Why are we so bad at selling?

Wednesday, Jun 05 2019 by
Why are we so bad at selling

Do you put more thought into buying a stock or selling it? Presumably, it’s the former because it’s official: we’re bad at selling stocks. 

Or at least professional portfolio managers (PMs) are, according to a report called Selling fast and buying slow: heuristics and trading performance of institutional investors.

The study tracks 4.4 million buy and sell events across 783 portfolios between 2000 and 2016. These professionally managed portfolios had, on average, some $573m of assets under management - so you would think the people in charge of these portfolios are pretty good at what they do.

And they are - when it comes to buying. But the study uncovers a striking pattern: PMs demonstrate clear skill in buying but their selling decisions underperform substantially. In fact, the study finds that the fund managers in the dataset would have been better off dumping shares completely at random.

This doesn't mean the next time you want to make a trade you should just put on a blindfold, start clicking, and see what happens - but If you think you have been faring better than these PMs, how do you know? Have you been tracking the performance of sold shares? Perhaps setting up a portfolio of sold holdings might make for an interesting experiment...

One of the most revealing insights in Selling fast and buying slow comes from the study’s interviews of fund managers. The anecdotal evidence from the sample points to PMs thinking very differently about the two decisions: buy trades represent their latest stroke of genius whereas selling is just a way to raise cash.

If that’s how they think of selling, why would they spend much time on it?

Why we're bad at selling #1: no focus and low conviction

The good news is that we have the skill to sell well, we just don’t focus on it. We put much more effort into buying things. 

Or, in the study’s own words (deep breath), ‘the stark discrepancy in performance between buys and sells appears to be driven by an asymmetric allocation of limited cognitive resources such as attention towards buying and away from selling.’

One way the study suggests this is by the fact that selling decisions on earnings announcement days outperform those on non-announcement days ‘by more than 200 basis points over a yearly horizon.’ The researchers interpret this to mean that,…

Unlock this article instantly by logging into your account

Don’t have an account? Register for free and we’ll get out your way


As per our Terms of Use, Stockopedia is a financial news & data site, discussion forum and content aggregator. Our site should be used for educational & informational purposes only. We do not provide investment advice, recommendations or views as to whether an investment or strategy is suited to the investment needs of a specific individual. You should make your own decisions and seek independent professional advice before doing so. The author may own shares in any companies discussed, all opinions are his/her own & are general/impersonal. Remember: Shares can go down as well as up. Past performance is not a guide to future performance & investors may not get back the amount invested.

Do you like this Post?
39 thumbs up
0 thumbs down
Share this post with friends

18 Comments on this Article show/hide all

Mike888 6th Jun 1 of 18

Simple rule, when you buy a stock, know what would make you sell that same stock. You only make money selling, you make no money holding.

| Link | Share
Jameshickman 6th Jun 2 of 18

For every holding ask yourself: Would I buy this today if I didn't already own it ?
If the answer is no then its time to sell

| Link | Share
wilkonz 6th Jun 3 of 18

Great article. Thanks Jack. I was particularly struck by the observation:

One way the study suggests this is by the fact that selling decisions on earnings announcement days outperform those on non-announcement days ‘by more than 200 basis points over a yearly horizon.’

This is something I hadn't realised - but I do wonder if the stark and significant contrast between announcement days and non-announcement days is supported by other studies.

| Link | Share | 1 reply
gmtrader 6th Jun 4 of 18

Nice read.
I'm guilty. Bad at selling. I.ve picked some real beauties and sold for 20% just for them to go on and multibag. Most recent was my 34p LUCE buy :((

| Link | Share
peachy47 6th Jun 5 of 18

I am a hold merchant! What goes down has got to come up - WRONG.

| Link | Share
Gromley 6th Jun 6 of 18

In reply to post #481361

This is a very interesting, and important, subject area and there are so many things that could be said, but for now I'll just pick up on one of the comments.

wilkonz :

One way the study suggests this is by the fact that selling decisions on earnings announcement days outperform those on non-announcement days ‘by more than 200 basis points over a yearly horizon.’

This is something I hadn't realised - but I do wonder if the stark and significant contrast between announcement days and non-announcement days is supported by other studies.

I have been looking at something similar to this over the recent past, off and on. Sadly more off than on so I don't have anything to report back yet although I do hope to roll out a minor sub branch of this work over the next month or so, which might provide a useful base for further research.

In the meantime though I think we have to be quite clear what Jack's quote actually says. The study looks at trades that were actually made not those that might have been made.

Sales made on an announcement day fall into three categories :

  1. Sales as result of a profits warning.
  2. Sales as a result of a sharp rise "outing" the value.
  3. Other.

I would imagine that the first categories are by far the most common, so perhaps this tells us that sales made for a specific reason are the best?

More specifically though I would expect that sales made as a result of profits warnings are the biggest category and if we compare these with sales made a while after a profits warning, I can think off the top of my head of two useful pieces of work :

Firstly Stocko's own Profit Warning Survival Guide - probably the key lesson from this is that on average initially under-react to profit warnings - so selling promptly on the day is usually better than hanging on to see what happens (or worse trying to 'catch a falling knife'.

Secondly the book by Lee Freeman-Shore The Art of Execution (other booksellers are available) - I cannot find the link currently, but there is also an execellent YouTube presentation on the subject by Leon Boros which is well worth a watch IMHO.

To horribly oversimplify the message of this book : Don't be a rabbit (dazzled in the headlights) execute and move on.

Perhaps a useful adage should be "don't stop and think - sell" (as opposed to it's sister : "don't buy - stop and think")

| Link | Share | 1 reply
Richard Cook 6th Jun 7 of 18

Sell immediately when stock would no longer would be selected as a buy, irrespective of length of time you owned it. I'd like to see this backtested on some of the top strategies.

| Link | Share | 1 reply
Gromley 6th Jun 8 of 18

In reply to post #481646

Sell immediately when stock would no longer would be selected as a buy, irrespective of length of time you owned it. I'd like to see this backtested on some of the top strategies.

That's a very similar observation to that made by jameshickman earlier in the thread.

I genuinely do not mean this to be rude (and I hope it won't be taken as such), but that is a 'sound-bite' not a strategy. (If you buy a stock because, for example, it's PE of 9.99 is less than 10, would you sell it at 1.01 for a 2% gain?)

However, you do both lean towards a very valid point - apply the same rigour to the decision as to whether to hold as you do to decision as to whether to buy. But you need to understand where to set your parameters.

If you are investing solely on ratios, that may be 'comparatively easy' - for example if a PE of <10 earns a "buy merit" then perhaps a PE of >12 should earn a "demerit". For some of the "pure screens" I do think that would be an interesting back-test to see some data on.

If there is an element of "qualitative interpretation" in your decisions I think it becomes a little more difficult, but the reminder to apply the same rigour IS I think important. - I'm sure I am not alone in the tendency to make or accept excuses for 'minor' failings in a company's performance once they are in the portfolio. To break that mental block is I would say one thing to focus on.

But also when do you conduct the review? Certainly on any trading or earnings announcement, but what about every 'X' days or on any 'Y%' movement in the share price.

Putting some structure to this I am sure would be of benefit - personally I know that I suffer from the dual sins of : holding too long and selling too soon.

| Link | Share
zeibots 7th Jun 9 of 18

In reply to post #481511

Yes Gromley, The Art of Execution is an excellent book and it demonstrates that there are lots of Rabbits even amongst the professionals. It is important to have the Assassins mentality when it comes to selling.
However it all starts when you buy, you need the confirmation that the stock is actually moving in the right direction when you press the button. For this, you need to refer to Minervini`s trade like a Stockmarket Wizard. Those of you who keep a complete record of all your buys and sells can do no better than read the advice contained in pages 276-282. You will be astounded how this discipline improves your portfolo`s performance.

Finally, all this discussion is about fundamental and emotional factors. Why not look at the technicals before buying. In this way the timing of sales becomes much more obvious. You really need to look at both the fundamentals as well as the technicals to overcome the risks associated with being a one legged man in a butt kicking contest.

| Link | Share
Maddox 8th Jun 10 of 18

Hi Jack, well done for putting this article together - must haven taken quite a bit of time to distill down the study.

Selling is obviously a very problematic challenge for investors and we don't seem to spend nearly as much time on it as buying - which is in essence reflected in the studies findings. One point that needs to be kept in mind with this report is that the PM's performance is measured against a random comparitor. How much worse the PM's performance might have been if compared against a well-designed sell-strategy?

One thing I do to try and improve my own approach is to input mirror (equal and opposite) trades into a Decision Monitoring Portfolio - so at least I can review my decisions over time. I also input potential buying decisions that I reject or do not act on. Investing is as much a battle with your own psychological biases as one with Mr Market.

Regards Maddox

| Link | Share
justinian 8th Jun 11 of 18

I presume the "selling stock to raise cash" may have something to do with Risk Management popping round, tapping you on the shoulder and advising that you have exceeded your VAR limits. Both using VAR limits and STOPs inevitably mean one has to sell when stock prices are under stress.

| Link | Share
mojomogoz 8th Jun 12 of 18


As we are emotional when buying but try to be rational when selling?

Whereas we should be rational when buying and emotional when selling.

| Link | Share
andrewdb 8th Jun 13 of 18

I have tracked my own decisions since '11 and can see (like others) that my buying decisions are a bit 'above average' and selling decisions were 'woeful' -> more recently 'nearly average'

There seems to be an assumption here:
When you sell, it is with the intention of reinvesting.
This is not true in this place (or if you are neil woodford :). Some people need the money to eat etc.  some people hold lots of cash.

The only way you can tell if you have sold 'well' or 'woefully' is in comparison with something else. The paper uses 'a broad based index' - I take that to mean something like the ASX - which is part of what i do.
The paper states it ignores cash - not a good idea.

This definition has a weakness:
The real test of whether or not you sold badly is what you did with the money.

My definition: over a 12 month period did what I repurchase perform more or less well than what I sold 

- often less well in the early years -> yes more recently.
(for me the test gets more flattering more recently and over longer timeframes) 

I am a big fan of rules (as you are all stocko subscribers, you probably are too).

The general guidance to sell on a profit warning has worked for me.

Other metric - based rules are less appealing ( like selling if the PE > 20 etc ) are not attractive as they only talking about that one stock and has no link to the value of money or the rest of the market.

I like the idea of 'sell if stock rank falls below 50' type rules, but the link between stock rank and performance is on a portfolio basis - not that one stock - and is not as clear as it was a few years ago.

The other qualitative rule that benefits me and would have helped immensely (particularity with small caps):

Get out v.v. quickly at a fixed stop loss without stopping to think or just buckle in for the long haul.
Do not hang on for a while, lose faith, get depressed, and then bail - as the article says.

| Link | Share
brinrich 9th Jun 14 of 18

I'm bad at selling, hate it, i tend to hold on to stocks far too long until they have a problem then i sell, i know i should have a plan . What i need is a plan which i can stick to,especially with my high yielding stocks,it becomes difficult to sell when dividends are due.

| Link | Share
Graham Ford 9th Jun 15 of 18

So, if selling on announcement days gives a better result my question is at what price are we selling on announcement days? Is the data looking favourable for this approach because it is from institutional investors who manage to sell as soon as the market opens when an announcement has poor results or just selling anytime during announcement day? If the share price hits your stop loss part way through announcement day rather than at the start have you lost the advantage of the effect described because the price has already fallen too far?

| Link | Share
Dickyfit 10th Jun 16 of 18

It would be interesting to see how a pure play criteria buying strategy like NAPS would work with a pure play sell/re-buy strategy e.g. if StockRank less than, say, 40 then sell and re-buy the highest ranked stock that meets the NAPS diversification criteria. The lower cut-off is arbitrary based on when historic performance starts dropping away badly but could be based more rationally on buy/sell costs balanced against the variance of the difference in expected returns, if that were available.

| Link | Share | 1 reply
Nick Ray 11th Jun 17 of 18

In reply to post #482571

I did a quick plot of the 2018 NAPS stocks performance and Stock Rank over the year 2018.


Only one stock gets as low as SR=40 (VOD).

INDV which is the worst performer over the year has a SR which briefly dips to 68 and then rises again.

Obviously you can tinker with the rules further. Maybe look at Q V M  separately, etc, etc. But the more interesting philosophical point is that inventing rules to exit a position early are far from intuitive.

(VLE is missing from the SR plot because I don't have any data for that stock (too small))

| Link | Share
Lincoln Green Imp 11th Jun 18 of 18

Feel like selling today. It's wet! Not stopped for two days. Good article.

| Link | Share

Please subscribe to submit a comment

Stock Picking Tutorial Centre

Related Content

Let’s get you setup so you get the most out of our service
Done, Let's add some stocks
Brilliant - You've created a folio! Now let's add some stocks to it.

  • Apple (AAPL)

  • Shell (RDSA)

  • Twitter (TWTR)

  • Volkswagon AG (VOK)

  • McDonalds (MCD)

  • Vodafone (VOD)

  • Barratt Homes (BDEV)

  • Microsoft (MSFT)

  • Tesco (TSCO)
Save and show me my analysis