Every result, read before it lands. Judged after.
Sign in
NIFTY 50 22,776.10 +0.98% NIFTY BANK 55,128.40 +0.76% NIFTY IT 28,136.05 -0.59% NIFTY MIDCAP 100 59,761.20 +1.08% NIFTY SMALLCAP 100 19,448.70 +1.56% NIFTY AUTO 25,541.95 +0.47% NIFTY PHARMA 26,549.60 +1.66% NIFTY FMCG 45,205.35 +1.40%
Closed Close 6 Oct

Track record

Every week from 29 Mar 2025 to 3 Oct 2026: the read made on the Saturday before, and what each share then did against Nifty over the ten sessions after its result. 70 of these 73 Saturdays are simulated — the engine was run on 13 Sep 2026 standing on each one in turn, with prices to the Friday close and only the results that had already happened. Nobody could have acted on those reads on the day; they show what the method would have said, not what it did say.

68weeks with results
5,972names read
5,810judged
Learning indexout of sample · 4,419 judged reads
+0.086
information coefficient of the model in force · t 4.5 · positive on 89% of 28 mornings
25Q3 25Q4 26Q1 26Q2 26Q3
hypothesis weights on the same mornings: +0.013 · last fit 1 Oct, applied
Profitability indextop fifth vs bottom fifth · ten sessions · vs Nifty · before costs
+28.7 pts
cumulative long-short excess over 29 result weeks · +17.7 pts after 20 bp a position · +1.18% per read · top fifth beat Nifty 42% of 886
hypothesis weights, same procedure: +20.0 pts · a description of the record, not a recommendation

The two indices, in full

every read scored by weights fitted only on the seasons before it · updated each Saturday
Model in forceHypothesis weightsEvery read
Information coefficient (mean over unseen mornings)+0.086+0.013—
t-statistic · mornings4.5 · 280.6 · 28—
Top fifth: mean excess per read · beat Nifty-0.75% · 42%-1.14% · 40%-1.55% · 37%
Bottom fifth: mean excess per read-1.93%-1.85%—
Long-short spread per read+1.18%+0.71%—
Cumulative long-short, summed week by week+28.7 pts+20.0 pts—
Cumulative long-short after costs (20 bp a position, assumed)+17.7 pts+9.0 pts—
Cumulative long only-18.1 pts-14.6 pts—
IC by season 25Q3 +0.118 · 25Q4 +0.047 · 26Q1 +0.102 · 26Q2 +0.043 · 26Q3 +0.156 25Q3 +0.008 · 25Q4 +0.108 · 26Q1 -0.050 · 26Q2 +0.014 · 26Q3 -0.006—
How to read this. "Model in force" is the weights the morning read runs on today (L1 0%, L2 15%, L3 0%, L4 0%, L5 85%, L6 0%), fitted from the post-mortems on 1 Oct 2026. To score a season it is refitted on the seasons before that one only, so nothing here was scored by weights that had seen its outcome. "Hypothesis weights" are the original 30/20/30/20 put through the identical procedure. "Every read" is the base rate. The first season on record (5 Jul 2025 was the first morning after it) has nothing before it to fit on, so both columns are measured from the first season a fit could exist. The profitability figures are an equal-weight basket of each morning's top fifth held ten sessions after the result, in excess of Nifty 50, before costs and slippage — a description of the record, not a recommendation. 4,419 judged reads, 5 Jul 2025 to 25 Jul 2026; computed 3 Oct 2026.

Did the calls carry information?

ten sessions after the result, against Nifty 50
39% favourable reads that rose
737 of 1874 · all names read: 38% rose
62% unfavourable reads that fell
1134 of 1826 · all names read: 62% fell
+0.0% spread
avg excess, favourable -0.8% minus unfavourable -0.8%

A favourable read (strong or favourable) says the name has tended to rise after its results; an unfavourable one (unfavourable or weak) that it has tended to fall. It is a hit when the ten-session move against Nifty went that way. The line to beat is the one beside each figure: what every name the engine read did, whatever its label. If the two hit rates sit on that line and the spread is near zero, the label carries no information — and that is what this page is for finding out. Neutral reads make no call and are not scored. For reference, of every result in the period including the ones never read (12,476), 38% rose against Nifty.

How the simulation stands

Knows on each Saturday: prices, volumes and index levels to the Friday close; every past result and its reaction that had already played out.

Overstates slightly: result dates come from the confirmed calendar as it stands today. A few of the week's dates would not yet have been intimated to the exchange by that Saturday. The surveillance list (ASM/GSM) is today's.

History floor: simulated weeks read a name with 8 or more clean quarters; the live rule is sixteen. Every read shows how many quarters it rested on.

Same code: the simulated read and the post-mortem that judges it are the live jobs, called with a past date. Nothing was tuned on this period.

Week by week

newest first · click a Saturday to stand on it
Result weekRead on ReportingReadJudged Favourable ↑ Unfavourable ↓ Spread Biggest mover
5–10 Oct 2026 Sat 3 Oct 2026 live 17 5 0 +5 open — — — —
28 Sep – 3 Oct 2026 Sat 26 Sep 2026 live 8 0 0 — — — —
21–26 Sep 2026 Sat 19 Sep 2026 live 6 0 0 — — — —
27 Jul – 1 Aug 2026 Sat 25 Jul 2026 simulated 367 233 233 49% 38/77 58% 39/67 -0.1% XPROINDIA -21.5% favourable
20–25 Jul 2026 Sat 18 Jul 2026 simulated 255 166 166 49% 27/55 62% 34/55 -0.2% BANDHANBNK -20.6% favourable
13–18 Jul 2026 Sat 11 Jul 2026 simulated 106 69 69 24% 5/21 63% 15/24 -0.6% NUVOCO +19.2% weak
6–11 Jul 2026 Sat 4 Jul 2026 simulated 15 9 9 0% 0/1 67% 2/3 -1.8% TCS +6.4% weak
29 Jun – 4 Jul 2026 Sat 27 Jun 2026 simulated 4 0 0 — — — —
22–27 Jun 2026 Sat 20 Jun 2026 simulated 1 0 0 — — — —
15–20 Jun 2026 Sat 13 Jun 2026 simulated 1 0 0 — — — —
8–13 Jun 2026 Sat 6 Jun 2026 simulated 1 1 0 +1 open — — — —
1–6 Jun 2026 Sat 30 May 2026 simulated 121 34 34 45% 5/11 64% 7/11 +2.5% PTCIL +19.7% neutral
25–30 May 2026 Sat 23 May 2026 simulated 844 283 186 +97 open 35% 21/60 58% 32/55 -2.4% TECHNOE -21.9% strong
18–23 May 2026 Sat 16 May 2026 simulated 470 225 224 +1 open 52% 39/75 63% 43/68 +3.7% KDDL +32.4% unfavourable
11–16 May 2026 Sat 9 May 2026 simulated 401 212 180 +32 open 32% 17/53 62% 32/52 -0.8% GLAND +20.5% strong
4–9 May 2026 Sat 2 May 2026 simulated 260 162 162 42% 20/48 60% 30/50 -0.7% SHOPERSTOP +21.2% weak
27 Apr – 2 May 2026 Sat 25 Apr 2026 simulated 190 132 132 55% 23/42 61% 23/38 +2.5% VEDL -61.9% neutral
20–25 Apr 2026 Sat 18 Apr 2026 simulated 97 69 69 52% 13/25 48% 11/23 +1.4% TANLA +15.2% strong
13–18 Apr 2026 Sat 11 Apr 2026 simulated 37 23 23 25% 2/8 67% 6/9 +1.4% VSTIND +9.3% neutral
6–11 Apr 2026 Sat 4 Apr 2026 simulated 6 4 4 100% 1/1 100% 2/2 +8.2% TCS -4.5% weak
30 Mar – 4 Apr 2026 Sat 28 Mar 2026 simulated 0Nothing due — between reporting seasons
23–28 Mar 2026 Sat 21 Mar 2026 simulated 0Nothing due — between reporting seasons
16–21 Mar 2026 Sat 14 Mar 2026 simulated 4 0 0 — — — —
9–14 Mar 2026 Sat 7 Mar 2026 simulated 3 0 0 — — — —
2–7 Mar 2026 Sat 28 Feb 2026 simulated 5 0 0 — — — —
23–28 Feb 2026 Sat 21 Feb 2026 simulated 8 5 5 50% 1/2 50% 1/2 -2.7% KSB +7.4% unfavourable
16–21 Feb 2026 Sat 14 Feb 2026 simulated 132 34 34 23% 3/13 79% 11/14 -4.9% EASEMYTRIP +43.7% weak
9–14 Feb 2026 Sat 7 Feb 2026 simulated 1056 367 367 42% 50/119 73% 88/120 +2.3% EASEMYTRIP +43.7% weak
2–7 Feb 2026 Sat 31 Jan 2026 simulated 571 298 297 +1 open 36% 33/91 53% 47/88 -1.1% GRWRHITECH +29.7% neutral
26–31 Jan 2026 Sat 24 Jan 2026 simulated 372 229 228 +1 open 45% 33/73 46% 31/67 -2.8% SERVOTECH +42.6% unfavourable
19–24 Jan 2026 Sat 17 Jan 2026 simulated 199 123 123 41% 15/37 50% 18/36 -1.6% HOMEFIRST +19.4% unfavourable
12–17 Jan 2026 Sat 10 Jan 2026 simulated 109 62 40 +22 open 35% 6/17 86% 6/7 +0.5% JINDALSAW +20.8% neutral
5–10 Jan 2026 Sat 3 Jan 2026 simulated 15 8 8 0% 0/3 100% 1/1 +0.1% ELECON -18.6% neutral
29 Dec – 3 Jan 2026 Sat 27 Dec 2025 simulated 3 2 1 +1 open 0% 0/1 — — SWARAJENG -1.1% strong
22–27 Dec 2025 Sat 20 Dec 2025 simulated 2 0 0 — — — —
15–20 Dec 2025 Sat 13 Dec 2025 simulated 5 2 2 100% 1/1 0% 0/1 +2.7% ORISSAMINE +3.4% strong
8–13 Dec 2025 Sat 6 Dec 2025 simulated 4 0 0 — — — —
1–6 Dec 2025 Sat 29 Nov 2025 simulated 6 0 0 — — — —
24–29 Nov 2025 Sat 22 Nov 2025 simulated 7 1 1 — — — PARSVNATH -2.8% neutral
17–22 Nov 2025 Sat 15 Nov 2025 simulated 3 0 0 — — — —
10–15 Nov 2025 Sat 8 Nov 2025 simulated 1197 445 445 22% 31/139 86% 137/159 +1.4% AHLUCONT +19.3% unfavourable
3–8 Nov 2025 Sat 1 Nov 2025 simulated 514 295 295 20% 20/102 89% 89/100 -0.4% LUMAXTECH +19.1% favourable
27 Oct – 1 Nov 2025 Sat 25 Oct 2025 simulated 271 177 176 +1 open 38% 21/56 76% 39/51 +1.7% LGBBROSLTD +28.0% unfavourable
20–25 Oct 2025 Sat 18 Oct 2025 simulated 63 36 36 62% 8/13 69% 9/13 +4.6% PFS +8.9% weak
13–18 Oct 2025 Sat 11 Oct 2025 simulated 186 126 126 55% 23/42 61% 22/36 +3.5% DCBBANK +22.5% favourable
6–11 Oct 2025 Sat 4 Oct 2025 simulated 19 6 6 50% 1/2 100% 3/3 +15.9% EIMCOELECO -6.0% weak
29 Sep – 4 Oct 2025 Sat 27 Sep 2025 simulated 0Nothing due — between reporting seasons
22–27 Sep 2025 Sat 20 Sep 2025 simulated 4 2 2 0% 0/1 — -14.2% SUDARSCHEM -5.5% strong
15–20 Sep 2025 Sat 13 Sep 2025 simulated 4 0 0 — — — —
8–13 Sep 2025 Sat 6 Sep 2025 simulated 5 0 0 — — — —
1–6 Sep 2025 Sat 30 Aug 2025 simulated 6 2 2 — 100% 1/1 +13.6% PARSVNATH +2.3% weak
25–30 Aug 2025 Sat 23 Aug 2025 simulated 3 0 0 — — — —
18–23 Aug 2025 Sat 16 Aug 2025 simulated 3 0 0 — — — —
11–16 Aug 2025 Sat 9 Aug 2025 simulated 789 286 286 39% 37/96 58% 51/88 +0.0% SPIC +24.0% favourable
4–9 Aug 2025 Sat 2 Aug 2025 simulated 593 292 292 34% 33/98 74% 59/80 +1.0% YATRA +43.2% strong
28 Jul – 2 Aug 2025 Sat 26 Jul 2025 simulated 393 246 246 36% 29/81 77% 58/75 +0.7% SARDAEN +23.0% unfavourable
21–26 Jul 2025 Sat 19 Jul 2025 simulated 241 157 157 19% 10/53 88% 45/51 +1.5% POCL +15.8% strong
14–19 Jul 2025 Sat 12 Jul 2025 simulated 96 64 64 26% 6/23 70% 16/23 -3.8% VIMTALABS +12.7% unfavourable
7–12 Jul 2025 Sat 5 Jul 2025 simulated 18 6 6 50% 1/2 100% 1/1 +9.0% ARIHANTCAP -4.9% favourable
30 Jun – 5 Jul 2025 Sat 28 Jun 2025 simulated 1 1 1 — — — MICEL -3.0% neutral
23–28 Jun 2025 Sat 21 Jun 2025 simulated 4 0 0 — — — —
16–21 Jun 2025 Sat 14 Jun 2025 simulated 5 0 0 — — — —
9–14 Jun 2025 Sat 7 Jun 2025 simulated 5 1 1 — — — ORISSAMINE -0.4% neutral
2–7 Jun 2025 Sat 31 May 2025 simulated 0Nothing due — between reporting seasons
26–31 May 2025 Sat 24 May 2025 simulated 775 273 273 41% 35/85 46% 42/91 -2.0% LUMAXTECH +29.0% strong
19–24 May 2025 Sat 17 May 2025 simulated 439 210 210 59% 38/64 33% 23/69 -0.2% HONDAPOWER +31.9% strong
12–17 May 2025 Sat 10 May 2025 simulated 384 215 215 54% 36/67 20% 13/65 -1.1% RAYMOND -65.0% unfavourable
5–10 May 2025 Sat 3 May 2025 simulated 229 148 148 67% 30/45 14% 7/49 -4.7% CCL +30.4% unfavourable
28 Apr – 3 May 2025 Sat 26 Apr 2025 simulated 168 112 112 47% 17/36 39% 15/38 -2.8% KSOLVES -19.0% strong
21–26 Apr 2025 Sat 19 Apr 2025 simulated 122 94 94 26% 8/31 62% 21/34 -2.0% RAJRATAN +34.7% unfavourable
14–19 Apr 2025 Sat 12 Apr 2025 simulated 34 18 18 0% 0/3 67% 4/6 -2.1% TATAELXSI +11.8% weak
7–12 Apr 2025 Sat 5 Apr 2025 simulated 4 2 2 0% 0/1 — -10.8% ANANDRATHI +0.3% strong
31 Mar – 5 Apr 2025 Sat 29 Mar 2025 simulated 0Nothing due — between reporting seasons
Biggest mover is the largest close-to-close move on the reaction day among the names judged that week, with the read it carried. A week where the read is refreshed daily (live weeks) is judged on the last read before each result; the Saturday row still counts every name it read.

The v1 record · five heroes a week, August–September 2026

What the earlier version of the screen published, kept exactly as published. Everything here is a hit rate on observed movement, not a return: the screen surfaced names likely to be volatile around their results and made no claim about direction.

Moved as much as expected

75.0%

8 names scored · bar set at 65.0%

Accumulation signature followed

50.0%

6 observed · bar set at 58.0%

Distribution signature followed

—

0 observed

Sector call correct

50.0%

4 scored

Biggest mover captured

0.0%

of 9 weeks reviewed

Weeks published

3

0 results postponed, excluded from scoring

What each number means

Moved as much as expected
The share of surfaced names that moved at least 60% of their own historical average move on results. This is the screen’s central claim, and the only one it is really making: that these names tend to be eventful around results.
Accumulation signature followed
Where Layer 3 observed a name decoupling upward from the index before its result, how often the reaction was upward. Reported as a base rate for comparison, not as a forecast — the signature is an observation about the past 90 sessions, and the screen states plainly that it may be wrong.
Biggest mover captured
How often the largest actual reaction of the week was among the names surfaced. This is the harshest number on the page and it is published on purpose. When it is low, the screen missed the story.

What this page will never show

No returns, no portfolio, no entry or exit levels, no price goals. This is a screen with a published record of how well it screens. What anyone does with that is entirely their own decision.

FY2026-27 · week by week

The fiscal year runs from 1 April, so week 1 begins Monday 6 Apr 2026. All 52 weeks are shown from the outset — a week that was never published stays visibly missing rather than disappearing from the record.

reviewed published, verdict pending not published still to come
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 3/5 19 3/5 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

3 of 52 weeks published so far this fiscal year. A figure computed from a handful of weeks is noise; these become meaningful across a full season, which is exactly why every empty slot is on show.

Weeks reviewed

WeekReached their barBiggest moverCaptured
10–15 Aug 2026 3 of 5 HLEGLAS no
3–8 Aug 2026 3 of 5 ATHERENERG no

How often a name crossed +5%

Of every name published, the share that traded 5% above its pre-result close at some point in the seven sessions after the result. This asks whether a name got there, which is a different question from the Gain rate below — that one asks whether it was still there on the seventh session. A stock that ran to +9% on day two and gave it all back is a yes here and a no there, and both readings are true.

Crossed +5%

60.0%

6 of 10 names measured

Median best point

+6.5%

the middle name’s high, against its pre-result close

Still open

12

result not in, or window still running

Measured on the session high, because “crossed” is about what the stock traded at rather than where it settled. Reaching a level at some point in a week is not the same as a return, and nothing here says it could have been realised.

Dates that moved after the pick

Names are picked before most companies have told the exchange when they will report, so a card is often frozen against a projected date. When the company then confirms a different one, the pick stands, the card is not edited, and the result is scored on the day it actually lands. Every such move is listed here.

Week Name Picked against Confirmed as Moved Date at pick time
12–17 Oct 2026 DMART
Avenue Supermarts Ltd.
15 Oct 10 Oct -5 d Projected
12–17 Oct 2026 ANANDRATHI
Anand Rathi Wealth Ltd.
13 Oct 9 Oct -4 d Projected
12–17 Oct 2026 WIPRO
Wipro Ltd.
14 Oct 15 Oct +1 d Projected