Statistical Investing

I recently read “Painting by Numbers – An Ode to Quan” by James Montier and Dresner Klienwort via a link from Richard Beddard to Greenbackd.  This paper, and the papers it refers to, have helped strengthen some of my existing opinions about stock picking and investing in general. 

My opinions are also those of Ben Graham towards the end of his life, that “I am no longer an advocate of elaborate techniques of security analysis in order to find superior value opportunities. This was a rewarding activity, say, 40 years ago, when our textbook “Graham and Dodd” was first published; but the situation has changed a great deal since then. In the old days any well-trained security analyst could do a good professional job of selecting undervalued issues through detailed studies; but in the light of the enormous amount of research now being carried on, I doubt whether in most cases such extensive efforts will generate sufficiently superior selections to justify their cost” (

Painting by Numbers backs the argument that human judgement typically offers no benefit to decision making where that decision can be made instead using simple quantatative methods.  Many examples are given of decisions made by doctors, judges, parole boards, purchase managers, wine experts, and in almost all cases the application of simple statistics produced better (or at least as good) outcomes.  A breif example which comes to mind is of a parole board deciding whether to let a prisoner out early.  Is the best decision (on average) made by the experts on the board, or by calculating the odds that the prisoner will re-offend based on data like sex, age, background, criminal record, offences commited inside prison, etc.  With hindsight the quantatative approach produces the better answer most of the time.  Although I’m sure there are many studies to counter this argument.

This tied in somewhat loosely with a TV program on the other night called The Secret Life of Chaos.  In the ‘secret life’ one of the scientists commented on how all the glorious patterns and structure in the universe can emerge from a relatively simple set of rules; or at least simple in relation to the patterns and structures they can generate.  He said, more or less, that if there was a job for God it it was to set the initial conditions for the system and then let it do its thing.  No further tweaking or interference is either desirable or required.

From a statistical investing point of view, this means it is probably better to play the odds.  By finding out the quantatative features of those companies that have, in the past, done best over the following 1-5 years (or whatever your investing timeline is), it may be possible to have an opinion on which companies are more likely to do better in the next 1-5 years.

Research gathered by Tweedy Browne in What Has Worked In Investing among many others, shows that this is possible (or at least, probably possible).  The lesson for me is: build a mechanical investing system; set it in motion; and as much as is humanly possible, leave it alone.

Author: John Kingham

I cover both the theory and practice of investing in high-quality UK dividend stocks for long-term income and growth.

6 thoughts on “Statistical Investing”

  1. Greenblatts simple formula is somewhat similar I would guess in that it works if followed over a period of time. Ron

  2. Hi RonI think the magic formula is a perfect example. It stands in stark contrast to the idea that if a simple method of outperformance is known then buyers will bid up the stocks until the outperformers no longer exist. You can't get much better known than the magic formula and yet it still outperforms over the long term.Let's wait 20 years and see if that's still true though. I think it will be because whenever you are buying bargains of any sort you are basically being paid to hold something smelly that others don't want to touch. For amateurs this is dinner party risk, where people will laugh at your crummy stocks, for professionals it is career risk.

  3. Great post.I also read the Graham article where Graham mentioned his equivalent of a magic formula.I really like the Greenblatt magic formula for its simplicity and track record.James Montier mentioned in the past the greatest challenge is to leave it alone.Dinner party risk for value investors are e given. I hardly ever talk about my holdings any more. Only if asked.Tim

  4. Thanks Tim. Simplicity is something I always enjoy. Selection criteria like net net, or just low P/B with one or two liquidity and debt measures are what get me excited. Your mention of check-lists here: an eyebrow as I'm a check-list user and am just getting into systemising other areas of my life. I've read Getting Things Done (love its non-fluffy mechanical nature) and have 80/20 and Work the System on order.

  5. I like the idea of Magic Formula Investing, but I am extremely nervous about its validity. By excluding intangibles, for example, you tend to end up with unrealistically high values for ROCE. So it's a case of GIGO (Garbage In, Garbage Out).

  6. Greenblatt's magic formula delivers good returns but other methods such as piotroski price to book and ERP 5 combined with momentum deliver much better returns. More info greenblatt and other techniques on Greenblatt Magic Formula

Comments are closed.

%d bloggers like this: