Five Rules for Picking Winning Stocks

FIve rules for picking winning stocks, taken from our How to Pick Winning Stocks course

In this blog, we look at five techniques which investors can use to hone their analytical skills and improve their investment performance. Some of them, notably the first, are in my view actually rules which every investor should follow. We have had lots of requests from investors to produce more free content of this type to help keen amateur investors improve their skills, so if it’s popular, we may make this a regular feature – let us know.

And if you are in any doubt about the importance of picking winners – Do Global Stocks Outperform US Treasury Bills? by Hendrik Bessembinder and colleagues concluded that the top performing 1% of stocks provided all the gains in stocks in the 1990-2018 period. $61tn in wealth was created, but only 1.3% of US stocks and under 1% of stocks outside the US created all the wealth. Over half of the 62,000 stocks in the sample underperformed one-month US government bonds. In today’s environment, with more gains accruing to winners, a framework for picking winners and avoiding losers has never been more important. Here are just a few tips from our How to Pick Winning Stocks course.


If a company is worth the sum of its future cash flows, then it’s obviously helpful to have an idea of what it is producing today as a sustainable level of cash generation – of course this may grow in future, but using a sustainable free cash flow measure gives you a more powerful valuation multiple than using the P/E, without the need to do a full blown DCF.

I am a bit geeky so you would expect me to make a lot of modifications to the reported numbers, for things like tax settlements, pension top-ups and other non-recurring items. You don’t need to go to this level of detail – if you make two simple adjustments (well they are simple in theory but not always in practice), you will produce a reasonable number.

1                     take out growth capex

2                     adjust for stock-based compensation

There are other adjustments you can make, but these are two of the most important, especially for modern tech stocks. The theory is quite simple:

  1. To the extent that a company is investing or future growth, incremental capex to build a new facility is a cash spend over and above what is required to achieve a steady state of revenue. Some companies will explain what their growth and maintenance capex is, the latter being the figure which is required for a steady state valuation.

  2. Stock-based compensation is generally added back to “adjusted” earnings, but it’s often a large item and it’s not as if this is cost-free to shareholders – there is a dilutive impact. This cost needs to be reflected in the sustainable cash flow calculation.

As I mentioned, the practical implementation is more difficult, but there are a couple of simple tricks that you can do to get a rough feel of the approximate level of sustainable cash flow. And these are short-cuts, only rough and ready indicators. In practice I would generally do a lot more work, but sometimes I will use these to give an initial view, for example when initially deciding if an idea is worth pursuing.

  1. Use the depreciation and amortisation number if there is not a better way of determining growth capex. It sounds slightly counter-intuitive, using a depreciation figure to derive a cash flow estimate, but I have found that it can often be a reasonable substitute. The amortisation is only that which related to capitalised intangibles – software and R&D usually.

  2. For the stock based compensation, this is much harder, and if this method gives an unusual result, I may take the average over several years or refine the calculation further – this is a complicated subject and worthy of a blog on its own. Again this may raise some eyebrows but I look at options granted last year times the average share price – this gives me an approximation of the equivalent cash cost of buying back the stock to offset the dilution. It’s not perfect but it’s better than ignoring the cost, in my view.


If you do want to calculate sustainable free cash flow, you need to have a sensible capex number. Don’t make the mistake this hapless Alphabet analyst did:



This analyst has forgotten capex entirely! Sentieo puts this number at $25bn in 2018 and $24bn in 2019 (and depreciation and amortisation at $9bn and $12bn, respectively). This makes quite a difference to the valuation. I actually feel slightly sorry for the analyst and I am not sure which firm is responsible – I picked this up from Twitter and forgot to note the name of the person (please let me know so that I can credit you!).

I use this quick estimate of sustainable free cash flow to give an indication of the steady state FCF yield, which is one of my initial checks as whether an idea is worth investigating further.


This is obviously particularly important for institutional investors. The young analyst at a very big hedge fund (both shall remain nameless here) who pitched Sotheby’s as his first idea was disappointed when it was instantly rejected. It was a great PA (Personal account) idea, but this fund needed to be able to deploy hundreds of millions in a single stock idea and Sotheby’s didn’t qualify because

  • It would have taken them months to acquire a position of any size.

  • Once acquired they would not have been able to get out.

This is also relevant for private investors. There are a number of reasons why small caps are often cheap:

  1. Fewer small cap funds (particularly in the UK)

  2. Fewer larger funds wanting to have illiquid holdings (Woodford legacy)

  3. Less visibility, reduced analyst coverage etc

  4. Insufficient volume of interested private investors

Some retail investors I know love such opportunities because they feel they are getting a bargain, but I believe you need to be adequately compensated for a number of risks:

  • The risk that things don’t change – the company does well operationally, but not well enough that the numbers become big enough to attract institutional investors. You may see the stock go up, but not the rerating required to make above average returns.

  • The liquidity problem remains, so you either accept a discount or cannot get out of the position.

  • The stock may be susceptible to sharp falls in an economic downturn as some people look to sell and there is a dearth of buyers.  

  • If the stock fails to deliver, there may be as much downside as for a more highly valued business as investors dump their positions.

Of course, these problems will correct themselves over time, but I think many investors fail to price these risks adequately. This is not to say that you cannot do well with a  couple of individual positions, and I acknowledge that sometimes the market prices small caps at ridiculous levels. Nevertheless, small caps and illiquid stocks are riskier and more volatile than liquid large caps, hence I tend to pay attention to liquidity early on in my analysis.


Analysts tend to focus obsessively on demand on the growth rates of a company’s markets and tend to talk a lot less about supply. This is a strange phenomenon to me. I covered the transport sector on the sell-side for many years and the one thing that was important was growth in capacity – demand growth tended not to vary that much actually, but capacity growth was highly variable and an increase in the growth rate was likely to prove toxic to profitability.

In contrast to most of my fellow sector analysts in the research department, I spent much less time worrying about demand growth and much more time with a laser focus on capacity growth. This was brought to the attention of a much wider audience in Edward Chancellor’s book of the writings of Marathon Asset Management – Capital Account. Sadly, it’s now out of print, but the sequel Capital Returns is still available, and is also recommended.


supply slide.png

As the chart shows, some of this data for the transport sector is readily available (Clarksons is excellent for the shipping industry), while for certain sub-sectors, it required a bit more work.  But I have found that spending time on the supply side, estimating how much new capacity is coming into the industry, or a specific sub-segment, yields much more valuable information than trying to guess the rate of demand growth. And the demand estimates are inevitably something of guess work, whereas you can often forecast supply with a high degree of precision.


Where employee numbers are available (they can be hard to source for some US companies), I always like to use per employee trends. They can offer a useful insight into productivity trends at a company, can give insight into an acquisition, and can be valuable for cross-sector comparisons.

The chart shows the trend in revenue per employee at Proctor and Gamble. Here is a business which has been undergoing some significant change with a rationalisation programme and business disposals, so looking at the trend in revenue per employee gives a sense of whether the portfolio quality is indeed being upgraded. Management will rarely tell you that they have sold a good business, and I have never heard a public company manager admit that what (s)he was left with was not as good as the businesses they sold – subsequent performance sometimes tells a quite different story.


PG revenue per employee.png

The Economist gave a great example recently when it reviewed Reed Hastings’ book about Netflix and its culture. A Netflix employee is worth nearly three times a Google employee and nearly twice an Apple employee, but generates only 30% more revenue. This type of comparison is useful in that it makes you think about the nature of the business more closely, and I think this is really helpful in clarifying company quality – data is always preferable to narrative. You cannot manage without the narrative, but you need data to ensure your conclusions are sensible.


economist per employee measures.png


I think history is quite an important subject and really helpful in understanding the culture and politics of a country. The same is true of businesses and I find it helpful to look back at the long term development of the business. Few companies have an annual report like this one, which lists sales, pre-tax profit, net income and book value for the last 90 years – but I think that you could learn a lot about the company from this single page. Before you read a single number, the fact that they proudly display this record tells the investor a lot about the character of the management and the culture of the business.


GPC Pic.png

History does not get as good an airing as it warrants, in my view. But don’t take my word for this – read this article by Lindsell Train on the enduring qualities of Unilever. This reveals two things to me – the fund manager really understands the businesses in which they invest, which is a great form of advertising. And it suggests that Unilever is in a much stronger position in India than its main developed market peer, by virtue of longevity – this is a valuable insight. The article is really worth reading.


It’s not a coincidence that our latest course is called How to Pick Winning Stocks. These five rules were picked randomly from the course and the charts and slides pictured above are actual examples taken from the course. The course is suitable for investors looking to improve their skills and has had an extremely positive reception. There is a special launch offer and you can learn more by clicking on the button below.