When Should You Sell? Twelve Rules Show the Way
January 25, 2024
11 min read
When Should You Sell? Twelve Rules Show the Way
Buying stocks is easy. But when should you sell? There is no holy grail. There are only methods to increase the probability of being right, preserving capital, and making money. The best offence is a strong defence — and that starts with sell discipline. Our original analysis for The Market by NZZ plus the extended investor's view on sell mathematics and discipline.
In 30 seconds — the core thesis
- Buying stocks is easy. When to sell is the hardest question — and it's structurally under-trained in investor culture. The best offence is a strong defence: without sell discipline, no one wins long-term.
- Three hard statistics calibrate the sell discipline: only 4% of stocks carry the total stock market performance over 100 years. 48% of all stocks perform worse than a 1-month US Treasury. The «point of no return» begins at -30% drawdown — recovery mathematics then becomes exponentially brutal.
- Twelve battle-tested sell rules reduce the risk of being trapped in the bad 96% and crossing the «point-of-no-return» threshold. These rules aren't a holy grail — they are the best available tool for disciplined probability optimisation.
The original analysis — the opening
The best offence is a strong defence.
A nice sports cliché. The funny thing about clichés is that they're often true. The best example is Greece, which won the Football European Championship 2004 because it defended incredibly well and somehow always scored a header from a free kick.
The same applies to investing. Without a strong defence that protects you from large losses, you can't win much investing.
Since I love statistics, here are three things to keep in mind when selecting stocks. First: in the last 100 years, only 4% of stocks were responsible for the total gains of the stock market. Second: 48% of all stocks barely match the performance of a one-month US Treasury bond. Third: the «point of no return» begins at about -30%, meaning the burden of negative performance acts exponentially negatively on your capital.
→ Read the full article with all twelve rules on The Market by NZZ
Chart 1: The law of large numbers — the point of no return begins at -30%
Source: NZZ The Market
01The uncomfortable question: why is sell discipline under-trained in investor culture?
Buy strategies fill libraries — sell strategies are rare. This asymmetry isn't accidental. It's structurally anchored in investor culture, the investment industry, and human psychology. This question doesn't fit into an NZZ column with the necessary depth — but it's the key to understanding: sell decisions activate psychological mechanisms that systematically lead investors to bad results.
Three structural mechanisms explain the underdevelopment of sell discipline:
- Loss-realisation aversion. As long as a stock is in the portfolio, the loss is a «paper loss» — psychologically tolerable. Once sold, it becomes a «real» loss — psychologically more painful, although mathematically identical. Investors hold losers longer than rational because holding avoids psychological pain.
- Identification with one's own selection. Anyone who selected a stock themselves emotionally identifies with it. Selling it means admitting a mistake — psychologically hard because it threatens the self-image as competent investor. This identification is an expensive trap.
- Hope beats probability. «Maybe it'll come back» is the most common irrational sell-avoidance reasoning. Statistically, structurally weak businesses rarely recover — but hope is stickier than statistics. Disciplined investors learn to act against their hope.
The statistical reality is brutal — and honest
If only 4% of all stocks carry the total market performance and 48% don't even beat a 1-month Treasury, then the fundamental investor question is not «which stocks will rise?» but «which stocks should I sell to not be trapped in the bad 48% half?». This reformulation of the investor question is the precondition for sell discipline. Anyone who rejects it fights against statistical reality — and loses against it over decades.
02The recovery mathematics — why -30% is the «point of no return»
The most brutal reality of sell discipline lies in the asymmetric drawdown-recovery mathematics. A 10% loss needs 11% recovery — negligible. A 30% loss needs 43% recovery — substantial. A 50% loss needs 100% recovery — exponentially hard.
The recovery mathematics — why -30% is the critical threshold
| Drawdown | Recovery needed | Reality |
|---|---|---|
| -10% | +11.1% | negligible, normal market cycle |
| -20% | +25.0% | significant, but surmountable |
| -30% | +42.9% | Point of no return — math becomes exponential |
| -40% | +66.7% | structurally hard, often years |
| -50% | +100.0% | doubling needed — rarely achieved |
| -70% | +233.3% | practically irretrievable for most |
| -90% | +900.0% | total loss in practice |
Recovery rate = 1 / (1 - drawdown) - 1. A -50% stock must recover from 50 to 100 = +100% needed.
The mathematics isn't subjective — it's universal. What it shows: before -30%, drawdowns are annoying but surmountable. After -30%, recovery becomes exponentially harder, and the probability of full recovery sinks dramatically. This is the mathematical justification for a sell threshold at or before -30%.
Empirical research over long time spans confirms: stocks that have once fallen -30% need on average 4-7 years for full recovery. Stocks that have fallen -50% need on average 8-14 years — if they reach recovery at all. Over 30 years of investor career, «sitting on» large losers costs dramatically much compound effect.
The uncomfortable consequence for sell discipline
Anyone thinking mathematically consistent sells quality losers at a clear stop-loss of -20% — Thierry's recommendation, well before the «point-of-no-return» threshold. Anyone relying on «sometime it'll come back» hope often loses structurally. The hard truth: realising a smaller loss disciplinedly is mathematically almost always better than hoping for a recovery that's statistically unlikely. That holds for quality losers with real fundamental problems — not for temporary valuation drawdowns with intact quality (cf. tennis-balls-vs-eggs companion).
03The twelve sell rules in overview
Thierry's twelve rules follow a clear structural logic: selling from strength (profit-taking) vs. selling from weakness (loss avoidance). A central accent of his approach: he weights technical signals (price, volume, moving averages) higher than purely fundamental reasons. Why? Because the stock market typically anticipates fundamentals 6-18 months ahead. By the time fundamentals become obvious, losses are often already a fact. That's the hard lesson from cases like Alibaba — since its 2014 IPO, around 20% loss despite revenues growing 1'150%.
Selling from strength — profit-taking (rules 1-2)
1
Never the «too expensive» reflexAs long as a stock behaves right and is in trend: hands in lap, do nothing. Many miss big winners by assuming something is too expensive.
2
Trim on exponential riseThree-stage acceleration, prices 50-70% above the 50-day line or 100-200% above the 200-day line: sell 1/3 to 1/2 — never the whole position.
Selling from weakness — fundamental reasons (rules 3-5)
3
I was wrongOriginal thesis disproved. Prerequisite: written purchase reasons before buying — only then objective review possible. Ego is the enemy.
4
Better opportunityOpportunity costs. Turnaround patience with 30% gain while the broad market made +100% = structurally bad investment.
5
Large M&A / «synergy»Major takeovers or mergers = management focus loss, hidden problems, culture collision. Retreat rather than wait-and-see.
Selling from weakness — technical reasons (rules 6-12)
6
Hard stop-lossSet before purchase, typically -20%. Automatic stop-out, no emotions. The «insurance premium» against -50% scenarios.
7
Trendline / moving average breaksWeekly close below long-term uptrend line, key support, or 50/200-day line (high volume). Weekly charts filter market noise.
8
200-day line turns downThierry's «if I had only one rule» rule. Never fight the trend. With falling 200-day line: no adding, ready to sell.
9
Late-stage signalsElevated volatility + sideways movement after long uptrend + break of key support = late stage. Subsequent corrections often sharp.
10
Don't let a gain flip into a lossAt 20-30%+ gain and beginning drop: set stop at breakeven. Good companies rarely fall so fast so hard.
11
Relative strength erodesUnderperformance vs. index 3-6 months = no more adding. 6-12 months underperformance = strict position review.
12
Bird-strike event~20% drop on bad news + subsequent trend deterioration → sell 50% of position. Halving creates objectivity.
The core philosophy behind the rules
Phil Fisher — one of Thierry's favourite authors («Common Stocks and Uncommon Profits») — put it this way: if the purchase was done right, it's almost never time to sell. The twelve rules therefore don't stimulate activity — they discipline wrong purchases and structural deterioration. Jesse Livermore added: markets are never wrong, opinions often are. When market and thesis diverge, the market is almost always right. Peter Lynch's image — «cutting the flowers and watering the weeds» — shows the most common investor mistake: selling winners too early, holding losers too long. The twelve rules are the tool against exactly this reflex.
The detailed justification of each rule — with concrete chart examples from market history (Alibaba as warning, Nvidia 2021 as exponential rise, Apple August 2023 as trend break, PayPal 2021 as late stage, Microsoft as intact relative strength) — you find in the original article on NZZ.
04What this means for your quality portfolio
The twelve rules must combine with quality selection — they don't replace selection discipline, they complement it. Three strategic implications:
| Implication | What to do |
|---|---|
| 1. Define sell rules before purchase | Before buying a stock, define specific sell triggers: at what price (-20% stop-loss per Thierry's Rule 6), at what fundamental deterioration, at what trend break would this stock be sold. Written down before purchase — not in the emotional heat of drawdown. |
| 2. Quarterly sell-rule review | Check each position quarterly systematically against all twelve rules. Time investment: 5-10 minutes per position. For 25-30 positions: 3-5 hours per quarter. This discipline prevents the gradual holding of losers. |
| 3. Distinguish quality tennis balls from quality-loser eggs | The twelve rules apply primarily to quality businesses that become losers (eggs, cf. tennis-balls companion). Quality businesses with temporary valuation weakness (tennis balls) should be held or added to, not sold. This distinction demands fundamental analysis — not just price observation. |
| Investor behaviour | Sell discipline status | What to review |
|---|---|---|
| "I never sell anything below purchase price" | Structurally trapped in loser trap | Abandon anchor on purchase price, ask fundamental quality question |
| "I sell winners early and hold losers" | Classic behaviour-based anti-pattern | Consciously act against reflex — hold tennis balls, sell eggs |
| "I follow written sell rules strictly" | Structurally superior positioned | Hold discipline, periodically refine rules |
| "I sell panicky in every drawdown" | Other trap — overtrading destroys compound | Hold quality businesses, only sell structural erosion |
How sell-rule logic combines with the tennis-ball-vs-egg selection lens is deepened in our Tennis-Balls companion. The behaviour-gap discipline against loss aversion is treated by the Emotions pillar.
05Three scenarios — how sell discipline impacts lifetime performance
Sell discipline isn't spectacular — but over 30 years of compound effect it's structurally decisive. Three plausible investor paths:
Bull Case
Consistent sell discipline — structural outperformance
Investors with consistent application of the twelve rules systematically avoid point-of-no-return drawdowns. Over 30 years they deliver structurally superior performance vs. investors without sell discipline — typically 1-2 percentage points annual difference, which over 30 years increases the end value by 35-80%. Statistically the best path with disciplined implementation.
Base Case
Partial discipline — average results with improvement potential
Most investors apply sell rules partially — they sell losers sometimes too early, sometimes too late, sometimes not at all. Results are average. Improvement potential is significant but demands written rule definitions and quarterly reviews. Our base case.
Bear Case
Investors without sell discipline — structural underperformance
Investors who never sell or only react panically accumulate losers structurally in their portfolios. Over 30 years an increasing share of their capital is trapped in the bad 48% half. The compound difference to disciplined investors becomes dramatic. Avoidable — but demands written sell rules and consistent discipline.
06What you should review now
A sell-rule inventory of your portfolio takes 60-90 minutes. Four concrete checks:
1. -20% drawdown inventory. Which of your current positions are 20%+ below their 12-month high? That's Thierry's stop-loss threshold. Each of these positions needs an honest quality assessment: temporary tennis ball or structural egg? For each egg diagnosis, sell discipline kicks in before the «-30% point-of-no-return» threshold is reached.
2. Written sell-trigger definition. Do you have written sell triggers for each of your positions — stop-loss level, trendline break, relative-strength threshold? If no — define today. Before the next market stress phase. Written rules help to remain objective in emotional moments.
3. Establish quarterly review routine. Block 3-5 hours per quarter for systematic sell-rule application to each position — especially the 200-day moving-average trend and the relative-strength line vs. index. This routine is the operational translation of sell discipline into portfolio practice.
4. Behaviour-gap self-diagnosis. Have you held losers longer than rational in the last 24 months? Have you avoided sell decisions out of hope instead of analysis? Have you had «too expensive» reflexes on winners that then continued to rise? This self-diagnosis is the starting point for discipline improvement.
What disciplined investors do
They define sell triggers in writing before each purchase. They conduct quarterly systematic sell reviews. They respect the -30% threshold as critical recovery asymmetry and typically sell quality losers before it. They clearly distinguish between tennis balls (hold) and eggs (sell). They accept that even the best investors have 30-40% sell decisions that appear «too early» in hindsight — the mathematics works over the totality, not over individual cases. This discipline isn't spectacular, but unspectacularly consistent. Exactly that makes it structurally profitable over 30 years of compound effect — and distinguishes disciplined investing from hopeful sitting still.
07Frequently asked questions
What are all twelve sell rules in detail?
The twelve rules are summarised above in Section 03. The detailed justification of each rule, with concrete chart examples from market history (Alibaba, Nvidia, Apple, PayPal, Microsoft), you find in the full original article on NZZ. Operational implementation benefits from combining both texts — the rules here as reference, the detailed reasoning and chart context in the original.
Should I also sell quality businesses at -30%?
Differentiated. If the -30% is caused by fundamental quality erosion (market-share losses, moat weakening, management failure), then yes — the sell rules apply regardless of former quality status. If the -30% is purely caused by market sentiment or sector rotation with intact quality, then no — possibly buying opportunity (tennis ball, cf. companion). The distinction demands fundamental analysis.
How do I apply tax-loss harvesting in Switzerland?
Swiss loss-offsetting rules differ from e.g. the US. Retail investors can typically not tax-offset securities losses (capital gains in CH are tax-free, so losses aren't deductible either). Tax-loss harvesting is therefore primarily relevant for investors with «commercial securities trader» status. When unclear: consult tax advisor.
How does arvy apply the sell rules?
arvy holds explicit sell criteria for each position and reviews them quarterly. Sales are transparently documented in the quarterly report (e.g. arvy's complete software sector divestment in Q1 2026 — applied sell discipline in practice). Concrete sell mechanics you find in the arvy Quarterly Report Q1 2026.
Further reading — the thematic anchors of this analysis
The Tortoise Problem
Quality + sell discipline
Tennis Balls and Eggs
Selection lens for sell decisions
Master Your Emotions
Overcome loss aversion
The Next Kodak Moment
Disruption as sell trigger
The Safety Margin
Asymmetry and recovery mathematics
arvy Quarterly Report Q1 2026
Sell discipline in practice
The best offence is a strong defence — the mathematics confirms it
If only 4% of all stocks carry the total market performance over 100 years, then the fundamental investing question is not «which stocks will rise?» but «which stocks should I timely sell to not be trapped in the bad 48% half?». This reformulation changes the entire investor discipline. It transforms reactive hoping into active probability optimisation.
What separates disciplined investors from average ones is not superior market forecasting. It's the willingness to intellectually accept the brutal recovery mathematics — that -30% drawdown marks the point of exponential recovery pain — and act accordingly. The twelve sell rules are no magical solution. They are the best available tool to increase the probability of staying in the good 4% half of the stock market and timely leaving the bad 48% half. This discipline isn't spectacular. It's defensive. But it's the structural difference between investors who build wealth over 30 years and investors who lose wealth over 30 years — trade by trade, loser by loser, recovery hope by recovery hope. Mathematics isn't subjective. It's universal. And it demands sell discipline from anyone who wants to invest successfully long-term.
Original written by Thierry Borgeat, Co-Founder of arvy, for The Market by NZZ. The extended arvy companion piece reviewed by Patrick Rissi, CFA and Florian Jauch, CFA. Data sources: NZZ The Market, own analyses, academic drawdown-recovery research. Last updated: April 2026.
Disclaimer: This article is for general educational purposes and does not constitute personal investment advice. arvy is a FINMA-supervised asset manager with a CISA licence (Art. 24). Imprint & Legal Notice.
