Investing, trading and gambling — and the mistakes that repeat
The answer
The 3 activities look identical on a screen. They differ in one thing: where the money you make comes from.
Investing takes a share of what a business earns. Trading takes money from other participants. Gambling pays a fixed cut to a house that has arranged the odds against you. The first is positive over time by construction. The second is zero before costs and negative after them. The third is negative by design.
Why this costs you money
Most people who lose money in markets are not gambling on purpose. They are doing one activity while believing they are doing another.
A person buys a stock intending to hold it for 5 years. Three weeks later it is down 12% and they sell. That was not investing. The holding period was 3 weeks, so the return had to come from other participants, which makes it trading — but without any of the things trading requires: an entry rule, an exit rule, a position size, and a measured edge.
A different person buys weekly index options because the amounts are small. The contract expires in 4 days. Nothing about a business can be analysed over 4 days. The outcome depends on a short-term price move plus a fixed cost per trade, which is the structure of a casino game with a slightly better payout table.
The cost is not that trading is wrong. Trading with a genuine edge is a profession. The cost is that doing one activity with the tools and expectations of another guarantees the worst version of both. You take the short-term risk of trading and the passivity of investing. You pay trading's costs and receive investing's patience without investing's holding period.
The published evidence on this is unusually clear in India, and it is in the case section below.
How it works
The test is a single question: who pays me if I am right?
| Where the money comes from | Sum across all participants | What produces the edge | |
|---|---|---|---|
| Investing | the profits, dividends and growth of a business | positive, because businesses create value | choosing businesses, and holding long enough for the value to show |
| Trading | the participant who took the other side | zero before costs, negative after | a repeatable pattern, sized and executed consistently |
| Gambling | the other players, minus a fixed cut for the house | negative, always | nothing. The cut is the design |
Two ideas make this usable.
Edge is a reason to expect a positive result that survives being written down and tested. "The stock will go up" is not an edge. "Companies with rising operating cash flow and falling debt outperform over 3 years, and here is the evidence" is a candidate for one.
Expectancy is the arithmetic of the edge:
expectancy = (how often you win × the average win) − (how often you lose × the average loss) − costs
Everything follows from that formula. You can be right 30% of the time and make money, if the wins are much larger than the losses. You can be right 80% of the time and lose everything, if the 20% are large enough. Being right is not the objective. Positive expectancy after costs is the objective.
Now the mistakes, which are the same 7 in every market and every decade. Each one destroys expectancy in a specific way.
- Acting on tips. A tip has no expectancy, because you cannot test it and it comes with no exit rule. When it falls, you have nothing to decide with.
- Following the crowd. By the time an idea is common, the price contains it.
- Leverage. Borrowed money converts a temporary fall into a permanent loss, because somebody else decides when you sell.
- No plan. Without written entry, exit and size rules, every decision is made in the emotional state the market is currently producing.
- Panic selling. Selling at the bottom converts a fall into a realised loss and removes you from the recovery.
- Chasing what has already risen. Buying because something went up, with no rule for when to sell, is momentum with the exit removed.
- Overtrading. Costs and taxes are certain and returns are not.
Numbers 1, 2, 4 and 6 are failures to define an edge. Numbers 3 and 5 are failures of size and behaviour. Number 7 is a failure of arithmetic. None is a failure to predict the market.
What it tells you, and what it does not
This distinction tells you which questions to ask about any activity you are about to start. If the money comes from a business, the questions are about the business and the price. If it comes from another participant, the questions are about your edge, your costs and your consistency.
It does not tell you that trading is bad or that investing is safe. A long-term investor holding an overpriced, badly financed company loses money slowly and completely. A trader with a tested rule and strict size control can do well for a career. The categories describe the source of the return, not the quality of the person.
It also does not divide the world neatly. Buying a stock for 6 months on a valuation argument is genuinely both. Market making is trading with a real, measurable edge. The test is not the label. The test is whether you can name the source of the return and whether the arithmetic works after costs.
And it does not say that a loss means you did something wrong. That is the subject of the next section.
The decision rule
Every loss is one of 4 things, and each has a different fix. Applying the wrong fix makes the next loss worse.
- Bad thesis. The reason you bought turned out to be false. Fix: improve
the analysis. Go back to step 2 and step 6 of the end-to-end pass.
- Bad sizing. The reason was reasonable, but the position was large enough
that being wrong hurt badly. Fix: the size rule. The analysis needs nothing.
- Bad timing. The reason was right and remains right, but you needed the
money, or you sold at the low. Fix: the plan, the emergency fund, the holding period.
- Bad luck. The reason was right, the size was right, and something
genuinely unforeseeable happened. Fix: nothing. Do it again.
The rule: name the type before you change anything. Most people respond to every loss by studying harder, which fixes only the first type.
Try this now
Five minutes, and it is uncomfortable, which is why it works.
- Open your trade history or your profit and loss statement for the last 2 years. Most brokers and demat providers give this as a report.
- Find your single worst loss in money terms. Not percentage. Money.
- Write down 4 facts: why you bought, what percentage of your portfolio it was at the time, how long you held it, and why you sold.
- Now classify it: bad thesis, bad sizing, bad timing, or bad luck. Pick one. If 2 apply, pick the one that cost the most.
- Write the specific fix for that type only, in one sentence, from the list above.
What you should see. Most people classify their worst loss as bad thesis, and most of them are wrong. Look at the percentage you wrote in step 3. If the position was more than 10% of the portfolio, the size decided the outcome more than the analysis did, and studying harder will not help.
If you sold within 3 months of buying something you meant to hold for years, the type is bad timing, and the fix is a written holding period and an emergency fund, not more research. And if you cannot remember why you bought it, the classification is already complete. There was no thesis to be bad.
Do this for your worst 3 losses. A pattern usually appears by the third one, and the pattern is the finding.
Three real cases
1. SEBI's studies of individual traders in equity derivatives, 2023 to 2026 (India) — measured, published, and repeated SEBI has published this analysis 3 times, and the figures are worth quoting exactly because they are usually repeated loosely.
- Press release 2/2023, dated 25 January 2023, covering FY 2018-19 and FY 2021-22, using all individual clients of the top 10 stock brokers, which accounted for 67% of individual client turnover in the equity futures and options segment in FY 2021-22: 9 out of 10 individual traders in the equity futures and options segment made net losses in both years. Loss-makers had average net trading losses of about ₹50,000 in FY 2021-22, and spent an additional amount equal to about 28% of those losses on transaction costs. 98% of individual traders in the segment traded options in FY 2021-22.
- Press release 22/2024, dated 23 September 2024, covering FY22 to FY24: 93% of more than 1 crore individual traders in equity futures and options made losses, averaging around ₹2 lakh per trader over the 3 years, with aggregate losses above ₹1.8 lakh crore and about ₹50,000 crore paid in transaction costs.
- Press release 50/2026, dated 20 August 2026, covering FY25 and FY26: 87.7% of individual traders made losses during FY26, with aggregate net losses of about ₹91,685 crore against about ₹1.12 lakh crore in FY25, and active individual traders falling about 20%, from 98.1 lakh in FY25 to 78.6 lakh in FY26.
There is no exact equivalent published in the United States. That is a real difference between the 2 markets, and it is discussed below.
2. The GameStop episode, January 2021 (United States) — all 3 activities in one stock GameStop shares started January 2021 around $17 and reached an intraday high of $483 on 28 January 2021. Melvin Capital, which held a short position, was reported to be down about 30% by 28 January and about 53% for the month, and received $2.75 billion from Citadel and Point72. On 28 January several brokers, including Robinhood, restricted purchases, citing collateral requirements at the clearing house. In the same stock, on the same days, some participants were investing on a view about the company, some were trading a squeeze with defined exits, and many were buying because the price was rising and had no exit at all. The third group produced most of the losses. The stock was never the variable.
3. The Indian securities scam, exposed 23 April 1992 — the tip that was true, and still ruinous On 23 April 1992 the journalist Sucheta Dalal published a column in The Times of India exposing the methods used by the broker Harshad Mehta to move money from the banking system into shares. Prices had risen enormously in the months before, and enormous numbers of small investors bought near the top on the basis of tips and reported success. The information circulating was, in a sense, accurate: prices really were rising and one man really was behind much of it. That is what makes it useful. A tip can be true and still be the worst possible basis for a decision, because a tip supplies an entry and never an exit.
The question that resolves it
A novice looks at a position and asks: will this go up?
An expert asks: if this goes up, who is paying me, and can they keep doing it?
If the answer is "the company's profits", you can hold for years. If the answer is "somebody who buys it from me later at a higher price", you need an exit rule before you enter, because you are dependent on a person who does not know you exist.
What would make this wrong
If the 3 activities were the same, then the results of long-term equity holders and short-term derivative traders would look similar over long periods. They do not, in either country, in any published data.
The honest limits.
The categories are not morally ranked. A skilled trader with a tested edge, strict sizing and low costs is doing something legitimate and difficult. An investor who buys a poor business at a high price and holds it for 10 years has done something that sounds respectable and lost money anyway.
The published Indian data covers the equity derivatives segment specifically. It says nothing about outcomes for people who buy shares and hold them, and it is frequently quoted as though it did. Do not extend it beyond what it measured.
The 4-way classification of a loss is a tool, not a science. Real losses are often 2 types at once. The value is in being forced to choose, because choosing directs the fix.
And a good decision can produce a loss while a bad decision produces a gain. Judging your process by a single outcome is the mistake underneath most of the others.
In India
Retail participation in India is heavily weighted towards derivatives, and the regulator publishes data on the outcome. That combination is unusual.
The instruments concerned are index and stock futures and options traded on the NSE and BSE. They are leveraged: a small amount of margin controls a much larger exposure, so the same percentage move produces a much larger percentage change in your money. Options expire, and an option that expires out of the money is worth zero, which means a position can go to a complete loss without the underlying index doing anything dramatic.
SEBI has taken a series of measures aimed at this segment, including changes to contract sizes, expiry structures and margin requirements.
Two Indian mechanics are worth naming, because they change the arithmetic. Securities transaction tax is charged on transactions, which raises the cost of high turnover. And the speculative business income treatment of intraday equity trading means intraday profits are taxed differently from delivery-based gains, and intraday losses can be set off only against speculative income.
In the United States
The structure of retail participation is different. A large share of American household equity exposure sits inside mutual funds, exchange traded funds and retirement accounts, much of it in target-date or index products bought automatically from salary.
Retail derivatives trading exists and has grown, including very short-dated index options. But there is no American regulator publishing a study equivalent to SEBI's, giving the percentage of individual traders who lost money in a defined segment over a defined period. What an American reader has instead is academic work using individual brokerage datasets, which is older and narrower, and firm disclosures in other jurisdictions that do not apply in the United States.
The United States does have a rule that sets a floor on who may trade actively: the pattern day trader rule requires a minimum equity of $25,000 in a margin account for somebody making 4 or more day trades in 5 business days. It does not stop losses. It sets a size below which the activity is not permitted at all, which is a different kind of protection from disclosure.
Where they differ, and what that tells you
The single most useful difference is the data itself.
India has an unusual asset: a regulator that studies the outcomes of individual traders in a specific segment and publishes the numbers with the period and sample stated. That means an Indian investor can check a claim about retail trading outcomes against a named document, rather than against an opinion.
The United States has no exact equivalent. American discussion of retail trading outcomes relies on academic studies of individual broker datasets, which are older, narrower and less current.
What that tells you is which kind of argument to trust in each market. In India, when somebody tells you what percentage of traders lose money, ask for the press release number, the financial years and the sample. Those exist, so the question is answerable, and a claim that cannot answer it is not evidence.
The second difference is where the money sits. Indian retail participation leans towards derivatives, which have an expiry and leverage built in. American retail participation leans towards funds and equities bought through payroll deduction, which have neither. The same person, with the same temperament, will produce different outcomes in the 2 systems, because the default product in each country carries a different amount of built-in risk.
Carry this
- Ask one question: who pays me if I am right? A business, another participant, or a house with a fixed cut.
- Expectancy, not accuracy. Being right often is not the objective.
- Every loss is bad thesis, bad sizing, bad timing or bad luck. Name it before you fix anything.
- Doing one activity with another's tools produces the worst of both.
- The 7 classic mistakes are all failures of definition, size or arithmetic. None is a failure to predict.