Let winners run or invest with margin of safety?

As you long term readers know by now this blog has two main themes, stock picks and evaluating my investment style. This is another one of those posts where I explore one aspect of portfolio management – should one hold on to winners as their valuation gets more stretched?

In the past year markets have both become easier and harder to invest in, depending on your style of investing. I try to keep an open mind, be adaptive but at the same time invest in a way that works long term. I do believe some very basic fundamentals of investing rules will always stay true. Valuation does matter for me, but so does other things as knowing your holdings well. My portfolio returns have been especially good in the past year (on both absolute and relative terms). This has been driven by a handful of my holdings which have re-rated substantially. As an investor thinking in terms of margin of safety, this margin has in many cases reduced significantly as the stocks have gone up. Fundamentals improved warranting an increase of 20-30%, but the stock has gone up 60%, in other words reducing margin of safety (multiple has expanded). What is the prudent thing to do in such a case? Keep the position and let your winner run, or immediately re-allocate your capital to something which you perceive to have a higher margin of safety? These are for me very difficult questions to answer as manager of my portfolio. Below I want to contrast two different approaches to investing to highlight the issue.

The Coffee Can portfolio

The Coffee Can portfolio concept harkens back to the Old West, when people put their valuable possessions in a coffee can and kept it under-the mattress. The story that explains the concept the best goes like this:

I had worked with the client for about ten years, when her husband suddenly died. She inherited his estate and called us to say that she would be adding his securities to the portfolio under our management. When we received the list of assets, I was amused to find that he had secretly been piggybacking our recommendations for his wife‘s portfolio. Then, when I looked at the total value of the estate, I was also shocked. The husband had applied a small twist of his own to our advice: He paid no attention whatsoever to the sale recommendations. He simply put about $5,000 in every purchase recommendation. Then he would toss the certificate in his safe-deposit box and forget it. Needless to say, he had an odd-looking portfolio. He owned a number of small holdings with values of less than $2,000. He had several large holdings with values in excess of $100,000. There was one jumbo holding worth over $800,000 that exceeded the total value of his wife’s portfolio and came from a small commitment in a company called Haloid; this later turned out to be a zillion shares of Xerox.

Full text can be found here: Coffe Can portfolio

So at its extreme it is not never sell anything you bought, but perhaps a slightly less extreme version is to never sell your winners. Would this be a reasonable investment strategy and something I could really commit to?

Every day is a blank sheet of paper

Another extreme way to see your portfolio was something I learned a long time ago talking to some seasoned portfolio managers. Every day is a blank sheet of paper, they pretended their whole portfolios was in cash. Basically they went through the exercise of pretending they did not own any stocks and they needed to deploy all of their funds in the market every day. If their current portfolio looked different than how they would deploy their money if they started from scratch they rebalanced their portfolio to look like the weights they actually wanted to have. In reality I don’t think they thought of their portfolio like this on a daily basis, but for sure some did this on a monthly basis or so. If one holding in the portfolio for example doubled in a months time and went from 2% weight to 4% weight, they would scale it back to what they then thought was appropriate. Perhaps after such a run-up it was only 1% position or something the sold entirely.

At its extreme this type of portfolio managed will constantly sell winners if the stock price increase was not based on fundamentals that would warrant to hold a larger position. If the stock price just increased its multiple with no new information that should move the stock, this type of investor would quickly sell down that holding and re-allocate the money into other holdings. A true mean-reverting strategy which would work very well if the stock that increased in price came back down again.

Middle road – Thesis Investing

A middle road to this would be to care less about current valuation and just focus on if your long term thesis for the company holds. As long as the thesis holds you never sell. This would mean you could hold a company that is short term grossly overvalued (but still within some limits of reasonableness). I think the investors who have been most successful in the past few years have fully allocated their portfolios into this type of holdings. Leaning on the powerful laws of compounding, for something growing 20-30% YoY valuation matters little if they keep this up for the coming 10 years. Obviously the hardest part is to figure out if they really will keep it up for the coming 10 years – which then means coming back to that the thesis for the investment still holds. The nice thing with thesis investing is that you filter out a lot of noisy and just focus on where the company is going long term – this in its simplicity is appealing to me.

Tesla as a use case

If I just play out the thought process of investing with a Coffee Can approach owning Tesla. Tesla could have been an investment in my portfolio since I looked at the company when it was still trading around 30 USD per share. I was early on the EV theme (as you long term readers know) and of course also looked at the Tesla stock. I even read the book about Elon Musk and how he created Tesla and SpaceX. I was very impressed with him after reading the book and especially impressed with SpaceX (and still am). With benefit of hindsight its obviously a big regret not buying the stock around 30 USD back when I started this blog. Let’s just play with the thought that I did. How would I then reacted when the stock suddenly popped from 35 USD up to 80-90 USD? Back then there were a lot of naysayers, this company is going bankrupt soon, huge pile of debt etc. – risks seemed high and the valuation was already “stretched”. I’m very sure I would have sold close to the 80 USD level and been very happy with such a quick big return, breaking both the Coffee Can and Thesis Investing rule. .

So what kind of investment mindset does one need to have to hold on to a winner like Tesla from 30 USD all the way to 800 USD? Is the Coffee Can strategy of letting your winners run and never sell, the approach one should take to investing? If I used Thesis investing it would have come down a lot to what my thesis is. With the successes Tesla had over the past few years probably the initial thesis would need to updated. An initial thesis back in 2013-2014 could probably have been that Tesla would be one of many future EV car makers (but no market leader). Few people back then believed Tesla would get even that far. I think this is not really a reasonable thesis anymore if one is long Tesla. That would be some type of bear case that Tesla would only be one of many EV makers. Todays valuation is pricing in more than that initial thesis, so it would have been time to sell some time ago, but perhaps one would have held on to the stock up towards 2-300 USD per share using thesis investing.

Talking about updating the thesis, I realize that is the big issue I have with Coffee Can approach. It basically assumes to take one decision at one point in time and then never update your thesis. Obviously there are many factors to consider if Tesla is a buy or sell at current levels. But let’s just take the argument around competition.

What drives Tesla valuation today?

Barring very special circumstances, competition will always eat away at successful companies. If the cake is too big and juicy, the market will never let just one or two companies enjoy the spoils for all eternity. Perhaps the entry barriers to get a seat at the table are high, perhaps the market leaders are extremely good at what they do. In the end people will keep trying to get a piece of that cake. This dynamic means that there is long term tendency for something to mean-revert. A new industry flares up, profits increase rapidly for early entrants and things looks great. More entrants enter the market offering products at lower margins and margins are eventually reduced. Sometimes there are special circumstances where this margin compression does not occur – for example by branding. It doesn’t matter how many watch companies enter the market, they will not be the same as Rolex, not until they built the brand value. So to every rule there are always exception. I try to follow basic principles like competition (and many others) to not get carried away with my investments and dream up blue sky scenarios which are highly unlikely to materialize. These type of arguments are closely linked to valuation. In my view a key point for someone that takes either a long or short position in Tesla is competition and brand value. I think most of us can agree that Tesla’s brand value is extremely strong, basically across all age groups and quite broadly in society. Two very relevant questions to ask is then, will it keep or even increase its brand value over time? And more importantly how much extra are people willing to pay to drive a Tesla instead of say a Mercedes, BMW, Volkswagen or even a Porsche?

These type of thoughts around a holding is in my view essential for investing, otherwise I might just throw darts at a board and put them in my Coffee Can. Just like the original article hints, I think the Coffee Can approach is more of an alternative to index investing and nothing I’m interesting in entertaining.


In the case of Tesla it would have been fantastic to hold it since I considered the stock until today but thinking through this hypothetical investment, makes it clear to me that my investment style if very far away from holding a stock like Tesla from 30 USD to 800+. Have I missed out on some big gains by selling out of some of my stocks? I few years back I evaluated how all holdings I exited performed after I sold them. The strong conclusion that came out of that was that I escaped more failing stocks than missed out on stocks that soared after I sold. Read the whole post here: Look in the rearview mirror.

Everyone wants to find and hold the next Tesla, but the likelihood that it ends up in your portfolio is very low to begin with. After that to hold it through all ups and down needs a very special investment approach which is close to Coffee Can investing or at least Thesis Investing where the thesis is updated to more and more bullish future scenarios as the company delivers and fundamentally gets stronger. I would have benefitted the past years to invest more based on thesis and less on fundamentals. But I believe we are going through a very extreme stock market cycle currently and valuations will come back as a stronger factor for investing. So currently I land somewhere between Blank Sheet of Paper investing and Thesis Investing. I try to give my holdings a little bit of room to be overvalued for shorter periods, but as soon as that overvaluation gets a little bit too stretched I mercilessly sell my holdings although I still think the company is doing great things. To become more long term, perhaps I should move a bit closer to Thesis Investing and a bit further away from the Blank Sheet of Paper approach. I would appreciate my readers input on the topic, always happy to hear your thoughts!

With these thought as a backdrop I will follow-up with a post on how I reason about some of my latest portfolio changes. If you take a peak at my Trade History page you can get a preview of what those decisions were.




6 thoughts to “Let winners run or invest with margin of safety?”

  1. Nice post!

    Curious on your thoughts on how the internet may be creating competitive dynamics that we may have not previously seen. Some new dynamics include scalable platforms / products due to low (almost zero) distribution costs, winner takes all markets, regulation that hasn’t been created for internet businesses.

    Facebook may be an example of a business that may different i) a consumer company that spends very little on marketing (and does not have a brand like Coke / Rolex) ii) a business model where users generate value for other users iii) automated ad-servicing platforms. Facebook generates 35-45% NPAT margins off 71bn revenue. Facebook’s network effects makes it very difficult for competitors to replicate their business model.

    1. Agree that network effects created a new type of moat which makes it very hard to get a seat at the table. I think examples like TikTok shows that its not impossible. For me who have been an internet geek for a very long time, for me its clear that these moats are not impenetrable.

      For example in early days of internet if you wanted to talk to other people online, everyone and I mean everyone used ICQ. Then Windows launched Messenger which was pre-installed with Windows, suddenly when the geeky guys wanted to talk to the pretty girls online they had to resort to Messenger. It didn’t take more than perhaps 5 more years and ICQ was out. Now Messenger is out and Whatsapp / WeChat has been king of the hill in the worlds largest markets as we moved to Smartphones. It didnt need much recently for Signal to make a serious dent in Whatsapps dominance. Similarly I can see that younger users do not really use Facebook anymore, Instagram, Snap etc is much more hot. And then TikTok comes along and users migrate again. I do think there is serious stickiness and strong moats with these businesses but many investors are also fooled that for the long term (10y+) that these companies will continue to have the market leading products.

  2. If you don’t take capital gains into account, then implication of coffee can investing is basically ‘valuation doesn’t matter’.

    I think you have to sell for the right reason. Often Amazon is used as an example. The problem with this is that Amazon never traded above 3x revenue between 2001 and 2018 (besides the dotcom bubble). And they did get kind of lucky with AWS, which is a major portion of why it trades so high now.

    Amazon often traded at 1-2x revenue. If you assume their real net profit margin would come out at about 7.5%, it never really traded above LTM 30x implied earnings. Which is still cheapish for a business that can grow revenue 20%+ a year for a long time while needing little capital to do so, with a long growth runway left to go. Especially if you take capital gains taxes into consideration as well.

    So it would have been a mistake to sell pretty much the entire period if you followed the above logic. But it clearly would not have been a mistake to sell during the dotcom mania when it went well above 3x revenue.

    So this nonsense of just sticking your head in the sand and not selling is not investing, it is gambling. If you cannot come up with a good reason to hold, and you held anyway, and you made a big return, you probably got lucky. And it can be dangerous advice even, when unskilled investors hold garbage stocks all the way down. Of course you will only hear about the ones who actually won big in the end.

    I never invested in Amazon though, so the above is easy for me to say of course 🙂 .

    1. I should probably have added a caveat about tax. For me personally I don’t have any capital gains tax, so I don’t focus much on it, but for others it might be very important. It’s interesting to play out this thinking with a lot of stocks. With my strict valuation assumptions I struggle to invest in something like Amazon. For me it would be a Speculative holding. That was part of the reason I extended my invest style to 3 buckets, allowing me to buy loss making companies

  3. Coffee can investing and never sell are trendy topics now due to the 10 year bull market. My own view is close to yours that valuation matters. For my winners when valuation gets stretched I sell and reduce the position size. I have stopped myself from completely exiting as I used to but I reduce the position. I am not comfortable with other approaches. This works for me. Investors have to know themselves.

Leave a Reply

Your email address will not be published. Required fields are marked *