Deciding whether to buy, sell or hold a stock comes down to testing a thesis written in advance against new information: results, valuation, the position's place in the portfolio. A simple framework has three steps: write the thesis before buying, set in advance what would invalidate it, then review it on fixed dates. Its aim is not to find the right answer every time, but to reduce the part emotion plays in decisions that remain the investor's own.
Why decide by rules rather than instinct?
Behavioural finance has documented several biases that weigh precisely at the moment of deciding:
| Bias | What it causes | Source |
|---|---|---|
| Loss aversion | A loss weighs about twice as much as a gain of the same size | Kahneman and Tversky (1979, 1992) |
| Disposition effect | Winners are sold too early and losers held too long | Shefrin and Statman (1985), Odean (1998) |
| Anchoring | The purchase price becomes a reference, although it says nothing about the future | — |
| Confirmation bias | We keep the information that supports our position | — |
| Overtrading | The more we trade, the more costs and mistakes pile up | Barber and Odean (2000) |
Across tens of thousands of US retail accounts, Barber and Odean found that the households that traded the most earned the lowest net returns. Rules set in a calm moment do not remove these biases, but they make them visible when they strike.
Step 1: write the thesis before buying
A thesis is a short, checkable answer to "why this company, at this price, now?". Writing it down forces the difference between a conviction and a hunch.
| Element | Question | Example (Company E, fictitious) |
|---|---|---|
| The idea | What does the market underestimate, in my view? | Operating margin can climb from 12% to 15% by end-2027 thanks to the new plant |
| The evidence | Which figures support it? | Last half-year margin, order book, announced capital spending |
| The horizon | How long should the thesis take to play out? | Three years |
| The price | What am I paying for this idea? | P/E of 14, against 18 for the sector median |
| The weight | What share of the portfolio to start, and what cap? | 6% to start, never more than 8% |
Step 2: set in advance what would invalidate the thesis
This is the most often skipped step, and the most useful. Invalidation criteria concern the company, not the share price:
- Company E's operating margin falls back below 11% for two quarters in a row;
- net debt exceeds three times EBITDA;
- the new plant is delayed by more than a year;
- a competitor wins the main customer.
A falling share price is not, in itself, an invalidation criterion: it may reflect the whole market. It does always deserve a question: "what has changed in the company?".
Step 3: review the thesis on fixed dates
Reviewing a thesis at every price move is exhausting and leads to overtrading. Reviewing on fixed dates (each results release, or each quarter) sets a rhythm. At each review a few situations come up:
| Situation | Useful questions |
|---|---|
| Results release | Do the figures go the way the thesis said? Has an invalidation criterion been hit? |
| Sharp price rise | Is the valuation still consistent with what the thesis can deliver? Has the position grown past its cap? |
| Sharp price fall | What has changed in the company, beyond the price? |
| A new investment idea | Would this capital be better used elsewhere, at comparable risk? |
| Need for cash | Which position to sell first, and with what tax consequences? |
One question helps neutralise anchoring on the purchase price: "knowing what I know today, would I choose to hold this position at this price and this weight?". The price originally paid plays no part in the answer.
Example: two reviews of Company E
One year after buying, two fictitious scenarios:
| Scenario 1 | Scenario 2 | |
|---|---|---|
| Price since purchase | +50% | −25% |
| Operating margin | 12.5% | 13.5% |
| Invalidation criterion hit? | No | No |
| P/E | 20 | 9 |
| Weight in the portfolio | 8.7% (cap: 8%) | 4.6% |
In scenario 1 the thesis is moving slowly but the price has already risen a lot: the valuation now exceeds the sector's and the position is above the cap set at the start. The questions are about price and weight, not about the quality of the business. In scenario 2 the thesis is progressing faster than expected while the price has fallen: the question becomes what the market sees that the thesis ignores. In both cases it is the grid written in advance that structures the thinking, not the emotion of the moment.
Holding is a decision too
When the information is mixed, doing nothing is often the most reasonable choice. Every trade has a cost: brokerage fees, the bid-ask spread, tax on gains depending on the account (see PEA or CTO). A decision forced on weak information often hurts performance.
Position size matters as much as the pick
A good idea badly sized can damage a portfolio. A few general principles recur among most investors: setting a cap per position and per sector, not confusing conviction with certainty, and looking at the portfolio as a whole rather than line by line (see diversification and risk management). The right level depends on each person's situation, which only the investor — possibly with an authorised adviser — can judge.
Where tools help, and where they stop
A research tool can feed every one of these steps: verified figures to write the thesis, opposing arguments to test it, tracking of results and weights to review it. In InvestIQ, a stock analysis gives a conviction score from 0 to 100 and what moves it (methodology), the War Room sets bull and bear analyses against each other, the portfolio audit shows concentration and weights, and alerts flag a price level being crossed. None of these makes the decision: they are data and readings, to be set against one's own thesis.
In short
Buying, selling or holding is best decided in a calm moment: a written thesis, invalidation criteria set before they are needed, reviews on fixed dates and one simple question to break free from the purchase price. The framework does not guarantee the outcome; it makes decisions explainable, and therefore correctable.
Educational guidance only, not investment advice.
Tools and data for this topic
Frequently asked questions
The purchase price says nothing about the company's future. One question helps break free from it: knowing what we know today, would we choose to hold this position at this price and this weight? Tax on the gain or loss, however, is a concrete part of the decision.
A short, checkable answer to "why this company, at this price, now?", with the expected horizon and the criteria that would prove it wrong. Writing it before buying lets it be judged on facts later.
Holding is a decision in its own right, often the most reasonable one when the information is mixed. Every trade has a cost in fees, bid-ask spread and sometimes tax.
Many investors review them on fixed dates, for instance at each results release or each quarter, rather than at every price move, which leads to overtrading.
