Maximum Drawdown and Loss Rules

First published 26 Aug 2026 · Last verified 29 Aug 2026

Ram Bahadur had a rule, or so he told himself. "If a stock falls twenty percent, I sell." He had said this to his brother-in-law over dal-bhat more times than he could count. It sounded firm. It sounded disciplined. It sounded like the kind of thing a serious investor says.

Then Nepal Reinsurance fell eighteen percent in a week during a market-wide correction, and Ram Bahadur did not sell. He told himself the twenty percent line hadn't been crossed yet — technically true — but he also knew, in the honest part of his mind he rarely visited, that if it had crossed twenty-two percent he still would not have sold, because by then he would have found a new reason: "it's due for a bounce," or "I already lost so much, no point selling now," or "let me just wait for it to recover to breakeven." The rule had sounded real when the market was calm. It evaporated the moment it was needed.

This chapter is about that gap — the distance between the loss rule you state in a calm moment and the loss rule you actually follow in a falling market. Chapter 81 dealt with exit rules for individual trades: the mechanical, backtested triggers that tell you when to sell one particular stock based on its own price action or valuation. This chapter is different. It is about drawdown, a word borrowed from professional portfolio management that simply means the drop from a peak value to a lower value, measured as a percentage. A drawdown rule is not about any single stock's chart. It is about you — your account size, your sector concentration, your entire portfolio, your nervous system — and about what you have pre-committed to do when the damage reaches a certain size. It sits closer to psychology and financial planning than to technical analysis. It is the last line of defence in your personal investment constitution, the rule that exists precisely because you cannot trust your future self, mid-crash, to think clearly.

Lesson 97.1 — What a Drawdown Rule Is, and Why It Is Not a Stop-Loss

Start with definitions, because this chapter will use several words that sound similar but mean different things.

A stop-loss, covered in Chapter 81, is an order or a rule attached to one position: "If Company X falls to Rs 450, I sell Company X." It is stock-specific. It is usually based on that stock's own price behaviour, its support level, its earnings outlook, or a backtested rule about how far a normal correction in that stock tends to go before it either recovers or breaks down further.

A drawdown limit is different in scope. It applies to a bigger unit — your entire portfolio, a whole sector you are exposed to, or a meaningful chunk of your net worth — and it is triggered not by what one stock's chart looks like, but by how much total value has been erased, measured against a peak. If your portfolio was worth Rs 25 lakh at its highest point and is now worth Rs 20 lakh, your portfolio has experienced a drawdown of 20 percent, regardless of which individual holdings caused it.

Why does this distinction matter enough to build a separate rule around it? Because a portfolio can be perfectly healthy stock-by-stock and still be dangerously exposed at the portfolio level. Imagine an investor who owns eight different NEPSE-listed companies, each with its own sensible stop-loss rule set according to Chapter 81's method. Every individual position is "under control." But six of those eight companies are hydropower producers, and NEPSE's entire hydropower sub-index falls 30 percent over two months because of a change in the government's power purchase agreement policy, or a bad monsoon season that damages river flow forecasts and spooks the sector. Each stop-loss may or may not trigger depending on how it was set, but the portfolio as a whole has taken a blow that no single stop-loss was designed to catch, because no single stop-loss is watching the correlation between your holdings.

This is the first job of a drawdown rule: it watches the forest, not the trees. It exists at three possible levels, and a complete personal operating system usually sets a threshold at each:

Position-level drawdown concentration: how much of your total portfolio is any single stock allowed to represent, and at what loss on that position do you reconsider — not sell mechanically as in Chapter 81, but reconsider your total exposure.

Sector-level drawdown: how much of your total portfolio can be concentrated in one sector (banking, hydropower, insurance, hotels, microfinance), and what happens if that sector as a whole falls sharply.

Portfolio-level drawdown: the big one — the total percentage fall in your entire investable net worth, from its peak, that triggers a pre-committed change in behaviour.

The second job of a drawdown rule, and arguably the more important one, is psychological. Chapter 81's stop-losses answer the question "when do I sell this stock?" A drawdown rule answers a harder question: "when do I admit that my current overall approach, allocation, or risk level is wrong for who I actually am, and change it?" That is not a trading decision. It is closer to a life decision, made about money. It is the difference between bailing water out of a leaking boat one bucket at a time (stop-losses on individual stocks) and deciding the boat itself needs to turn back to shore (a portfolio drawdown limit).

Lesson 97.2 — Sizing Your Threshold to Real Life, Not a Round Number

Ram Bahadur's mistake was not that twenty percent is a bad number. It is that twenty percent was not connected to anything about his actual life. He picked it because it sounded disciplined, the way "I'll start my diet on Monday" sounds disciplined. A drawdown threshold that is not rooted in your real financial situation is a slogan, not a rule, and slogans do not survive contact with a falling market.

A genuine personal maximum drawdown threshold should be built from three inputs, in this order.

First, your time horizon — how many years before you actually need this money. This is the single most powerful input, because time is the resource that heals almost every drawdown that is not caused by permanent business failure. NEPSE's history includes the 2016 correction, the 2021 to 2022 bear market that took the NEPSE index down more than 50 percent from its peak, and various sharper but shorter corrections in between. Every one of those episodes eventually resolved, for the index as a whole, given enough years — though individual companies inside the index sometimes did not recover, which is a separate risk covered elsewhere in this book. An investor with fifteen years until retirement can tolerate a much deeper drawdown than an investor who will need to withdraw money for a daughter's wedding in eight months, because the first investor has time on their side and the second does not.

Second, your need for the money — not just when you will use it, but how essential it is and how replaceable it is. Money earmarked for genuine emergencies, a child's near-term school fees, or loan repayments should arguably not be in equities at all, a point covered in earlier chapters on asset allocation. But even within the portion that is properly invested in NEPSE, some money is more load-bearing than other money. A retired schoolteacher living on a pension supplemented by dividend income from her share portfolio has less room for drawdown than a salaried bank employee in his early thirties who is investing a portion of his monthly income and has years of future paychecks to fall back on if the market falls.

Third, your other income sources — remittance income, a salary, rental income, a spouse's earnings, a family business. An investor whose household receives steady remittance income from a family member working in the Gulf or Malaysia has, in effect, a source of new capital that keeps arriving regardless of what NEPSE does. That is a buffer. It means a drawdown in the portfolio does not automatically mean a drawdown in the household's ability to pay rent, buy rice, or cover a medical bill. An investor with no other income, who is depending on trading gains as their primary livelihood, has no such buffer, and should set a far more conservative threshold.

This table is a starting scaffold, not a universal answer — the household with strong remittance income and a ten-year horizon still needs sector and position limits underneath the big number, because a portfolio can stay under its overall ceiling while being badly concentrated in one dangerous sector the whole way down.

There is a further honesty check worth applying once you have a candidate number: ask yourself not "does this number sound right" but "have I actually lived through a drawdown of this size before, and how did I behave." Many investors who have only ever experienced rising or gently correcting markets dramatically overestimate their own tolerance. They set a 30 percent threshold in their head while having never sat through even a 10 percent one without panic-selling or panic-buying more to "average down" without a plan. If you are newer to NEPSE, set your first real threshold lower than you think you need, and revisit it — as Chapter 98 will describe — once you have actually lived through at least one real correction and observed your own behaviour rather than your intentions.

Lesson 97.3 — Paper Loss Versus Realised Loss, and Why the Difference Changes Everything

Here is a distinction that sounds obvious once stated but trips up even experienced investors constantly.

A paper loss, also called an unrealized loss, is a loss that exists only on your portfolio statement. You bought a stock at Rs 600, it now trades at Rs 450, so your holding shows a loss of Rs 150 per share — but you have not sold. No cash has actually left your net worth in a final, locked-in way. The loss is real in the sense that if you sold today, that is what you would receive, but it is not yet permanent. The price could recover tomorrow, next month, or next year.

A realised loss happens the moment you sell. Once you sell at Rs 450, the Rs 150 per share is locked in. It no longer matters what the stock does afterward — whether it recovers to Rs 700 the following month or falls further to Rs 300. That outcome is no longer yours to gain or lose from, because you no longer own the shares.

Why does this distinction deserve its own lesson in a chapter about drawdown rules? Because it explains two opposite and equally common mistakes, and a good drawdown rule is built to prevent both.

The first mistake is refusing to realise a loss that should be realised. This is loss aversion in its purest form, a concept covered in earlier chapters on investor psychology: humans feel the pain of a loss roughly twice as intensely as the pleasure of an equivalent gain. Because a paper loss "doesn't count yet" in the mind of the person holding it, investors will hold a badly deteriorating position indefinitely, telling themselves it is not a real loss until they sell, so they simply do not sell, ever, and call this patience. Ram Bahadur, from the start of this chapter, was doing exactly this. The stock's fundamentals may have genuinely worsened — a hydropower company losing a power purchase agreement, a bank facing a spike in non-performing loans, a hotel group hit by a prolonged tourism slowdown — but because the loss is "only on paper," the investor keeps waiting for a recovery that a rule-following investor would have already priced out.

The second, less discussed mistake is realising a loss that should not have been realised — panic-selling a fundamentally sound holding purely because the portfolio-level drawdown number looked frightening on a particular red day, without checking whether that specific holding's business case had actually changed. A drawdown rule exists to force a decision at a threshold; it is not meant to force you to sell everything indiscriminately the moment any threshold is crossed. This is why Lesson 97.5 will draw a firm line between what a drawdown rule requires you to do (stop, review, decide deliberately) and what it does not require (sell everything reflexively).

The practical consequence for building your own rule is this: your drawdown threshold triggers a mandatory review, not a mandatory sale. What crossing the threshold actually requires of you is that you sit down, ideally away from the trading screen, and re-underwrite every major holding as if you were buying it fresh today, using the same research standards from earlier chapters on fundamental analysis. Holdings that pass this fresh test can be kept even after the threshold is crossed. Holdings that fail it should be sold — turning the paper loss into a realised one — not because the price fell, but because the re-underwriting showed the business case genuinely no longer holds. The drawdown rule's job is to force the review at a moment discipline requires it. It is not to auto-liquidate your judgment.

Lesson 97.4 — Circuit Breakers and the Illusion of Control

Chapter 90 covered NEPSE's circuit-breaker mechanics in the context of market-wide volatility controls. This lesson returns to that mechanism from a different angle: not "how does the circuit breaker work" but "what does the circuit breaker mean for your ability to actually execute your drawdown rule when you need to."

As a reminder for this chapter's purposes: NEPSE, like most exchanges, has rules that automatically halt trading, either for an individual security or for the market as a whole, once price movement in a session crosses a defined percentage band. These bands exist to slow down panic, give information time to be absorbed, and prevent disorderly price discovery. They are a genuinely useful piece of market infrastructure, put in place by NEPSE and overseen by the Securities Board of Nepal (SEBON) for good reasons.

Here is why this matters for a drawdown rule and not just for a single stock's stop-loss. A stop-loss on one stock, from Chapter 81, is already vulnerable to a circuit halt: if a stock gaps down past your stop-loss level in a single session and hits its lower circuit before you can sell, your stop-loss simply cannot execute that day, because there are no buyers being matched at a price near yours, or the counter is frozen. This is frustrating for a single position, but a drawdown rule multiplies the problem, because a drawdown rule is often triggered precisely during the kind of market-wide event — a sharp macro shock, a monetary policy surprise from Nepal Rastra Bank, a political disruption, a regional shock affecting remittance flows or the broader economy — that is most likely to produce circuit-breaker halts across many stocks simultaneously, not just one.

Consider the mechanics honestly. Your portfolio-level drawdown rule says: "If my total portfolio falls 20 percent from its peak, I will reduce my equity exposure by a third." On the day your portfolio actually crosses that threshold, it is very plausible that a meaningful number of your individual holdings are themselves down their daily circuit limit, meaning trading in them is halted or severely thinned for that session. You cannot sell what you cannot get matched on. Your rule says "act now." The market's plumbing says "you may not be able to, not today, and possibly not for several sessions if the decline continues and each session opens down-limit again before you get an order filled."

What does this mean practically for how you should design and think about your own drawdown rule?

First, it means the rule should trigger on a review and a decision, not on an assumed instant execution, exactly as Lesson 97.3 argued for a different reason. Because you often cannot sell immediately even if you want to, the useful part of the rule is the discipline of stopping to decide while the market is doing the deciding for you through halts, rather than the fantasy that you will cleanly exit at precisely your threshold price, which was already the wrong expectation for a stop-loss and is doubly wrong for a portfolio-level rule spanning many circuit-constrained securities at once.

Second, it means position sizing and diversification, covered in earlier Part XVII chapters, are doing real work here that a drawdown rule alone cannot replace. If your portfolio is concentrated in a small number of thinly traded scrips, a circuit halt genuinely traps you — there may be no exit at any reasonable price for days. If your portfolio is spread across enough liquid names, some of your holdings will likely still be tradable even on a day when others are frozen at their limit, giving you at least partial ability to act on your rule while you wait for the frozen names to open up.

Third, it means the rule should specify not just a threshold but a plan for the days after the threshold is crossed, since execution may be staggered across several sessions rather than completed in one. A well-built rule says something like: "Once my portfolio drawdown crosses 20 percent, I begin the re-underwriting review immediately, and I execute any resulting sales across whichever sessions allow me to, prioritizing the positions I have already decided to exit, without waiting for a 'better' day to start."

Fourth, and this is worth stating plainly because it cuts against a natural instinct: circuit breakers, by design, slow down exactly the kind of panic-driven, indiscriminate selling that a poorly designed drawdown rule might otherwise trigger. If your rule had said "sell everything the instant the portfolio falls 20 percent," the circuit-breaker system would have partially protected you from executing that reflexive decision at the worst possible moment, by simply making it impossible to do so instantly. This is one more argument, on top of the paper-loss-versus-realised-loss argument from Lesson 97.3, for building a rule that triggers a considered review rather than a reflexive mass sale. The market's own volatility controls are, in effect, nudging you toward the more disciplined version of the rule whether you designed it that way or not.

Lesson 97.5 — Building Your Personal Drawdown Rulebook

It is time to put the pieces together into something you can actually write down. A complete personal drawdown rulebook has four components: the thresholds themselves at each of the three levels described in Lesson 97.1, the mandatory action tied to each threshold, an execution acknowledgment reflecting the circuit-breaker reality from Lesson 97.4, and a re-underwriting checklist reflecting the paper-loss discipline from Lesson 97.3.

Start with position-level concentration and drawdown. Decide, in calm conditions, the maximum percentage of your total portfolio that any single stock should represent — a common range for individual investors is 10 to 20 percent per position, tighter for less liquid or more speculative names, looser for a small number of core holdings you have researched deeply and hold with high conviction. Separately, decide what a severe single-position loss — say, a stock you hold falling 30 to 40 percent on its own, independent of the wider market — should trigger: not necessarily an automatic sale, since Chapter 81's stop-loss framework already governs mechanical stock-specific exits, but at minimum a mandatory re-underwriting review of that specific holding within a set number of days.

Move to sector-level exposure. Nepal's listed market is heavily weighted toward a handful of sectors — commercial banks, development banks and finance companies, microfinance institutions, life and non-life insurers, and hydropower — and it is very easy for an investor's portfolio to become far more concentrated in one or two of these than they realise, simply because that sector had been performing well and the investor kept adding to winners. Set a maximum percentage of total portfolio value for any single sector, commonly somewhere between 25 and 40 percent depending on how correlated you judge your other holdings to be with that sector, and set a rule for what happens if that sector as a whole — tracked through NEPSE's published sub-indices — falls sharply: a mandatory review of every holding in that sector together, since a sector-wide decline often reflects a shared cause (a regulatory change from NRB affecting bank capital requirements, a change in power purchase agreement terms affecting hydropower broadly, a monsoon or drought pattern) that is worth understanding as a group rather than stock by stock.

Finally, the portfolio-level rule, built from Lesson 97.2's inputs: your overall maximum acceptable drawdown from peak portfolio value, and the specific action tied to it. The action should be concrete and specific enough that you cannot talk yourself out of it later through vague language. "I will reduce my equity exposure by moving 20 percent of remaining equity value into fixed deposits or government securities" is concrete. "I will be more careful" is not a rule at all.

The re-underwriting checklist itself, referenced twice in the table above, should be short enough that you will actually use it under stress rather than abandon it. A workable version asks, for each holding under review: has the specific reason I bought this company changed (its earnings trend, its management, its regulatory environment, its competitive position)? If I had cash instead of these shares today, would I buy this company at the current price? Is the price decline explained by this company's own fundamentals, by its sector, or by the whole market — and does that distinction change my answer? What would I need to see to change my mind again in either direction?

One more component belongs in the rulebook: an explicit statement of what the rule does not require. It does not require selling a fundamentally sound holding purely because a threshold was crossed. It does not require ignoring the circuit-breaker reality and assuming instant execution. It does not require perfection — a rule followed imperfectly, three weeks late because of circuit halts and a slow re-underwriting process, still beats no rule at all followed with perfect hindsight-driven excuses.

Lesson 97.6 — Rupa's Rule: A Worked Example From a Real Downturn

To make all of this concrete, follow one investor through an actual application of a drawdown rule.

Rupa is a 34-year-old high school administrator in Pokhara. She has been investing in NEPSE-listed shares for six years, funded partly from her salary and partly from remittance support her younger brother, working in Qatar, sends home to the family, a portion of which the family agreed she could invest on behalf of household savings. She has no debts, a small emergency fund in a savings account separate from her investments, and does not expect to need this invested money for at least seven years, when her daughter will begin university.

In early 2021, following the earlier framework in Lesson 97.2, Rupa sat down and wrote her personal drawdown rulebook. Her time horizon was long (seven-plus years), her dependence on the invested money for near-term needs was low, and her household had a second income stream through remittances that was not tied to NEPSE at all. Based on that honest assessment, she set her portfolio-level maximum drawdown threshold at 25 percent from peak value, with a pre-committed action: at 25 percent, reduce total equity exposure by moving one-quarter of remaining equity value into a fixed deposit at her bank, and conduct a full re-underwriting review of every remaining holding. She also set a position-level rule (no single stock above 15 percent of portfolio value; any stock falling more than 35 percent on its own triggers review) and a sector rule (no sector above 35 percent of portfolio value; a 20 percent sub-index decline in any sector she held triggers a group review). She wrote all of this in a notebook, dated it March 2021, and did not look at it again for months, because the market was doing well and there was no reason to.

By late 2021, NEPSE had reached record highs, and Rupa's portfolio, which had grown to include several banking stocks, two hydropower companies, and one life insurance holding, had grown well past her original contributions. Her peak portfolio value, reached in September 2021, became the reference point her drawdown rule would measure against going forward.

Then the correction that ran through 2021 into 2022 began. NRB's tightening of margin lending rules and liquidity conditions, alongside broader macroeconomic pressure including a widening trade deficit and pressure on foreign exchange reserves that prompted import restrictions, pulled the index down sharply and steadily. Rupa's portfolio fell alongside it — not evenly, since her hydropower holdings fell faster than her banking holdings in the early phase, then banking holdings fell further as the lending-rule tightening bit specifically into that sector.

By January 2022, Rupa's portfolio had fallen 22 percent from its September peak. Her rule had not yet triggered. This was, in its own way, useful information: she was watching the number weekly (not daily — she had also decided, wisely, that checking a falling portfolio every single day was its own form of self-harm and had committed to a weekly check-in instead), and she noticed that at 22 percent, her instinct was already to want to sell everything, well before her own pre-committed 25 percent line. This is worth pausing on, because it illustrates something important about drawdown rules generally: the rule's value was not only in what it eventually triggered, but in giving her a benchmark against which to notice that her emotional urge to act was running ahead of her own considered judgment. She held, because her rule — written by a calmer version of herself eight months earlier — had not yet said to act, and she trusted that earlier version of herself more than the frightened version reading the portfolio statement in January.

By March 2022, the portfolio had fallen 27 percent from peak. Her rule had triggered.

Her re-underwriting review produced three different outcomes across her holdings, which is itself an important lesson: a drawdown rule triggering does not mean selling everything. Her two largest banking holdings, after review, still looked sound — solid deposit bases, provisioning that appeared adequate given disclosed non-performing loan figures, and a valuation that now looked, if anything, more attractive than it had at the September peak. She kept both, in full. One of her hydropower holdings had genuine company-specific trouble layered on top of the sector-wide decline: a delay in its plant's commissioning timeline that had been disclosed in a company filing, pushing expected revenue out by over a year. She sold this one, realising a loss of roughly 40 percent on that specific position — a loss that had existed on paper for weeks but that she now, deliberately, converted into a realised one because the re-underwriting review showed the original investment case no longer held. Her life insurance holding and remaining hydropower holding she kept, after concluding their declines reflected sector-wide sentiment rather than company-specific deterioration.

The proceeds from the sale, together with the pre-committed quarter of her remaining equity value, went into a fixed deposit at her bank, exactly as her March 2021 rule had specified. This was not a large sum in absolute terms, but it served its intended purpose: it stopped the household's total drawdown from deepening further on that portion, and it gave Rupa something concrete to point to — a decision made, not merely worried about — during a period when the news and her friends' WhatsApp groups were full of far more dramatic, far less useful reactions.

The lesson from Rupa's experience is not that a 25 percent threshold, or her specific sector and position rules, are the correct numbers for every investor. They were correct for her situation: her time horizon, her other income, her low dependence on this specific money. The lesson is the shape of the process — a threshold set honestly in calm conditions and written down, a trigger that produces a review rather than a reflexive sale, an acceptance that execution would be slower and messier than the rule implied because of circuit-breaker mechanics, and a willingness to actually use the rule when the moment came rather than finding reasons, the way Ram Bahadur did, why this particular fall was different and the rule did not really apply yet.