKYLAMINE on 2020-04-01 · TANLA on 2020-06-15 · ICICIPRULI on 2019-01-25 · ICICIPRULI on 2019-01-21 (one day before the call — the prior point serves, nothing leaks) · RELIANCE on 2026-07-09 (recent — the event bus reconstructed by event date)
Replay timing for RELIANCE
when the inputs this replay used were recordedSource: recorded inputs.
How we know this2 of 5 fields not recoverable
- Rule
- No selection rule - this surface selects nothing. It replays what was knowable on a chosen date and withholds everything else.lens_registry: replay-any-date
- Source
- NSE bhav copy; NSE (nsearchives)redistribution: public-record · feed licence: public-archiveprovenance.PROVENANCE chain: bhav_eq · feed_manifest.FEEDS['bhavcopy']
- Knowable at
- AS_TRADED · 2026-10-01As-traded As-traded equity OHLC, NSE bhav 2026-10-01; EOD publish same trading day; row knowable at publish, never intradayprovenance.PROVENANCE chain: bhav_eq
- Base rate
- not recoverable from available datano database connection was supplied to this render, so no base rate was looked upseasonal_cells · scope=stock · axis=month
- Falsifier
- not recoverable from available datano settled test and no blocking failure model is on record for this surface - read it as unproven, not as working. What is on record is a handling fence, which constrains how the data may be used but states nothing that could turn out to be false: bhavcopy: raw-archive (the pre-compute doctrine keeps it never normalised away)our failure-model record, the strategy reference, and the feed manifest
What the API answered
raw, uneditedSource: /v1 · verbatim envelopes.
requested as-of 2026-07-09 knowable from 2026-04-24 basis EVENT rule applied concall-event-date
| Measure | On that date |
|---|---|
| Track record | mixed |
| Effective period | Apr 2026 |
| Promises resolved | 17 |
| Guidance accuracy | 61.8 |
| Momentum | 6.0 |
| Trend | IMPROVING |
What the concall-credibility engine could honestly say about RELIANCE that morning.
{
"data": {
"symbol": "RELIANCE",
"credibility": {
"track_record": "mixed",
"n_resolved": 17,
"as_of": "Apr 2026",
"guidance_accuracy": 61.8,
"momentum": 6.0,
"trend": "IMPROVING",
"robust": true,
"_raw_tier": "B",
"_raw_level": 68.4,
"knowable_from": "2026-04-24",
"knowable_basis": "EVENT"
},
"note": "DESCRIPTIVE guidance track-record for diligence in confluence — §C-backtested: NO validated return edge. NOT a recommendation, NOT a ranked buy/sell signal.",
"pit": {
"as_of": "2026-07-09",
"knowable_rule": "concall-event-date"
}
},
"_provenance": {
"cci_series": {
"data_class": "cci_series",
"label": "PIT credibility level + momentum + tape",
"source": "computed PIT",
"db": "hermes",
"grain": "derived",
"has_symbol": true,
"basis": "DERIVED",
"modeled": false,
"node_basis": "DERIVED",
"chain_basis": "DERIVED",
"derives_from_chain": [
"concall_settlement"
],
"node_redistribution": "owned",
"chain_note": "",
"redistribution": "owned",
"as_of": "Apr 2026",
"as_of_kind": "derived",
"effective_as_of": "Apr 2026",
"lag_days": null,
"method_versioned": true,
"display": "PIT credibility at Apr 2026, no look-ahead",
"warnings": []
}
},
"_meta": {
"as_of": {
"cci_series": "Apr 2026"
},
"knowable_as_of_oldest": null,
"basis": {
"cci_series": "DERIVED"
},
"coverage": "descriptive credibility track-record (no validated return edge — §C)",
"methodology_version": "v1.0",
"request_id": "682c9948f2c748f9912b9c0e62b64fb4",
"tenant": "dev",
"degraded": false,
"disclaimer": "Descriptors / percentile rank-gaps and point-in-time evidence — NOT investment advice and NOT a performance claim."
},
"_entitlement": {
"tier": "research",
"scope_used": "research"
}
}run it yourselfcurl -s -H "X-API-Key: $PATEARN_KEY" "https://srv1704897.hstgr.cloud/v1/securities/RELIANCE/credibility?as_of=2026-07-09"
requested as-of 2026-07-09
Retrospective event-date reconstruction: the API selects the newest event date on or before the requested day. A row may have been detected later; its recorded detection time is shown here. This is not an as-was alert inbox.
| Stock | Lens | Event | From, to | Event date | Recorded detection |
|---|---|---|---|---|---|
| ACSTECH | mep | phase_flip | ACCUM → NEUTRAL | 2026-07-09 | 2026-07-10 12:51:44 |
| ADVENTHTL | mep | phase_flip | NEUTRAL → ACCUM | 2026-07-09 | 2026-07-10 12:51:44 |
| ADVENZYMES | mep | phase_flip | DISTRIB → STRONG_DISTRIB | 2026-07-09 | 2026-07-10 12:51:44 |
| AEROFLEX | mep | phase_flip | NEUTRAL → DISTRIB | 2026-07-09 | 2026-07-10 12:51:44 |
| AMBALALSA | mep | phase_flip | NEUTRAL → DISTRIB | 2026-07-09 | 2026-07-10 12:51:44 |
| ANANDRATHI | mep | phase_flip | NEUTRAL → ACCUM | 2026-07-09 | 2026-07-10 12:51:44 |
A retrospective event-date view; recorded detection may be later than the date you requested.
{
"data": {
"attention": [
{
"id": 3,
"symbol": "ACSTECH",
"lens": "mep",
"event_type": "phase_flip",
"direction": "flip",
"from_state": "ACCUM",
"to_state": "NEUTRAL",
"magnitude": 1.0,
"as_of": "2026-07-09",
"detected_at": "2026-07-10 12:51:44",
"note": "ACSTECH: accumulation flipped ACCUM → NEUTRAL"
},
{
"id": 5,
"symbol": "ADVENTHTL",
"lens": "mep",
"event_type": "phase_flip",
"direction": "flip",
"from_state": "NEUTRAL",
"to_state": "ACCUM",
"magnitude": 1.0,
"as_of": "2026-07-09",
"detected_at": "2026-07-10 12:51:44",
"note": "ADVENTHTL: accumulation flipped NEUTRAL → ACCUM"
},
{
"id": 6,
"symbol": "ADVENZYMES",
"lens": "mep",
"event_type": "phase_flip",
"direction": "flip",
"from_state": "DISTRIB",
"to_state": "STRONG_DISTRIB",
"magnitude": 1.0,
"as_of": "2026-07-09",
"detected_at": "2026-07-10 12:51:44",
"note": "ADVENZYMES: accumulation flipped DISTRIB → STRONG_DISTRIB"
},
{
"id": 7,
"symbol": "AEROFLEX",
"lens": "mep",
"event_type": "phase_flip",
"direction": "flip",
"from_state": "NEUTRAL",
"to_state": "DISTRIB",
"magnitude": 1.0,
"as_of": "2026-07-09",
"detected_at": "2026-07-10 12:51:44",
"note": "AEROFLEX: accumulation flipped NEUTRAL → DISTRIB"
},
{
"id": 10,
"symbol": "AMBALALSA",
"lens": "mep",
"event_type": "phase_flip",
"direction": "flip",
"from_state": "NEUTRAL",
"to_state": "DISTRIB",
"magnitude": 1.0,
"as_of": "2026-07-09",
"detected_at": "2026-07-10 12:51:44",
"note": "AMBALALSA: accumulation flipped NEUTRAL → DISTRIB"
},
{
"id": 11,
"symbol": "ANANDRATHI",
"lens": "mep",
"event_type": "phase_flip",
"direction": "flip",
"from_state": "NEUTRAL",
"to_state": "ACCUM",
"magnitude": 1.0,
"as_of": "2026-07-09",
"detected_at": "2026-07-10 12:51:44",
"note": "ANANDRATHI: accumulation flipped NEUTRAL → ACCUM"
}
],
"pit": {
"as_of": "2026-07-09"
}
},
"_provenance": {
"signal_events": {
"data_class": "signal_events",
"label": "Typed state-change events",
"source": "computed",
"db": "hermes",
"grain": "event",
"has_symbol": true,
"basis": "DERIVED",
"modeled": false,
"node_basis": "DERIVED",
"chain_basis": "DERIVED",
"derives_from_chain": [],
"node_redistribution": "owned",
"chain_note": "",
"redistribution": "owned",
"as_of": "2026-10-02",
"as_of_kind": "derived",
"effective_as_of": "2026-10-02",
"lag_days": null,
"method_versioned": false,
"display": "Typed signal event on 2026-10-02",
"warnings": []
}
},
"_meta": {
"as_of": {
"signal_events": "2026-10-02"
},
"knowable_as_of_oldest": "2026-10-02",
"basis": {
"signal_events": "DERIVED"
},
"coverage": "descriptive state-changes (not signals to trade)",
"methodology_version": "v1.0",
"request_id": "409c86acfb274e94a1d95d5f7c9c3da9",
"tenant": "dev",
"degraded": false,
"disclaimer": "Descriptors / percentile rank-gaps and point-in-time evidence — NOT investment advice and NOT a performance claim."
},
"_entitlement": {
"tier": "research",
"scope_used": "research"
}
}run it yourselfcurl -s -H "X-API-Key: $PATEARN_KEY" "https://srv1704897.hstgr.cloud/v1/attention?as_of=2026-07-09"
requested as-of 2026-07-09
2,742 companies were listed and tradeable that day, delisted names included, counted as the date falls between its first and last trading day.
Who was listed and tradeable, including names since delisted.
{
"data": {
"status": "ok",
"policy": "Survivorship-correct: universe built from RAW bhavcopy_rows (delisted/suspended names RETAINED), not the current nse_equity_list snapshot.",
"total_securities": 4718,
"active": 2962,
"delisted_or_inactive": 1756,
"archive_floor": "2004-04-17",
"archive_ceiling": "2026-10-01",
"left_censored_at_floor": 736,
"as_of": "2026-07-09",
"universe_on_as_of": 2742,
"as_of_mechanism": "first_date <= d <= last_date (security_master.universe_on)",
"continuity_breaks": {
"AMALGAMATION": 4,
"DEMERGER": 113,
"MERGER": 1,
"SCHEME": 118
},
"renames": {
"confirmed": 1076,
"candidates": 8
},
"disclosures": [
"Survivorship is correct only from the archive floor 2004-04-17; delistings BEFORE the archive start are not captured (left-censoring).",
"Asymmetric: the price/survivorship spine keeps delisted names, but the fundamentals universe is the ~listed-today set — the two universes differ.",
"Universe = NSE mainboard EQ/BE/BZ series; SME (SM/ST) excluded.",
"Renames auto-confirmed via shared-ISIN clean handover; ISIN is sparse in the historical bhav feed, so rename resolution is biased toward currently-listed names until the authoritative NSE symbol-change feed is loaded."
],
"_provenance": {
"data_class": "survivorship",
"label": "Universe construction policy",
"source": "raw bhav archive",
"db": "hermes",
"grain": "derived",
"has_symbol": true,
"basis": "DERIVED",
"modeled": false,
"node_basis": "DERIVED",
"chain_basis": "DERIVED",
"derives_from_chain": [],
"node_redistribution": "owned",
"chain_note": "",
"redistribution": "owned",
"as_of": "2026-07-09",
"as_of_kind": "derived",
"effective_as_of": "2026-07-09",
"lag_days": null,
"method_versioned": false,
"display": "Survivorship-correct universe, delisted retained",
"warnings": []
}
},
"_provenance": {
"survivorship": {
"data_class": "survivorship",
"label": "Universe construction policy",
"source": "raw bhav archive",
"db": "hermes",
"grain": "derived",
"has_symbol": true,
"basis": "DERIVED",
"modeled": false,
"node_basis": "DERIVED",
"chain_basis": "DERIVED",
"derives_from_chain": [],
"node_redistribution": "owned",
"chain_note": "",
"redistribution": "owned",
"as_of": null,
"as_of_kind": "derived",
"effective_as_of": null,
"lag_days": null,
"method_versioned": false,
"display": "Survivorship-correct universe, delisted retained",
"warnings": []
}
},
"_meta": {
"as_of": {
"survivorship": null
},
"knowable_as_of_oldest": null,
"basis": {
"survivorship": "DERIVED"
},
"coverage": "survivorship-correct universe-as-of (delisted retained); see disclosures",
"methodology_version": "v1.0",
"request_id": "b62c0b808c8440d387c8b1e34aaee98d",
"tenant": "dev",
"degraded": false,
"disclaimer": "Descriptors / percentile rank-gaps and point-in-time evidence — NOT investment advice and NOT a performance claim."
},
"_entitlement": {
"tier": "research",
"scope_used": "compliance"
}
}run it yourselfcurl -s -H "X-API-Key: $PATEARN_KEY" "https://srv1704897.hstgr.cloud/v1/universe?as_of=2026-07-09"
Past-date evidence, with its clocks shown
the proof you can drive yourselfSource: /v1.
Pick a past date to ask the live API for its historical records. The credibility panel applies its knowability clock; the attention panel reconstructs events by their historical event date and may include events detected later. The universe counts names listed then, including names since delisted. Read each panel's time basis before treating it as something a subscriber could have seen that morning.
For credibility, the two-tier knowable clock uses EVENT when the period's real public clock was captured (the call was held, or the transcript was published — whichever is later). MODELED means no clock was captured, so the point only counts as knowable once its label month has completed. Result filings are never used as a clock for call-derived content, because filings usually come first — that is the leak direction.
The same answer, at a glance
two numbers from the replaySource: summary.
Zero look-ahead, proved on demand: the API is asked for a past date and answers with only what was knowable then.
How many companies were listed that morning, delisted ones included, and the earliest date the credibility read could honestly have been used.
What happened next
the revealSource: NSE bhav copy · split/bonus adjusted.
Replayed close on 2026-07-09: ₹1,279.80. The path enters at the next session's open, 2026-07-10: ₹1,291.90.
| After | Date | Adjusted close | Change from entry |
|---|---|---|---|
| 1 month | 2026-08-10 | ₹1,327.30 | +2.7% |
| 3 months | not yet | — | not yet |
| 6 months | not yet | — | not yet |
| 12 months | not yet | — | not yet |
Within 58 sessions after entry: the highest close was ₹1,334.80 on 2026-08-07 (+3.3%); the lowest was ₹1,167.70 on 2026-10-01 (-9.6%).
What followed is shown for any symbol and date you pick — winners and losers alike. It is the history after the date, never what the page predicted; the worked examples above were chosen after the fact.
Check this another way
the same evidence, from another sideSource: published boundary.
Cross-checks for a diligence reader: the per-dataset boundary · the pre-registered studies · the assembled procurement pack · the machine contract.