Geoff Gannon June 21, 2012

How to Become a Better Analyst – One Hour at a Time

Geoff here.

Over at Portfolio14, there is a good question about spending time researching new stocks as opposed to just adding to the same old positions:

Charlie Munger always says diversification is diworsification. My dilemma here is whether I should diversify in order to reduce my exposure to one single company. No matter how high my conviction is, there are always “unknown unknowns”. There is also this unhelpful thought urging me to divest: “Earning outstanding returns requires hardwork. If I keep on adding to just the same old position and not spending time to dig deep into other companies, I’m not working hard enough.”

(Portfolio14) 

This is a great question to ponder. On the one hand, I agree that diversification often leads to diworsification. On the other hand, it is important to practice, practice, practice:

A little over a year ago, Geoff Gannon wrote a post where he gave readers the salient financial information of company, but didn’t give the ticker/name of the company. He then had readers guess the stock price. It was an amazing little experiment derived from a quote of Warren Buffett where WEB goes on to say he likes to guess the stock price before looking at the actual price when he analyzes investments.

As always, WEB was well ahead of his time. Much work and study from behavioral finance/economics, like that of Daniel Kahneman, discusses the effects anchoring has on each of us. If we see a stock price before valuing the company, we will unconsciously fix our valuation near the actual price.

Ever since Geoff’s original post I have been fascinated by the experiment. I even went as far as making an Excel program that would randomly generate ticker from the Russell 3000, display the financial information with ticker and price hidden. I could then go about valuing the company and check my work to see how I was doing.

(Distressed Debt Investing)

What’s the solution here?

 

Be Focused in Your Buying – But Omnivorous in Your Research

Should you buy a lot of different stocks? Analyze a lot of different stocks but don’t buy them? Or take the Peter Lynch approach and buy tiny amounts of many stocks just to keep them on your radar – then only load up on the ones you really love?

It’s a tricky question. To gain experience you need to do something similar over and over again. But it can’t be exactly the same. Otherwise, you will become experienced just at that one task. If you become an expert at analyzing Microsoft (MSFT) – you will know how to analyze Microsoft, not how to analyze stocks generally.

 

Use Real World Examples to Master Abstract Concepts

There’s another complicated issue that has to do with patterns.

I think two of the surest signs of real mastery of a subject are:

  • Ability to talk fluently about the subject in everyday language without use of jargon; ability to use short sentences; having a low frequency of big clichéd word chunks common to the field in your writing, speech, etc.
  • Ability to group examples and problems in unusual categories; to see connections between aspects of examples that are not usually the primary basis of categorization within the field

Let’s take movies as an example. We can all group movies according to whether they are:

  • Drama
  • Comedy
  • Horror
  • Romantic comedy
  • Adventure
  • Thriller
  • Etc.

I want to point out something obvious but important here. No definition or rule is really being used here. We are naturally grouping things by a strong “genre” element we see that reminds us of some exemplar – either concrete or abstract.

And there is some abstraction going on here. My clichéd idea of a James Bond movie – the Bond formula in my head – is actually more formulaic than any single Bond movie I can point to. All of the actual movies are missing one or two elements I’d include on a list of what constitutes a James Bond movie.  So I’ve abstracted this idea of a James Bond movie that shares maybe 80% of its DNA with every actual James Bond movie but doesn’t share 100% of its DNA with any single James Bond movie. The abstract concept overlaps with every concrete example – but it never overlaps 100% with any one example.

This is important. Because it means we can’t learn a single definition of a category separately from learning examples in the category. We actually have to get some first or second hand experience with the aspects of various objects belonging to that class to really understand the class at all.

 

How to See Connections Most People Can’t

And as we get better and better at analysis, it’s actually possible to see congruent elements present in objects that belong to different classes. We are no longer limited to thinking in terms of an object belonging only to a single class.

For example, if you’ve seen a lot of romantic comedies and you’ve seen a lot of comedies – you’ve noticed that maybe four-fifths of the way into the movie, there’s going to come a moment (it’s a pretty low point) that’s going to feel very familiar. And maybe you’ll even be able to predict that from this point – where the two leads are not together – someone’s going to have to make a decision that’s going to put the two of them together and resolving something from earlier in the story in the next few minutes.

There’s a French movie playing in (parts) of the United States right now: “The Intouchables”. It’s not a romantic comedy. It is a relationship movie. And it follows that formula I just described.

A critic can criticize it for doing that. An audience can be satisfied by the formula. But an analyst has to see the similarity with other movies – and other stories – that aren’t about the same subject, aren’t from France, aren’t rated R, etc. And hopefully the analyst can understand how the pieces fit together, what their purpose is, why they are where they are, etc. This is necessary for the analyst to do more than just say the movie is French, rated R, a comedy, and about a guy in a wheelchair.

In investing, we face the same analytical problem. There are classes of stocks – value and growth, specific industries, brand companies versus commodity companies, big caps vs. small caps, recognized names versus obscure stocks, etc.

We can certainly spend some time analyzing stocks on the basis of such categories. If you’re analyzing a railroad, you want to spend a little time talking about railroads.

But you don’t want to spend all your time talking about the industry.

There are some key concerns that span stocks from many different industries like:

  • Competitive position
  • Product economics
  • Return on capital

Let’s start with the kind of simple, innocent sounding question that – when asked about a real stock out in the wild – can lead you on an investigation that shapes both the way you see one stock and the way you’ll analyze other similar stocks in the future.

The company is Paradise (PARF).

And the question is: Does this company have industry leading market share?

 

A Real World Example That Illustrates Abstract Concepts

That question alone won’t tell you if return on capital is good. Paradise (PARF) has an 80% share of the U.S. candied fruit business.

It earns low returns on capital. That’s because the product economics of candied fruit suck.

The company has to build up raw material inventory whenever the fruit is ready throughout the year, produce throughout the year, and then rush the product to market so the grocery stores have it on shelves in time for the roughly 6 weeks or so which constitute the only time the end user actually seeks out the product.

To understand Paradise, it helps if you know:

  • Other companies with 80%+ market share
  • Other products where virtually all retail sales are made in under 2 months
  • Other products where it is necessary to acquire raw material at set times
  • Other products where it is necessary to produce a product without selling it for most of the year

It may also help to know a little about societal trends and how they usually play out in business. Candied fruit goes into fruit cake. Most people analyzing this company start with no knowledge of fruitcake other than thinking it a rather antiquated product. So – before analyzing Paradise, it helps to have some prior experience learning about a product you’re not that interested in, might be biased against, etc. Having that experience gives you an advantage over analysts who let their pre-conceived notions of a product dominate their analysis of a stock.

Analyzing other stocks can help with all of this. For example, when I first looked at Paradise – which was a net-net – I had plenty of experience with other net-nets. I had analyzed dozens of net-nets in depth. And I’d glanced at hundreds of net-nets over the years.

I also was familiar with a handful of companies that had 50% to 100% market share. Especially in tiny industries. This is a particular interest of mine. You’ll remember that Hidden Champions is one of my favorite books.

 

Collect an Archive of Informative Examples – Then Abstract From Them

How do you start to build up little mental file cabinets full of ideas about stuff like companies with 80% market share?

For the most part, you can only do it by analyzing stocks. Analyzing different kinds of stocks. But most importantly – analyzing specific stocks. Not abstract ideas about stocks.

Should you analyze stocks you aren’t going to buy?

Yes.

It’s great practice. And sometimes it’s hard to know what you will and won’t buy before you start analyzing the company.

Now, some folks will argue that it makes no sense to analyze companies you know are not especially cheap – because there are conspicuously cheap stocks out there. If one stock has an EV/EBITDA of 4 and another has an EV/EBITDA of 10 – why waste time on the stock with an EV/EBITDA of 10? If you look hard enough, can’t you find something with an EV/EBITDA of 4 that you really like? And isn’t there a lot of academic research, backtests run by bloggers like me, etc. that shows it’s better to hunt for a decent company trading at an EV/EBITDA of 4 than trying to outguess the competition when it comes to well known companies with an EV/EBITDA of 10?

In other words, shouldn’t you just focus all your time on obviously cheap and obscure stocks?

That’s where the best bargains are. And you’ll learn a lot analyzing those stocks.

But, if you put in enough time – you’ll learn from analyzing almost any stocks.

Some people are worried about having too little time. I’m going to sound harsh here – and maybe this really doesn’t apply in your case. Maybe your life really is too busy to add another minute of stock analysis to your day.

But I doubt it.

 

How to Sneak in an Extra 250 Hours of Investing Practice a Year

Most people spend about an hour – or more – of their time each weekday night on something non-critical. It may be watching some mediocre TV show. I’m not judging. I’m just saying you can probably find a way to go to sleep about one hour earlier than you do right now.

That means you can – with the same amount of sleep – get up an hour earlier. If you go to sleep one hour earlier tonight than normal, and you normally wake up at 6 a.m. – tomorrow you can be just as rested waking up at 5 a.m.

When you first wake up is a very productive time. An hour of analyzing some stock from 5 a.m. to 6 a.m. each morning is likely to be as effective as an hour and a half (or more) spent analyzing some stock in the evening.

Try it. You’ll be amazed how productive you are when you focus your mind on a single task very early in the morning.

Anyway, you can add 5 hours of stock research to your week this way. And that’s just on weekdays. I think this is something almost anyone could manage to add to their life.

Five hours is enough time for most people to research most stocks. That means you can research 52 stocks a year. We’ll give you a vacation each year and call it 50 stocks a year.

That’s a lot of practice. You build up a big library of examples by analyzing an extra 50 stocks a year.

I recommend keeping two lists:

  1. A list of stocks you find interesting but know almost nothing about – this is your research list
  2. A list of the stocks you researched and now know best – this is your watch list

The great things about knowing a company really well is that the knowledge stays fresh longer than you’d think. Good businesses change pretty slowly. Much slower than stock prices.

You’ll often find a company you researched – but thought too expensive – has seen its stock price drop 50% or more since you spent a week analyzing it.

In this way, research you thought was purely for practice suddenly becomes very practical.

Very actionable.

Share: