Monday, 8 August 2016

Revisiting my Greenblatt Experiment One Year Later, and a New Experiment with Real Money

Ok, I'm about a month early, but I thought I'd chime in how the overall magic formula dart throwing exercise unfolded.

Last year on September 25, I selected 20 hypothetical darts using the magic formula system straight off of the magicformulainvesting.com site.

The ideas are straight out of Greenblatt's book, The Little Book that Still Beats the Market.

Here are the results through August 8, 2016:




So how did the darts do?

With just over a month left, the portfolio has returned 9.3% before dividends, and 11% including dividends. Annualizing, I get 12%.  Over the same period the S&P 500 as measured by SPY has returned 14.86% in USD's (Sep 25 through current) before dividends, with nominal year/year change in USD/CAD (surprisingly enough).  Including dividends, SPY has returned closer to 17%.

So, just by buying and holding SPY, an investor would have improved on the magic formula darts by about 500 bps. There's still a month to go, but I don't see the above darts improving on SPY comparatively in this short a time period.

So what went wrong?

Firstly, the magic formula darts had too much concentration in retail/apparel. Buckle, Cherokee, and Gamestop, represented 15% of the total dart portfolio. Were I to construct an actual portfolio, I'd probably limit retail/apparel exposure to a much lower weighting (likely zero).

Secondly, the magic formula darts had too much concentration in biotech.  Enanta, Gilead, and Lannett, represented another 15% of the total dart portfolio.  Again, were I to construct an actual portfolio, I'd probably limit biotech exposure to one position instead of three.

Winners included:

Higher One - Acquired by  Blackboard
Argan and Flour Corp (both engineering and construction companies)
CA Inc.
HPE (spinning off enterprise unit and merging it with Computer Sciences)
Interdigital
Liberator Medical  - Acquired by CR Bard
Liquidity Services

Overall, I believe that the construction of a magic formula investing geared portfolio has to consist of more than just arbitrary dart throwing.  I believe that consideration has to be given to overlapping exposure to companies in the same or similar lines of business or sectors, as it's not enough for a company to just be good and cheap.

The New Experiment

This time around, I've started a new experiment with actual money, but I've chopped the overall value of the portfolio in half.  I'm using $10K.  My rules are as follows:


  • The portfolio is to be equally weighted in 20 separate positions using a combination of Greenblatt's magic formula methodology and supplemented by the occasional special situation (blue chip on sale based on my research, merger/arbitrage candidate, spin-off, etc.)
  • Each position cannot initially exceed $500 in total capital at risk
  • Each position must be held for 1 year.  After 1 year, the entire portfolio will be reevaluated in terms of good and cheap. If a position has become cheaper vs. the prior year, and there has been no deterioration in the fundamentals or the outlook/prospects, the position will be maintained, otherwise it will be sold.  Similarly, if a position has done really well and it appears expensive based on fundamentals or the outlook/prospects, the position will be sold, otherwise it will be maintained.
  • The worst positions with deteriorating fundamentals and the best performing positions with deteriorating fundamentals will have to be turned over in order to free up capital for new additions to the portfolio.

The current portfolio is as follows (I'm also adding a new tab to the top of the blog with these holdings).





A brief synopsis of each position below:

Commerce Hub - Spun off recently, I believe the company has a potential reinvestment moat in terms of facilitating drop ship fulfillment for ecommerce

AMC Networks - One of the top good and cheap stocks on the magic formula ranking, and for good reason.  The market is worried about how AMC can generate future cash flows once The Walking Dead fizzles out

Franklin Resources - Asset management, Fidelity, pressure on fees, almost 1/3 of the capitalization is cash, tightly controlled float

H&R Block - I believe it's undervalued and another of the top good and cheap stocks on the magic formula ranking

MSG Networks - Another recent spin off, has performed poorly since the spin off. I believe the company's networks are valuable (MSG sports etc.), and another of the top good and cheap stocks on the magic formula ranking

Neustar - Lost a contract to maintain its position as the main number portability incumbent provider. Has since sued the FCC. Another of the top good and cheap stocks on the magic formula ranking

Oaktree Capital - Howard Marks, need I say more?

Boston Beer - I believe the company could get bought out.

Teradata - Has fallen by the wayside in the data analytics and storage space and is trying to reinvent itself. Another of the top good and cheap stocks on the magic formula ranking.

United Therapeutics - Another of the top good and cheap stocks on the magic formula ranking, and at $4.4B, a possible acquisition target in the biotech space

Depomed - In receipt of recent letters from Starboard Value looking to shake up the board. 

Cogeco - Ontario / Quebec cable provider, I believe they could be bought.

CI Financial - Small position, I'm probably early and would prefer buying closer to $20.

Bristol Myers - Is down close to 20% on news of a late stage trial failure for the company's lung cancer drug, Opdivo. 

I'm up to 14 positions, so I'm looking to add another six positions between now and the end of the year in order to round out the portfolio.








No comments:

Post a Comment