Most traders can name the stock they sold too early. Fewer can tell you how much they lost holding a stock they should have exited, because that number is spread across months of “it’ll come back”. Both habits often grow from the same root: regret aversion, the urge to pick whichever option protects us from feeling foolish later, even when it isn’t the option with the better odds.
This piece looks at what the research actually says about regret, why it pushes us towards selling winners and sitting on losers, and how a simple regret table can make a messy exit decision a lot clearer. The numbers in the worked example are hypothetical. The research figures come from the sources linked in each section.
In this article
- What regret aversion means
- Where the theory came from
- We overestimate how bad regret will feel
- How regret aversion drives the disposition effect
- Where it shows up on an Indian trading screen
- A regret table for a partial exit (hypothetical example)
- Practical ways to stop regret from running your exits
- Questions to ask before your next exit
- Key takeaways
What regret aversion means
In decision theory, regret aversion (also called anticipated regret) describes how the emotion of regret shapes choices made under uncertainty. The Wikipedia entry on regret in decision theory sets it out neatly. When we decide without full information, we often feel regret if we later find that another choice would have done better. That regret can be measured as the gap between what our actual decision delivered and what the best decision would have delivered in hindsight.
The key word is anticipated. We don’t just feel regret after the fact. We see it coming and steer away from it in advance. That anticipation can push us towards choices that minimise the chance of regret even when they aren’t the best bet on a plain expected-value basis.
Regret can come from two directions. There’s regret of commission, where we acted and it went wrong, and regret of omission, where we passed on something that worked. Traders know both. The first is the stock you bought at the top. The second is the breakout you watched and didn’t take. We covered the second flavour in detail in our piece on omission bias.
Where the theory came from
Regret theory was developed in 1982, independently, by Graham Loomes and Robert Sugden, by David E. Bell, and by Peter C. Fishburn. Their models add a regret term to the usual utility calculation. That term falls as the outcome you actually got improves, and rises as the best alternative you missed improves. In plain terms, how you feel about a result depends not only on what you got but on what you could have had.
The research summary lists several patterns that anticipated regret helps explain where standard expected utility theory struggles, including status quo bias, inaction inertia and the habit of dodging any decision that makes a better “what if” easy to picture.
One experiment described in the same entry is a good illustration. Offer someone a sure $40, or a coin toss that pays $100 for a correct call and nothing otherwise. Taking the sure $40 cuts risk, but it also cuts regret, because the coin usually isn’t tossed and you never learn what you missed. Choose the toss and lose, and the $0 stings. If the coin is tossed regardless of what you chose, there is no option that removes the possibility of regret.
That last line matters for traders. In markets the coin is always tossed. Whatever you do, the screen shows you what the price did next.
We overestimate how bad regret will feel
Here’s the twist. Anticipated regret tends to be bigger than the regret people actually experience, especially for choices they feel responsible for, and especially when they miss an outcome by a narrow margin.
The research summary describes a study of commuters. They predicted they would feel more regret missing a train by one minute than by five minutes. Commuters who actually missed their trains by one or five minutes reported equal, and lower, regret than predicted. They had underestimated how readily they would blame outside causes.
Swap the train for a trade and it sounds familiar. We imagine the agony of booking profit a day before a 10% move, and that imagined pain is enough to keep us in a position well past our plan. When it actually happens, most of us shrug and move on faster than we expected.
In markets the coin is always tossed. Whatever you decide, the screen shows you what you missed, so no choice is regret-proof.
How regret aversion drives the disposition effect
The best-documented place regret turns up in investing is the disposition effect: the tendency to sell winners too early and hold losers too long. We have a full explainer on the disposition effect itself, so here we’ll stick to the regret angle.
Hersh Shefrin and Meir Statman coined the term in a 1985 paper. They argued the behaviour rests on four psychological pieces together: loss aversion from prospect theory, mental accounting against the purchase price, regret avoidance tied to pride, regret and responsibility, and self-control conflicts. In their view, none of the four on its own produces the full effect.
The regret piece works like this. While a position is open, a loss is only on paper. Selling closes the mental account and turns it into a realised loss, and that is the moment the regret lands, made worse because we have to admit we made the call. Selling a winner does the opposite. It locks in a little hit of pride. So the books get trimmed on the green side and left alone on the red side.
Terrance Odean put numbers on this in a 1998 study of 10,000 accounts at a nationwide discount brokerage, covering January 1987 to December 1993. He compared the proportion of gains realised (PGR) with the proportion of losses realised (PLR). Outside December, PGR was about 14.8% and PLR about 9.8%. Put another way, a stock that was up was almost 60% more likely to be sold than one that was down. In December, the month Shefrin and Statman had tied to tax-loss selling, the pattern flipped: PGR fell to 10.8% and PLR rose to 12.8%.
That December reversal is a useful clue. When people are given a reason to reframe a loss, here as a tax saving, they find selling losers much easier. The regret hasn’t vanished; it’s been given a different label.
Where it shows up on an Indian trading screen
None of these studies used Indian data, so treat what follows as observation, not statistics. Still, the patterns are easy to spot.
Booking profit at the first green day
A swing trade is up 4% against a planned target of 10%. The fear of watching it give back gains gets louder than the plan, and we sell. Regret of commission avoided. The trade’s actual edge never got a chance to play out.
Averaging down to “fix” a loser
A position is down 12%. Selling would make the mistake official. Buying more lowers the average price and keeps the account open, which feels like action without the sting of admitting error. This often overlaps with the sunk cost fallacy.
Freezing before a results announcement
Quarterly results are due tomorrow. Exiting means regret if the numbers are good. Holding means regret if they disappoint. Many traders resolve this by doing nothing and calling it a decision, which is regret aversion choosing the option that feels least like a choice.
Chasing after missing a move
The flip side is regret of omission. A stock runs without us, and the urge to “not miss it again” pulls us into an entry far from any sensible stop. The regret of the earlier miss is now dictating a fresh trade.
A regret table for a partial exit (hypothetical example)
Regret isn’t only a bias to fight. Decision theorists also use it as a tool. The minimax regret approach, first presented by Leonard Savage in 1951 according to the same research summary, picks the option whose worst-case regret is smallest. The source’s own example compares stocks, bonds and money market returns under three interest rate scenarios and finds bonds give the lowest worst-case regret. We can run the same logic on a trade.
Say we bought 200 shares at ₹500 and the stock now trades at ₹560, a paper profit of ₹12,000. We’re torn between selling everything, holding everything, or selling half. Picture three possible outcomes over the next few weeks: the price rises to ₹640, stays at ₹560, or falls to ₹480. All of these numbers are made up for illustration.
First, total profit on the trade in each case:
| Choice | Price goes to ₹640 | Price stays at ₹560 | Price falls to ₹480 |
|---|---|---|---|
| Sell all 200 now | ₹12,000 | ₹12,000 | ₹12,000 |
| Hold all 200 | ₹28,000 | ₹12,000 | −₹4,000 |
| Sell 100, hold 100 | ₹20,000 | ₹12,000 | ₹4,000 |
The best possible result is ₹28,000 if the price rises, and ₹12,000 in the other two cases. Regret is the gap between that best result and what each choice delivered:
| Choice | Price goes to ₹640 | Price stays at ₹560 | Price falls to ₹480 | Worst regret |
|---|---|---|---|---|
| Sell all 200 now | ₹16,000 | ₹0 | ₹0 | ₹16,000 |
| Hold all 200 | ₹0 | ₹0 | ₹16,000 | ₹16,000 |
| Sell 100, hold 100 | ₹8,000 | ₹0 | ₹8,000 | ₹8,000 |
Selling half halves the worst-case regret in this setup. That’s why scaling out feels so good psychologically. It doesn’t mean selling half is the profit-maximising choice. That depends on how likely each scenario is, which the regret table deliberately ignores. As the source points out, minimax regret has the advantage of not needing probabilities, but probabilities are exactly what a trading edge is built on.
So use the table to see what you’re really trading off, not as an autopilot. If your setup has a genuine edge to the upside, a full hold may be right even with a bigger worst-case regret. If you have no view, a partial exit at least makes the trade-off explicit instead of leaving it to whichever fear is loudest that afternoon.
A regret table won’t tell you which way the stock goes. It tells you what you’re really afraid of, and that is often enough to make the decision on purpose.
Practical ways to stop regret from running your exits
Decide exits before entry. Write the stop and the target, or the rule for trailing it, when you place the trade. Shefrin and Statman described stop-loss orders as a self-nudge against exactly this problem: a decision made in advance, so there’s less of a fresh, painful choice later. Our risk management guide covers how to size those stops.
Let rules do the selling. In an experiment by Weber and Camerer (1995), cited in the disposition effect summary, subjects tended to sell winners and keep losers, but when shares were sold automatically after each period, the effect was greatly reduced. Good-till-triggered stops and pre-set target orders are the retail trader’s version of that.
Reframe the loss. The December finding in Odean’s study shows that a new frame can make selling losers easier. Ask “would I buy this today at this price?” instead of “how do I get back to my buy price?” The buy price is history. Only the next move matters.
Journal the trades you didn’t take. Regret of omission thrives on selective memory. We remember the one breakout we skipped that ran 20%, and forget the five that failed. A plain log of skipped setups and what they did next brings the real picture back.
Judge the decision, not the outcome. A sale made according to plan is a good decision even if the stock rallies afterwards. If you grade yourself on hindsight, you’ll keep feeding the regret you’re trying to manage.
Scale out on purpose, not by mood. If partial exits suit your style, build them into the plan, for example half at the first target and the rest on a trailing stop. Then the split is a rule, not a nervous reaction.
Questions to ask before your next exit
- Am I selling because the setup is done, or because I’m scared of giving back gains?
- Am I holding because the thesis still stands, or because selling would mean admitting I was wrong?
- If I had no position, would I buy here?
- Which outcome am I most afraid of, and is that fear bigger than its actual probability?
- Did I write the exit rule before I entered, and am I following it?
Key takeaways
- Regret aversion is the tendency to choose options that limit future regret, even when they aren’t the best bets on expected value.
- Regret theory was developed in 1982 by Loomes and Sugden, Bell, and Fishburn, and it helps explain status quo bias and inaction inertia.
- People tend to overestimate how much regret they’ll feel, especially over near misses.
- Regret avoidance is one of the four elements Shefrin and Statman linked to the disposition effect, and Odean’s 1998 data showed winners were sold far more readily than losers outside December.
- Regret tables and the minimax regret idea make trade-offs visible, but they ignore probabilities, so they support decisions rather than make them.
- Pre-set exits, automatic orders, a skipped-trade journal and judging process over outcome are the practical defences.
Disclaimer: This article is for educational purposes only and is not investment advice. The worked example uses hypothetical figures. Markets carry risk; please do your own research or consult a SEBI-registered adviser before making investment decisions.





Leave a Reply
Join the conversation
Commenting is open to registered readers. Sign up with just your email — no password needed, and it takes under a minute.
Sign up to comment Log in