SV

Street Verdict

Analyst Scorecard

Methodology

This page explains, in plain language, exactly how Street Verdict decides an analyst call is a HIT or a MISS — and where the process is imperfect. It is written for skeptics, not for marketing: every number below is queried from the live database at the moment this page renders, not typed in by hand. If a claim here can't be independently checked, it doesn't belong on this page.

How we grade, in brief

Street Verdict collects public statements by analysts and firms from named news sources, with a link to each. We grade a call only when its own time frame has passed, using closing prices and entering at the next trading day's close. A call that never reached its stated entry is not scored, and a call we cannot check exactly is left ungraded.

Every hit rate we show comes with the number of graded calls and a 95% confidence range, because small samples mislead. We publish our rules and our restatements. You can report an error on any call, and we review every report and correct the record when we find a mistake. We do not give advice or recommendations, and we are not a SEBI-registered investment adviser or research analyst. The detail follows below.

Verdicts were restated on 19 September 2026

Grading used to be decided by a language model reading each quote and judging the price move. It is now decided arithmetically, from closing prices in our own database, by the rules set out in Section 2 — with no model involved in the verdict at all.

Every previously published call was recomputed under those rules, rather than leaving old verdicts standing beside new ones. Some calls changed from HIT to MISS and some from MISS to HIT, and the overall hit rate fell, because the new rules are stricter: entry moved to the next day's close, a target has to be reached by a close rather than an intraday touch, a setup whose stated entry was never reached is no longer scored at all, and a directional view has to be right at both one and three months.

We think restating is the honest choice. A scorecard that quietly keeps verdicts it no longer stands behind is worth less than one that changes them in public and says so.

1. What gets graded

An analyst opinion starts as a quote pulled from a tracked news source (see Section 5 for the current source list), tagged with a direction — bullish, bearish, or neutral — and, where possible, a specific tradeable instrument. Every opinion is then classified as trackable, long-horizon, or commentary.

Only trackable calls are scored, and every figure in this section counts trackable calls only. A call is trackable when it names an instrument we carry price history for and states a claim specific enough to check against that history — a target, a level, or a plain directional view. Long-horizon and commentary opinions (general market colour, macro views with no price call, conditional setups) stay logged and visible on the site, and on the analyst's own page, but never count toward anyone's score. Of the opinions we hold, a minority are trackable — so these counts are much smaller than the total we publish.

Trackable calls are not all gradable straight away either: before grading starts, a check confirms the claim can be resolved against real price data, and one that cannot is left ungraded rather than pushed to a shaky HIT or MISS.

16,263

trackable calls logged

11,292

currently graded (HIT or MISS) — 70.5% of non-suppressed trackable calls

3,825

still pending their resolution window — 23.9% of non-suppressed trackable calls

Most trackable calls are still waiting for their window to elapse, not stuck or ignored — see Section 3 for how that window is set. A further 2.8% of non-suppressed trackable calls were reviewed and found not-gradable (Section 2) rather than published as a shaky HIT or MISS. The remainder were removed from the public record entirely — see Section 4 on suppression.

2. How HIT / MISS is decided

Verdicts are decided arithmetically from closing prices only — never an intraday tick. A price that touches a target during the day but closes below it has not reached that target, and a wick through a stop is not a stop-out. The chart you see on a call uses the same closing series the verdict is computed from, so you can check our working.

Entry is the next trading day's close. Whatever time a call was published, we enter at the close of the first trading session after that date — never the same session, and never a price that printed before the call was made.

From there, what counts as right depends on what was actually claimed:

  • A stated target — a closing price has to reach it within the window.
  • A trade setup with a stated entry — a level, or a zone such as “accumulate between ₹2,160 and ₹2,100” — we wait for a close to reach it: for a zone, the first close inside it. If no close does in the 30 days that follow, the setup never triggered (“entry zone ₹2,100–2,160 never reached”): no hit, no miss, and it counts toward nobody's score. We will not credit a trade that was never available to take, and a call that merely closed above the zone the next day was not a trade.
  • A directional view with no level — both the one-month and three-month returns have to go the way the analyst said. One out of two is a miss.
  • A conditional call — it says nothing until a daily close goes beyond a level it states. The first side to fire is judged on its own target, which has to be reached on a close. If neither side fires inside the window, or the side that fires gives no target of its own (“triggered, but no target given for this side”), the call is not graded and counts toward nobody's score.
  • A relative call — judged on the gap between the pick's return and the comparator's return, from the first close after the call to the end of its window (three months when none is stated). A gap above zero is a hit for a call that favours the pick, a gap of zero or less is a miss, and a call that marks the pick down is read the other way. The comparator is the one the quote itself names; where we cannot name it, the call is not graded.

Where we refuse to grade. A demerger, bonus or split moves a share price without anything happening to the business. Where the action can be adjusted for exactly we do so; where it cannot — a demerger hands shareholders separate securities, with no honest single ratio — we mark the call ungradable and leave it out. An unscored call costs nothing. A confident wrong one costs someone their record.

3. Resolution window

A call isn't graded early, and it isn't graded on one company-wide schedule — it's graded once the analyst's own time frame has passed, read from their words and the same for every kind of call: target calls, trade setups, directional calls and conditional calls.

  • A stated time frame — “next two weeks”, “by December”, “12-month target”, “year-end” — is the window, counted from the day the call was published. If a quote states several, the longest counts.
  • “Within a month”, “near term” and “short term” set a 30-day window, including when the phrase sits before an “if” (“in the short term if it closes above 1,760”). “Medium term” is 90 days; “long term” is 12 months. A level called “intraday support” is a level, not a one-day claim.
  • No time frame, but a target and a stop loss — the call is tracked until a close reaches one of them. The target first is a hit, the stop first is a miss, and there is no fixed window: until one is reached the call is shown as open (“tracking until ₹2,400 or ₹2,000 is reached”) and counts toward no hit rate.
  • No time frame and no stop — a technical setup gets 30 days; a call with a price target gets 12 months; a relative call gets 3 months; a conditional call may take up to a year to trigger, and its target then gets 30 days. Vague phrases such as “in the coming weeks” do not set a window.

Every change to these rules is written down, with its date and what it did to the record, in the rule changes below.

4. Suppression

Opinions can be manually removed from the public scorecard by the editorial/moderation team. This is a simple boolean fact about a row — suppressed or not — independent of whatever grading status it has underneath; a suppressed opinion doesn't disappear from our database, it's just excluded from every public number on this site.

1.5%

of all logged opinions (242 of 16,263) are currently suppressed

Why opinions get suppressed. Based on what our own cleanup tooling actually does, the likely reasons are: duplicate or conflicting extraction (the same analyst quote pulled twice, or two directly contradictory reads of the same article — we suppress rather than guess which one is right), misattributed or unresolvable text (a quote that doesn't map to a real tradeable instrument), and junk or too-short extractions that don't clear a basic quality bar. Most suppressed rows — 69.4% of them — were still awaiting grading when they were suppressed, so suppression is mostly a data-quality cleanup step, not a way of hiding unfavorable verdicts after the fact.

Reason codes. Every suppressed call carries a reason code that only our reviewers see (for example: no outside evidence of which company it is about, a duplicate, or not a call). Of the 1,371 calls suppressed today, 341 have a specific reason and 1,030 are labelled “other”, mostly calls hidden before reason codes existed whose reason we could not rebuild from the record. You can report an error on any call; we review every report and correct the record when we find a mistake.

5. Data sources

Price data comes from Yahoo Finance — end-of-day/delayed quotes, not tick-level or real-time data. Historical end-of-day price data for grading older calls is cross-checked against NSE/BSE bhavcopy archives. Neither source is instantaneous, and we don't claim otherwise anywhere on this site.

Opinions are extracted only from a tracked allowlist of news sources — currently 12 domains:

  • economictimes.indiatimes.com
  • livemint.com
  • cnbctv18.com
  • zeebiz.com
  • financialexpress.com
  • businesstoday.in
  • thehindubusinessline.com
  • etnownews.com
  • goodreturns.in
  • timesofindia.indiatimes.com
  • outlookbusiness.com
  • fortuneindia.com

This list will grow as coverage expands — it is queried live from the same allowlist constant the extraction pipeline itself gates on, so it can never silently go stale on this page.

6. How a call played out

A HIT or a MISS says what happened to one call. We judge only what the analyst actually said, so beside a verdict we show only what follows from their own words and the price path, on closing prices, from the same entry the grade used. We do not set a call against an index, a market or a statistical baseline: those are comparisons the analyst did not make.

Time to target

For a hit: the trading sessions and calendar days from the entry to the close that reached the target.

The worst close against a call

The furthest a closing price went against the call, in per cent of the entry price, before the call was decided: for a bullish call the lowest close below the entry, for a bearish call the highest close above it. An intraday dip is never counted, in step with every verdict.

Target or stop first

For a call that states both a target and a stop loss: which of the two a close reached first, or that neither was reached by the end of its window.

On an analyst's page

The median time to target over their hits and the median worst close against a call, each with the number of calls behind it, and only from five calls. They describe past calls; they do not rank analysts and are never advice.

7. Known limitations

How we tag direction, and why bullish still leads

Of 28,588 non-suppressed opinions, 80.5% are tagged bullish, 10.8% bearish, and 8.7% neutral. Some of that gap is real: Indian brokerage research genuinely skews toward initiations, upgrades, and buy calls more than outright sell calls, and analysts hedge positive views in print far more often than negative ones — independent sentiment baselines run bullish-leaning too, just not always this wide. We work to keep our own extraction and counting process from adding to that real-world skew on top of it: the model is shown equally rigorous worked examples of bullish, bearish, and neutral calls, so it isn't primed to reach for one direction more readily than another; when a single article's signals are genuinely mixed, we tag whichever direction is actually dominant, with a true tie landing on neutral rather than being awarded to conviction; and every sentiment number on this site counts each analyst's latest stance on an instrument once, no matter how many times they've repeated that same call across different articles — a restated view is still one opinion, not several. We validate this quote-by-quote against real published calls, not just by checking whether an aggregate percentage looks better. The number above will keep drifting as new opinions replace older ones in our rolling windows — it won't land at an even three-way split, and shouldn't, because the underlying commentary itself isn't evenly split either.

The sample-size caveat

A hit rate computed from only a handful of calls can look impressively high (or low) simply because the sample is small. Take our own worked example: an analyst with 4 hits out of 6 graded calls has a headline rate of 67% — but the real 95% confidence interval on that number, computed with a Wilson score interval, is roughly 30–90%. That is a huge range — wide enough that this analyst could plausibly be a genuine coin-flip or a genuinely strong performer, and 6 calls alone can't tell you which. Anywhere you see a percentage on this site with a Low sample badge next to it, this is why — look for the confidence-interval range next to the number, not just the number itself.

Rule changes

2026-11-14

Comparisons the analyst didn't make (market, chance) are not shown

We judge only what the analyst actually said. Beside a verdict we show only what follows from their own words and the price path, on closing prices and from the same entry the grade used: the entry, the target, the stop and the window as stated; the time a hit took to reach its target; the worst closing move against a call on the way; and, for a call that states a target and a stop, which of the two a close reached first. A call is not set against an index, a market or a statistical baseline, because those are comparisons the analyst did not make. No verdict is changed.

What it did to the record: Nothing in the record moved. Of the graded Calls with these figures, 4,141 hits state a target and reached it in a median of 12 trading sessions, and the median worst closing move against a call before it was decided is 3.7%.

2026-11-14

A call with no time frame, a target and a stop is tracked until a close reaches one of them

A call that states no time frame but gives both a target and a stop loss is tracked until a close reaches one of them: the target first is a hit, the stop first is a miss, and until then it stays open, with no fixed window. A stated time frame or a horizon word still sets the window. The 30-day default for a technical setup (and the 12-month default for a target call) now applies only to a call that gives a target without a stop. A setup that states an entry still has 30 days to reach it.

What it did to the record: 5,264 setups state no time frame and give both levels. On today's prices a 30-day window would give them 1,172 hits and 3,099 misses (27.4%); tracked to the target or the stop they have 1,458 hits and 2,795 misses (34.3%), with 281 still open. 261 setups that missed at the end of 30 days went on to close at their target before any close reached their stop.

2026-11-14

An entry zone the analyst states is read, and the call has to reach it

A call that says "accumulation between ₹2,160 and ₹2,100", "in the range of ₹126-127", "a buying range of ₹482-492" or "on dips to ₹1,240" states an entry. Many were stored with no entry and so graded as a target call entered at the next close, which could already be above the zone: a hit for a trade that never reached its entry. The entry (a zone or a level) is now read from these phrases, the call starts at the first close that reaches it, and a call that never gets there in the 30 days allowed is not triggered, graded neither hit nor miss. A call that enters at the current market price and adds on dips keeps the next close as its entry, and a time frame such as "within 3-4 weeks" is no longer mistaken for a price zone.

What it did to the record: 296 setups and target calls had an entry the reader had missed (233 zones, 63 levels); 19 more had a time frame stored as a zone, now cleared. Of the 296: hits 71 to 75, misses 183 to 142, pending 42 to 15, and 64 became not triggered (24 hits, 30 misses and 10 pending before), among them the TCS call of 12 October 2026. With the open-ended rule, the hit rate over all Calls moved from 36.5% to 38.7%, and 76 analysts' counts changed.

2026-11-14

A stated or implied short time frame sets a 30-day window for every kind of call

"Within a month", "in the next two weeks", "near term" and a horizon word that sits before an "if" ("in the short term if it closes above 1,760") now set the call's window, the same way "short term" always did: 30 days for "within a month", "near term" and "short term", the stated length for "next two weeks". This applies to target calls, trade setups, directional calls and conditional calls alike; until now only some of these phrases were read, and a call that said "within a month" could be graded on the 12-month default. A level that is described as "intraday support" is a level, not a one-day claim, and a month written next to a valuation ("Sep-27 adjusted PB") is not a time frame. Vague horizons ("in the coming weeks") are unchanged and keep the default window.

What it did to the record: Reading the phrases changed the window of 117 target-bearing calls (108 to a shorter one) and they were regraded: 12 hits became misses and 2 misses became hits, 33 pending calls were decided (28 misses, 5 hits), 2 became not graded. Those calls went from 43 hits and 16 misses to 39 hits and 54 misses; the hit rate over all Calls moved from 36.6% to 36.5%.

2026-11-14

A conditional call whose triggered side gives no target is not graded

A conditional call makes no claim until a daily close goes beyond a level it states. The first side to fire is judged on its own target. If that side gives no target of its own, the call is shown as "triggered, but no target given for this side", is not graded, and counts toward no hit rate; the other side does not take over.

What it did to the record: This was already how conditional calls were judged; the label is now said in those words on every call page.

2026-11-11

Conditional and relative calls are judged on their own rules, with official index closes

Conditional calls (R9): the trigger is a daily close beyond the far edge of a stated zone, the target has to be reached on a close, nothing that never fires is graded. Relative calls (R6): the pick's return minus its comparator's return from the first close after the call to the end of its window (3 months when none is stated); a spread above zero is a hit for a call that favours the pick, exactly zero is a miss, and a call that marks the pick down is mirrored. The comparator comes from the quote (a sector, the market, a named company); a call whose comparator cannot be named, or whose preferred side is not the instrument it is filed under, is not graded. Index calls are judged on the exchanges' own closing values from September 2024.

What it did to the record: Conditional calls in the Calls table went from 156 hits and 90 misses to 445 hits and 226 misses once the index history and the readings were fixed (pending 374 to 31); relative calls from 6 to 33. The hit rate over all Calls moved from 35.2% to 36.6%.

Have a question this page doesn't answer, or spot something that looks wrong? Every call on the scorecard links back to its original source article — that's the ultimate check on any number here. If something is wrong, use “Report an error” on the call and we will look at it.