Trading Activity: Work Out What Actually Trades

Average daily traded value is the volume for a day multiplied by the price, averaged over a stated window. Delivery share is delivered quantity divided by traded quantity. A turnover ratio is annual traded value divided by a capitalisation, and which capitalisation is picked changes the answer substantially. Days for a size is that size divided by daily traded value, and it is a floor.

Four computations, and the instrument for all four is immediately below. The meaning of the four measures, and what a reader is entitled to conclude from any of them, is covered separately. Below are the arithmetic, the counter each number is fetched from, and a calculator that will not report a turnover ratio without also reporting the number it divided by. Without the divisor and the day count that produced it, an output cannot be compared with anything, so every output it prints carries both. Carrying both is not a design preference. A figure published without its divisor is the single most common way a correctly computed trading activity figure goes on to mislead somebody, and this calculator will produce that failure on demand rather than describe it.

The figures it opens on are the illustrative Sarvani Coatings ones, worked through in full further down. On a window of five sessions with nothing unusual in any of them, 8,64,000 shares a day at Rs 486/- is an average daily traded value of Rs 41,99,04,000. Of that day, 2,76,480 shares, or 32.0 per cent, went to delivery, worth Rs 13,43,69,280 in which holdings actually changed. Across 14,400 trades the average trade came to Rs 29,160. Scaled up over 250 stated trading days the year comes to Rs 1,04,97,60,00,000: 90.0 per cent of a whole capitalisation of Rs 11,664 crore, and 189.1 per cent of the Rs 5,552 crore available to trade. A size of Rs 250 crore is a floor of 6.0 days against all traded value and 18.6 days against the delivered value.

Two of its controls exist only so that the two ways these numbers mislead people can be produced rather than described. Push the heaviest session slider to eight times and one session takes 66.7 per cent of the window: the average jumps to Rs 1,00,77,69,600 a day, the turnover ratio on the whole capitalisation goes from 90.0 per cent to 216.0 per cent, and the floor for Rs 250 crore falls from 6.0 days to 2.5 days, on a share whose ordinary session has not moved at all. Strip the divisor out of the headline sentence and the 189.1 per cent it prints becomes indistinguishable from the 90.0 per cent that the very same day of trading also supports.

What do the four computations need before they can start?

Five quantities go into the four computations. Three come off a published report. The other two are decisions the analyst makes and then has to write down. The calculator groups its fields that way and names a document and a line beside every one of them.

The distinction matters more than it sounds. A fetched number can be checked by anybody who goes back to the same report on the same day. A decided number cannot be checked at all unless it is stated, and an output built partly on a decided number is only as reportable as the record of that decision. Consider a shopkeeper reporting that the stall took eleven thousand rupees a day. The very first question is over how many days. Eleven thousand a day across a wedding week and eleven thousand a day across a wet Tuesday month are different claims about the same stall. Same arithmetic, different window, and the window was never in the sentence.

Where do the volume and the price come from, and what is decided?

Both are on the report the exchange publishes for each trading day, side by side, for every scrip that traded. The traded quantity and the price are taken as published. Neither is an estimate and neither is adjusted before use. For a single day, this input is already complete.

The decision arrives the moment more than one day is wanted. Average daily traded value is an arithmetic meanAdd the values up and divide by how many there are. An arithmetic mean is the only kind of averaging done in this calculator, and it gives every day in the window exactly the same weight. of the daily figures, which means every day in the window counts for the same amount, and it means a single enormous day counts for the same amount as a quiet one. Equal weighting is fine over sixty days. Over five days it is not fine at all. One outlierA single value sitting a long way from the rest of the set. In a short window one of these can carry most of the average by itself, so the answer describes that one day more than it describes the stock. can carry most of the answer by itself.

The window length is a decision made and then recorded, and an average daily traded value quoted without its window is an incomplete number rather than a wrong one. A block deal, an index rebalancing, a result day: any of these can put several times the usual quantity through in a single session. Including it makes the average largely a description of that one event, and the calculator shows on screen what share of the window that one session takes. Excluding it quietly decides that the event does not count. Neither choice is the correct one in general, and only the recorded choice is defensible.

Where does the delivered quantity come from?

The same report, from the same venueThe particular exchange a trade was executed on. India has more than one, and each publishes its own day report covering only what happened on its own order book., on the same line. The exchange publishes a deliverable quantity alongside the traded quantity, and the delivery share is simply one divided by the other. There is no adjustment, no lookup and no judgement in this computation at all.

There is one thing that will break it, and it breaks it silently. India has more than one exchange, and each publishes for its own order book only. Taking the traded quantity from both venues and the deliverable quantity from one puts a numerator from one population over a denominator from a larger one, and the ratio that results is not a delivery share of anything. The result will still look like a percentage, still somewhere between nought and a hundred. Nothing on the screen will mark it as meaningless.

Both figures in the delivery share must come off the same report for the same venue, or the ratio is comparing one population with a different one. A genuinely combined picture needs all four numbers, both traded quantities and both deliverable quantities, added before the division.

The division is trivial, and the quantity underneath it is not. Delivery is the part of the day that survives settlementThe step after a trade at which the shares and the money actually change hands. Settlement is the step that turns a purchase into a holding rather than a position closed out the same day. as a change in who holds the share. The rest of the day was bought and sold inside the session. Picture a wholesale vegetable market at four in the morning. The same crate of tomatoes can change hands between five traders before the sun is up, and the market can honestly report a large day, but only one household ends up cooking with it. Traded quantity counts every handover. Delivered quantity counts the kitchen.

Which capitalisation goes underneath the turnover ratio?

Whichever one is stated as used. There are two in circulation for any listed issuer, and both are legitimate. One is the whole capitalisation, the share count multiplied by the price. The other is the free floatThe part of the share count that is actually available to be bought and sold, as against the part held closely and not traded. Working out that proportion is a separate job, done beforehand. capitalisation, which counts only the part of the share count that is actually available to trade. Both are produced by a separate working, covered elsewhere. The number is fetched, typed in, and which of the two was typed is recorded.

The choice cannot be left implicit. The two figures are far apart for most listed companies, and they are far apart here. Dividing the same annual traded value by two numbers of very different size gives two very different answers, and neither is more correct than the other. The two ratios answer different questions. A turnover ratio is a fraction whose denominatorThe number divided by, sitting underneath the line. Changing it changes the answer even when the number on top has not moved at all. is a choice, so the choice travels with the answer or the answer travels alone and useless.

There is a household version of this that makes the trap obvious. Two people both say they spent forty per cent of their money on rent. One means forty per cent of what lands in the account, the other means forty per cent of what is left after tax and the loan instalment. Both sentences are true and the two people are not describing the same situation at all. Nobody would compare those two numbers if the two definitions were written next to them. The comparison only happens because the definitions were left out.

How many trading days is the figure scaled up over?

A turnover ratio compares a year of trading against a stock of value, so the daily figure has to be pushed up to a year before the comparison is made. Pushing it up is one multiplication, by the number of days the market was actually open. The multiplication is the smallest step in the whole computation and the one most often left silent.

The day count is not a constant, it is an input, and the turnover ratio moves proportionally with whatever it is fed. The effect shows plainly when everything else is held at the case figures and nothing moves but the number of days stated.

How is each of the four numbers actually worked out?

1. Average daily traded valueMultiply the traded quantity for a day by the price for that day, then take the mean of those daily values across the window. Not the mean quantity multiplied by the mean price, which is a different number whenever quantity and price move together.
   Record the window length beside the answer.
2. Delivery shareDivide the deliverable quantity by the traded quantity, both from the same report for the same venue. Multiply by a hundred for a percentage. Applying that percentage to the traded value gives the value in which holdings actually changed.
   Record the venue beside the answer.
3. Turnover ratioMultiply the average daily traded value by the number of trading days being scaled up over, then divide by the capitalisation chosen. Doing it twice, once for each version, is the honest thing to do.
   Record the day count and the capitalisation beside the answer.
4. Days for a stated sizeDivide the size in question by a daily value. Use the whole traded value for one answer and the delivered value for the other. Both answers are floors.
   Record which daily value the answer sits on.

What do the case figures give, worked all the way through?

Sarvani Coatings Limited is an invented maker of decorative paints and industrial coatings, and every figure below is illustrative and stamped 28 August 2026. On the stated date it traded about 8,64,000 shares a day at Rs 486/-, of which 2,76,480 shares were taken to delivery. Its capitalisation was Rs 11,664 crore and the part available to trade was Rs 5,552 crore. The day count used is 250 trading days, typed in rather than assumed, and the size being tested is Rs 250 crore.

Read the annual traded value line twice. Rs 1,04,97,60,00,000, or about Rs 10,497.6 crore, is a year of trading in a company whose whole capitalisation is Rs 11,664 crore. The annual traded value is the number sitting on top of the fraction, and it does not change between the next two rows. Only what sits underneath it changes.

How does all of that turn into days for a stated size?

One division, done twice. The size in question is divided by a daily value. Dividing by the whole traded value gives one answer. Dividing by the value in which holdings actually change gives another. The choice between them belongs to the question being asked. The calculator prints both and picks neither.

32.0 per cent of the day is delivered, and one divided by 0.32 is 3.125, so the delivery floor is exactly 3.125 times the traded value floor. The second answer is not a separate estimate arrived at by a different method: it is the first answer divided by the delivery share, and the calculator checks that identity on screen every time a field changes.

How does an analyst actually use these four numbers?

Meghna Iyer, an invented analyst covering coatings, is not computing these to put a paragraph in a note. She is computing them because a portfolio manager has asked whether a position of a stated size can be put on, and every part of that question turns into one of these four outputs, with the days figure the one that answers it in the unit the question was asked in. Her first move on any new name is the one the slider imitates: look at the window before the average. A figure built on a window with one enormous session in it will size a position she cannot actually build.

The person receiving her answer will put it next to a figure computed by somebody else, so what she reports back is never a single number but always a number with its divisor and its window attached. A lender sizing a loan against pledged shares does the same thing for a different reason: the days figure sets how long a sale would take at best, and a floor that arrives without its denominator cannot be stress tested by the credit committee that has to sign it. A household investor holding a small quantity may find the whole computation irrelevant to their own position. Irrelevance is itself a useful conclusion.

What can these four numbers never be made to say?

The calculator computes four numbers and stops. Enough is a property of the question being asked rather than of the share, so no output says whether Sarvani Coatings trades enough. A comparison needs both sides computed on the same definitions, and only one side is ever on screen, so no output compares the issuer with any other issuer. The cost of an exit is a question about price impact, and not one of these five inputs carries any information about price impact at all.

Above all, the two days outputs are floors and not forecasts. The division treats a single participant as the sole buyer on each of those days, and that is never the case. Six days assumes one participant takes every rupee that trades for six consecutive days and nobody else takes any. The 18.6 day figure assumes the same thing about the delivered portion. Both are the best possible case, computed exactly, and the real answer is always worse.

Sarvani Coatings Limited, Nandivarman Paints Limited, Kesaria Surface Solutions Limited, Thottam Chemicals Limited and the analyst Meghna Iyer are invented.
Educational material. Not advice on any investment, tax, budget or market position.