Statistical analysis will improve investment outcomes

Particularly in commodity based companies like energy

The late Jim Simons built the most successful hedge fund in history by using mathematics to guide his trading decisions. Medallion Fund, which he ran for 30 years, had an average annual rate of return of 66% for that 30 year period. He understood that uncertainty creates both risk and opportunity, applied statistical analysis, and made fortunes.

Statistical analysis is particularly valuable for commodity based companies. Consider oil stocks.

A look at the historical distribution of oil prices adjusted for inflation creates this distribution (prepared by Aswath Damadoran in an article he published¹). In real terms, he found oil prices ranged from $8 a barrel to $120 a barrel in 2009 dollars.

That was fifteen years ago, and in 2024 dollars that range is more likely $12 dollars to $170 dollars (inflation averaged about 2.5 percent from 2009 to 2015). The mean of the adjusted distribution is approximately $65 and today’s WTI price is ~$80 a barrel. That suggests (other things being equal) oil prices are more likely to fall than rise, although it is pure speculation to make short term predictions given the macroeconomic forces that affect oil supply and demand.

Valuation of energy companies based on the assumption of a $65 a barrel oil price is in my opinion a more reliable valuation than one that assumes prices about two standard deviations above the mean (the standard deviation of this distribution is approximately $7)

So what does that mean for popular oil names? For this analysis, I am assumiing light oil commands a $10 per barrel premium to heavy oil (i.e. WTI at US$75 a barrel), that condensate is priced as if it were light oil, and that natural gas is priced at $2.50 per gigajoule. I will use a CDN$ exchange rate of $0.73 U.S.

I won’t belabor the point with an analysis of the whole flock of companies I follow. Instead, I will focus on just two - MEG Energy, a heavy oil producer and Saturn Oil, a light oil producer.

With WCS at US$65 (assumed) or its equivalent CDN$87 a barrel, and applying an arguably generous 6 x EBITDA valuation metric to reflect the long lived reserves and low decline rate of MEG’s reserves, I come up with a per share price of about CDN$25.66 for MEG.

For Saturn which is primarly light oil, the US$75 for WTI is equal to CDN$103 per barrel (which when blended with Saturn’s natural gas and NGL output approximates CDN$80 a barrel) and applying a 4 x EBITDA valuation metric to reflect Saturn’s high debt leverage, I get a valuation of CDN$8.66

I believe these are more reliable estimates of the per share value of these companies than the ones I see from most sell-side analysts. The market price of MEG is higher than my estimate of value by 20-25% and the market price of SOIL is 70% lower than the value I see.

There is an old adage “markets can stay irrational longer than you can stay solvent”, so don’t take out a mortgage to rush to buy Saturn shares or panic and sell your MEG holdings. Just think it through, discuss it with you own advisors, and don’t fall victim to fads, short term sentiment, or some “belief” in future commodity prices. Common sense and patience pay off.

1. Ups and Downs. Valuing Cyclical and Commodity Companies, Damodaran, 2009