---
title: "HFT, MFT and LFT compared on Nifty data | RupeeCase Learn"
description: "HFT, MFT and LFT as holding periods: a typical Nifty move by horizon against futures charges, SEBI's FY24 algo profit figures and the retail algo limits."
source_url: "https://www.rupeecase.com/learn/path-19/module-19-3-hft-mft-lft"
---

[Path 19 3 of 16](/learn/path-19/)HFT, MFT and LFT compared[Next 19.4](/learn/path-19/module-19-4-colocation-and-latency)

Tuesday 15 Sep 2026. The Nifty closed at 23,118.60. At the average volatility of the 300 sessions in the files, a typical one minute move was 8.67 index points, and a typical one second move 1.12. A futures round trip on that price costs about 13.07 points in statutory charges before any brokerage. So one minute of price movement is smaller than the bill for trading it. That one comparison is why speed is a different business from patience. HFT, MFT and LFT are three answers to it. This lesson sets them side by side with dated NSE numbers and SEBI's own figures, and shows where a retail algo in India can sit.

3By the end you can

1. Say what HFT, MFT and LFT describe and what decides the result at each speed.
2. Scale a volatility to a holding period and set the move against the charges of a round trip.
3. Read SEBI's FY24 algo profit figures for what they count, and place a retail algo on the ladder.

Watch in the terminal: how closely did two lines move together? A short tutorial on sample data.

## The idea, plainly

### What do HFT, MFT and LFT mean?

They are market shorthand for the holding period of a strategy and for what decides its result. They are not licence classes. SEBI's circulars regulate an algo order by how it is placed, tagged and risk checked, whatever label its owner uses. A program is not HFT because it is automated. The label has to match what the system does, and its own logs say that.

* **HFT, high frequency:** positions last seconds or less, and the orders run in microseconds to milliseconds. The edge is speed: where the server sits, how short the code path is, how early an order joins the queue at a price.

* **MFT, medium frequency:** positions last minutes to hours. Signal and execution matter about equally.

* **LFT, low frequency:** positions last days to months. The signal, the costs and the risk carried overnight decide the result. Speed hardly matters.

Each step towards speed moves the edge from the signal to the plumbing

The three labels as a ladder of holding periods. A method diagram with no market figures. Where a real system sits depends on its logs, not its name. Source: RupeeCase illustration of trading frequency bands, no market figures. As of 17 Oct 2026.

| Style and hold | Decided by |
| --- | --- |
| HFT: seconds or less | Latency, queue, feed |
| MFT: minutes to hours | Signal and fills |
| LFT: days to months | Signal, size, gaps |

### Why does the holding period change everything?

Because price moves with the square root of time, while costs are paid per trade. Halve the holding period and the typical move falls by only about 30%, but the charge for the round trip stays put. Shorten it enough and the move is smaller than the charge. The fold below has the arithmetic.

The maths: square root of time
Move = S x vol
x sqrt(t / year)

Take the Nifty close S and annualised volatility vol. A year has 252 sessions of 375 minutes, which is 94,500 trading minutes, and the square root of 94,500 is 307.4. One minute then moves about S x vol divided by that root, one standard deviation. For one second use 252 x 22,500 = 5,670,000 and a root of 2,381. For one session use the root of 252, 15.87.

The rule assumes returns are independent and volatility is the same all day. Neither holds: the open and the close are livelier than the middle. Read the numbers as scale, not as a forecast.

Read one more paragraph, 43 words
Both parts of the comparison are rates, so they can be set side by side. The move is a standard deviation in basis points of the price, one [basis point](/learn/glossary#basis-point) being 0.01%. The charge is statutory, from the rate tables by date.

## Worked example: the move against the bill

Take the 20 day realised volatility of the Nifty, averaged over the 300 sessions to 15 Sep 2026: 11.5% a year. That is calm next to the 29.5% peak in the same files. Scale it down to each holding period and set the result beside the cost of one futures round trip.

A minute of Nifty movement is smaller than the charges to trade it

One standard deviation of the move at the average 20 day realised volatility of 11.5%, scaled by the square root of time with 375 minutes a session, in basis points of the price. The charge bar is a sell and a buy of one Nifty future at the 15 Sep 2026 close: STT, exchange fee, SEBI fee, stamp duty and GST, with no brokerage. Source: NSE end of day data, computed by RupeeCase. As of 15 Sep 2026.
Read 2 more paragraphs, 146 words
The picture is plain. A one second move is 0.48 basis points, a minute 3.75, and the charges of the round trip 5.65. Only from a holding period near fifteen minutes does the typical move run to about 2.6 times the charges. A strategy that holds for a minute and has no edge beyond the move pays more in charges than it can expect to win. It does not mean an edge cannot exist at that speed. It means the edge has to be larger than the charge, every time.

Most of the charge is STT. The rate on the sell side of a futures contract is 0.05% from 1 Apr 2026, which is 11.56 index points at the close and Rs 751.35 on one lot of 65 units. The full table of charges is in [the real cost of an F&O trade](/learn/path-12/module-12-6-cost-of-an-fo-trade).

### Who earns at the fast end?

SEBI's September 2024 study of profits and losses in equity derivatives covers the whole F&O population for FY24. It found that algo entities, defined as any entity that did at least one trade in the year using an algo order, earned 97% of FPI gross profits and 96% of proprietary traders' gross profits. Of 376 FPIs, 306 used algos, and of 626 proprietary traders, 279, by the study's Table 6. Its text says 347, which is the table's count of proprietary traders who did not use algos. Only 13% of the 95.7 lakh individual traders did, and SEBI says that share is overstated because it includes clients whose brokers squared off their positions using algos.

The profits of the professional groups ran through algos, those of individuals did not

SEBI study, Analysis of Profits and Losses in the Equity Derivatives Segment (FY22 to FY24), 23 Sep 2024: gross profit before transaction costs, all FPIs and proprietary traders in the NSE F&O segment in FY24. Counts of algo users from the study's Table 6. An algo entity did at least one trade in the year using an algo order. Algo use is not the same as high frequency. Source: SEBI study of 23 Sep 2024, FY24 data, computed by RupeeCase from the study's counts. As of 23 Sep 2024.
Read one more paragraph, 49 words
Read the chart for what it says. The study counts any algo order, so it does not isolate HFT, and it reports gross profit, before costs. It says that where money was made by the professional groups, an algo was involved. It does not say that algos make money.

### What does each speed have to get right?

Each end has its own way to fail. At the HFT end the risk is operational. A bug or a bad feed can send thousands of orders before a person looks, which is why exchanges keep a kill switch for a single algo identifier and why risk checks sit in the order path itself. At the MFT end the risk is decay and slippage: a signal that worked in a test fades once real fills arrive a few ticks away from the price assumed. At the LFT end the risk is the one the file of closes hides: the gap overnight, the roll every month and the margin that has to be funded while a view plays out. A speed is a choice about which of these risks to carry.

### What did SEBI change about speed?

The rules most about speed date from the circular of 9 Apr 2018, which was written for the fast end. It asked exchanges to give every trading member the tick by tick data feed free of charge, to publish latency statistics including the 50th and 99th percentile, to offer a managed co-location service for smaller members, to tag every order with the identifier of the algo behind it, and to narrow the band around the last traded price in which orders escape the order to trade ratio penalty, from 1% to 0.75%. The aim, in its words, was a level playing field between algo and manual trading. The case behind some of that is in [the colocation module](/learn/path-19/module-19-4-colocation-and-latency).

Watch in the terminal: why did my broker refuse an order? A short tutorial on sample data.

### How do I tell which one my system is?

Read the trade log, not the code. The median holding period tells the band. The gap between a signal and the order reaching the exchange tells whether latency is part of the edge. The ratio of orders to trades tells how much the system quotes and cancels, a habit of the fast end. A log with a median hold of three weeks and a handful of orders a day is LFT, however it was built. Write the band down before testing, then check that the tested edge is bigger than the charge at that band.

### Where can a retail algo sit?

Under the retail framework in force for all brokers from 1 Apr 2026, an investor's own algo that stays at or below ten orders a second per exchange needs no registration with the exchange and runs through the broker's API from the investor's own whitelisted static IP. Algos from empanelled providers are hosted on the broker's servers instead. Either way the orders go through the broker's risk checks and carry an exchange tag. None of that is built for microseconds. A retail API algo sits at the MFT and LFT end by construction. [Building a compliant retail algo](/learn/path-19/module-19-5-building-a-compliant-algo) covers the framework and [the colocation module](/learn/path-19/module-19-4-colocation-and-latency) the exchange side of speed.

## The common mistake

The belief: Faster execution means more profit, so my strategy should be as fast as possible. Speed is a cost, and it pays only where the edge is made of speed. At the average volatility the typical Nifty move in a minute was 3.75 basis points against 5.65 for the charges of one round trip. A strategy whose edge is a slow signal gains nothing from shaving milliseconds off its orders and loses the hours it needs to work. Ask three questions. What is the holding period in the logs? What decides the result at that period? Is that smaller than the charge on every trade? SEBI's study of FY24 shows that the profits sat with a few professional groups that used algos. It does not measure speed, and its profit is gross, before costs.

2 terms this section turns on

From the module glossary; every term is in the Learn glossary too. Source: Module 19.3, HFT, MFT and LFT compared.

## Key takeaways

* HFT, MFT and LFT describe holding period, not a licence class.
* Price moves with the square root of time. Charges are paid per trade.
* A one minute Nifty move was smaller than the charges of a futures round trip.
* SEBI's algo profit shares are gross and count any algo order.
* A retail API algo sits at the MFT and LFT end by construction.
* Automation does not make a program high frequency.

**Before you trade.** Derivatives trading carries a high risk of loss. SEBI's study "Profitability of Individual Traders in the Equity Derivatives Segment (FY25 to FY26)" (August 2026) found that 87.7% of individual traders in equity F&O made net losses in FY26, with aggregate net losses of Rs 91,685 crore. [Read the SEBI study](https://www.sebi.gov.in/reports-and-statistics/research/aug-2026/study-profitability-of-individual-traders-in-the-equity-derivatives-segment-fy25-fy26-_103835.html)

What does HFT, MFT and LFT compared cover?Three speeds of trading, and the one comparison that sorts them: the move you can expect against the bill to trade it. It works through The idea, plainly, Worked example: the move against the bill, The common mistake, Key takeaways. It is module 19.3 of Path 19, Algo, HFT and Backtesting, about 10 minutes of reading, free.

TTry it

## F&O costs and STT, on the same example.
Every statutory charge on a trade at the 2026 rates. Dated NSE end of day data, nothing live, no orders.

**F&O costs and STT**, Nifty, 10 Jun 2026, 30 Jun 2026 expiry.

[Open the f&o costs and stt](/learn/tools/fo-cost-calculator?chain=NIFTY_2026-06-10_2026-06-30&preset=future-round-trip)

  AF&O on RupeeCase

## Hedging only, in development
F&O at RupeeCase will do one job: protect a portfolio you already own, by a rule published before it runs. No speculation, no naked option selling, no leveraged bets. It is not available yet.

  [How hedging will work](/hedging)[Sign up for RupeeCase](https://invest.rupeecase.com/signup)

  [No hedgeFalls with the marketWith a putA floor, for a premiumIllustration with made up numbers, not a product and not a recommendation.](/hedging)

QQuick check

## 10 questions, 7 to pass.
Each attempt draws 10 of the 15 questions in this module's bank, and a retake draws a fresh set. A pass gives this module's Certificate of Completion. Your result stays on this device and counts towards the Path 19 certificate.

7 of 10 correct passes module 19.3

Every answer is explained when you check; a retake is always free. Source: RupeeCase Learn quiz rules (70% to pass).

  Passed

###

  Certificate ID

Download PNGPrint or save as PDF[Check the ID](/learn/certificates/verify)
Self paced and unproctored. A record of completing free education by RupeeCase; not a NISM certificate and not endorsed by NISM, NSE, BSE or SEBI. It qualifies no one to deal, advise or publish research.

CCheck yourself

## Path 19 so far.
Your quick check results on this device, stop by stop.

Stop 3 of 16 on Path 19

Algo, HFT and Backtesting: every live stop, this module ringed. Source: RupeeCase Learn, Path 19.

1. [19.1What a backtest is, and five biases that ruin itNot yet](/learn/path-19/module-19-1-backtest-biases)
2. [19.2Strategy metrics: CAGR, Sharpe, Sortino, drawdown and expectancyNot yet](/learn/path-19/module-19-2-strategy-metrics)
3. [19.3HFT, MFT and LFT comparedNot yet](/learn/path-19/module-19-3-hft-mft-lft#quiz)
4. [19.4Colocation, latency and fairness: NSE's historyNot yet](/learn/path-19/module-19-4-colocation-and-latency)
5. [19.5Building a compliant retail algoNot yet](/learn/path-19/module-19-5-building-a-compliant-algo)
6. [19.6Backtesting options on Indian dataNot yet](/learn/path-19/module-19-6-backtesting-options)
7. [19.7Walk forward and out of sample testingNot yet](/learn/path-19/module-19-7-walk-forward-testing)
8. [19.8Monte Carlo and bootstrap of trade resultsNot yet](/learn/path-19/module-19-8-monte-carlo-and-bootstrap)
9. [19.9Microstructure models: spreads, informed trading and price impactNot yet](/learn/path-19/module-19-9-microstructure-models)
10. [19.10Market making: inventory and quotingNot yet](/learn/path-19/module-19-10-market-making)
11. [19.11Optimal execution and market impactNot yet](/learn/path-19/module-19-11-optimal-execution)
12. [19.12HFT economics and the latency raceNot yet](/learn/path-19/module-19-12-hft-economics)
13. [19.13Backtest overfitting: deflated Sharpe, PBO and multiple testingNot yet](/learn/path-19/module-19-13-backtest-overfitting)
14. [19.14Financial machine learning done carefullyNot yet](/learn/path-19/module-19-14-financial-machine-learning)
15. [19.15Expiry day market integrity: a SEBI interim order, FutEq limits and surveillanceNot yet](/learn/path-19/module-19-15-expiry-day-market-integrity)
16. [19.16Capstone: a research note on an Indian derivatives questionNot yet](/learn/path-19/module-19-16-capstone-research-note)
17. [TPath test, 30 questionsNot yet](/learn/path-19/module-19-16-capstone-research-note#pathTestSection)

Path 19 in reading time: 172 minutes over 16 modules

Minutes of reading per module, this one in amber. Source: RupeeCase Learn, Path 19.

    Explore next

## Put the module to work.

* [ToolOption payoff builderLegs on a dated Nifty chain, breakevens and charges](/learn/tools/payoff-builder)
* [Free module 12.6The real cost of a trade: STT, exchange fees, SEBI fee, GST and stamp duty7 min read](/learn/path-12/module-12-6-cost-of-an-fo-trade)
* [Free module 19.1What a backtest is, and five biases that ruin it9 min read](/learn/path-19/module-19-1-backtest-biases)
* [Free module 19.2Strategy metrics: CAGR, Sharpe, Sortino, drawdown and expectancy9 min read](/learn/path-19/module-19-2-strategy-metrics)
* [Free module 19.4Colocation, latency and fairness: NSE's history11 min read](/learn/path-19/module-19-4-colocation-and-latency)
* [Free module 19.5Building a compliant retail algo10 min read](/learn/path-19/module-19-5-building-a-compliant-algo)

      Community

## Join my WhatsApp group.

      Scan the code, fillFill one short form, and join my WhatsApp group. Slides, sources, factsheets and the rebalance logs live there, and members join a daily Google Meet with me.

      New to RupeeCase? [Getting started](/getting-started) [See the terminal](/terminal)

        [Sign up for RupeeCase](https://invest.rupeecase.com/signup)
        [Join the WhatsApp group](https://forms.gle/C6HMtM1Z2GfeocQV7)

      [One form, one group](https://forms.gle/C6HMtM1Z2GfeocQV7)Scan with your phone camera

SSources

## Go to the source.
8 sources, NSE the most cited

The publisher of each source listed below. Source: Module 19.3 source list.
1. [SEBI, Analysis of Profits and Losses in the Equity Derivatives Segment, FY22 to FY24 (23 September 2024)](https://www.sebi.gov.in/reports-and-statistics/research/sep-2024/study-analysis-of-profits-and-losses-in-the-equity-derivatives-segment-fy22-fy24-_86905.html) SEBI
2. [SEBI circular SEBI/HO/MIRSD/MIRSD-PoD/P/CIR/2025/0000013, Safer participation of retail investors in algorithmic trading (4 February 2025)](https://www.sebi.gov.in/legal/circulars/feb-2025/safer-participation-of-retail-investors-in-algorithmic-trading_91614.html) SEBI
3. [SEBI circular SEBI/HO/MRD/DP/CIR/P/2018/62, Measures to strengthen algorithmic trading and co-location (9 April 2018)](https://www.sebi.gov.in/legal/circulars/apr-2018/measures-to-strengthen-algorithmic-trading-and-co-location-proximity-hosting-framework_38605.html) SEBI
4. [NSE circular NSE/INVG/67858, implementation standards for retail algo trading (5 May 2025)](https://nsearchives.nseindia.com/content/circulars/INVG67858.pdf) NSE
5. [NSE circular NSE/FATAX/73524, STT rates from 1 April 2026](https://nsearchives.nseindia.com/content/circulars/FATAX73524.pdf) NSE
6. [NSE circular NSE/FA/73061, transaction charges for equity derivatives from 1 March 2026](https://nsearchives.nseindia.com/content/circulars/FA73061.pdf) NSE
7. [NSE, algorithmic trading and decision support tools](https://www.nseindia.com/static/trade/platform-services-non-neat-decision-support-tools-algorithm-trading) NSE
8. Aldridge, I., High-Frequency Trading: A Practical Guide to Algorithmic Strategies and Trading Systems, 2nd ed., Wiley, 2013 Book

GTerms

## 5 terms from this module.
High frequency tradingTrading that holds positions for seconds or less and depends on speed: latency, queue position and the data feed.
Holding periodHow long a position stays open. It sets what decides the result of a strategy, from speed to signal and costs.
[Basis point](/learn/glossary#basis-point)One hundredth of one percent. Moves and charges are compared in basis points of the price.
Algo entityIn SEBI's FY24 study, any entity that did at least one trade in the year using an algorithmic order. The term is not a measure of speed.
Threshold orders per secondThe order rate, ten a second per exchange to begin with, at or below which an investor's own algo needs no registration.
[All 230 terms in the glossary](/learn/glossary)

[Question people askHFT, MFT and LFT: what is the difference?A short answer with a dated chart](/learn/questions/hft-vs-mft-vs-lft)[Related module12.6 The real cost of a trade: STT, exchange fees, SEBI fee, GST and stamp duty7 min, Beginner](/learn/path-12/module-12-6-cost-of-an-fo-trade)[Related module19.1 What a backtest is, and five biases that ruin it9 min, Intermediate](/learn/path-19/module-19-1-backtest-biases)[Pillar guideDerivatives and Quant115 modules across paths 12 to 19](/learn/derivatives)

Education only, not investment advice, a research report or a recommendation to buy, sell or hold any security. RupeeCase (QC Alpha Technologies Pvt Ltd) is not registered with SEBI as an investment adviser or research analyst. Contracts and prices shown are for illustration only, come from NSE end of day files at least 30 days old, and are not recommendatory. Derivatives trading carries a high risk of loss.