The boundary on our own headline number
ZEME-101 teaches the upper-lock base rate on a closing definition: a lock is a session that closed at its cap. That definition has an edge, and this lesson measures exactly how much sits on the other side of it.
Of the sessions in our intraday forensic record where a stock genuinely touched its upper cap during the day:
202 of 414 = 48.792% still closed there.
Roughly half of everything that reaches the band gives it back before the bell.
Two things this changes
One - "it hit upper circuit today" is a much weaker statement than it sounds. On this sample it is close to a coin flip whether the stock is still there at the close. If you are reading a message saying a stock "circuited", you are being told about an event that reverses about half the time.
Two - our closing base rate is the RIGHT number for a swing horizon and the WRONG number for an intraday one. A closing definition understates how often the band is reached. If your horizon is a single day, the touch rate is your number, not the close rate. The definition has to match the decision.
When it happens, and how much has already traded
49.517% of cap touches happened within 45 minutes of the open, and by the moment the cap was touched an average of 25.480% of that day's volume had already traded. The event is concentrated early, and most of the day's liquidity is not available to you at the price you saw.
Evidence - History - Confidence - Counterargument
Confidence - moderate, and the reason is the sample. This forensic record covers 414 touches, a genuinely small window next to the ten-year price panel behind the rest of this course. Treat the roughly-half figure as an indication, not a constant.
Counterargument - the strongest case against: this record is recent and short, and cap-hold behaviour plausibly varies with the market regime. A calmer or a more speculative market could move it materially. It is published with its n precisely so you can see how much weight it can bear.
Evidence - where every number came from
Every figure in this lesson is listed below with its definition, its sample size, its window and the exact query that produced it, so you can re-run it rather than take our word for it.
touch_held_count = 202 (n = 414)
- what it is: of those, the number that closed at the cap
- window: 2026-07-15..2026-07-31
- source: psx_data.db:lock_forensics
- the exact query:
``sql SELECT COUNT(*), SUM(held_close), SUM(CASE WHEN mins_to_lock<=45 THEN 1 ELSE 0 END), AVG(pct_before) FROM lock_forensics WHERE touched_cap=1 ``
touch_count = 414 (n = 414)
- what it is: sessions observed touching the upper cap intraday
- window: 2026-07-15..2026-07-31
- source: psx_data.db:lock_forensics
- the exact query:
``sql SELECT COUNT(*), SUM(held_close), SUM(CASE WHEN mins_to_lock<=45 THEN 1 ELSE 0 END), AVG(pct_before) FROM lock_forensics WHERE touched_cap=1 ``
touch_hold_rate = 48.792% (n = 414)
- what it is: of the sessions in which a stock actually TOUCHED its upper cap intraday, the share that still CLOSED locked
- window: 2026-07-15..2026-07-31
- source: psx_data.db:lock_forensics
- the exact query:
``sql SELECT COUNT(*), SUM(held_close), SUM(CASE WHEN mins_to_lock<=45 THEN 1 ELSE 0 END), AVG(pct_before) FROM lock_forensics WHERE touched_cap=1 ``
touch_early_rate = 49.517% (n = 414)
- what it is: share of cap touches that happened within 45 minutes of the open
- window: 2026-07-15..2026-07-31
- source: psx_data.db:lock_forensics
- the exact query:
``sql SELECT COUNT(*), SUM(held_close), SUM(CASE WHEN mins_to_lock<=45 THEN 1 ELSE 0 END), AVG(pct_before) FROM lock_forensics WHERE touched_cap=1 ``
touch_pct_before = 25.480% (n = 414)
- what it is: average share of the day's volume already traded BEFORE the cap was touched
- window: 2026-07-15..2026-07-31
- source: psx_data.db:lock_forensics
- the exact query:
``sql SELECT COUNT(*), SUM(held_close), SUM(CASE WHEN mins_to_lock<=45 THEN 1 ELSE 0 END), AVG(pct_before) FROM lock_forensics WHERE touched_cap=1 ``
ZEME Academy is educational analysis of Pakistan Stock Exchange data. It is not investment advice, not a recommendation, and not a solicitation to trade. Every number here is measured history, and history is not a forecast. ZEME is not SECP-registered.