Reader's Comment: Oil Stocks vs. Oil Trading Strategy

This is a drilldown (pun intended) on the strategy I shared last week that traded the oil sector using a ratio of oil stocks versus oil.

As a refresher, the graph below shows the (frictionless) results of the strategy in green versus buy and hold in blue, and for comparison, the inverse of the strategy in red, from 1986.

click to enlarge

[logarithmically-scaled]

The strategy went long the oil sector at today’s (December 29) close when the 9-day exponential moving average of the oil sector vs. oil ratio was falling, indicating that oil stocks were becoming cheap (in an intermediate timeframe) relative to oil itself.

Reader Russ Abbott of Los Angeles, California responded with a very smart comment. In a nutshell, Russ questioned my oversimplification that oil stocks were becoming cheap relative to oil.

Was it that both were rising but oil stocks had risen less (i.e. oil leading oil stocks up), or that both were falling, but oil stocks had fallen more (i.e. oil stocks leading oil down), or were the two diverging (i.e. oil stocks heading down while oil itself heading up)?

My Response

In the graph below I’ve rerun the original strategy in four different flavors: when the 9-day exponential moving average of oil stocks was rising (blue) or falling (dotted blue), or of oil prices was rising (red) or falling (dotted red).

click to enlarge

[logarithmically-scaled]

At first glance it appears that the strategy was most effective when either oil stocks or oil itself was rising (solid lines relative to dotted lines), but that ignores the fact that one of those flavors might spend a lot less time in the market. So, the next table looks at daily returns (only when invested):

click to enlarge

What can we conclude from the graph and table above?

The “Number of Days” column shows that most of the time the strategy is buying the oil sector when oil prices have been rising but that rise hasn’t yet been matched by the oil sector (i.e. oil leading oil stocks up).

In terms of pure daily return, the strategy has been most effective when oil prices have been falling (but oil stocks have been falling even more), but in terms of volatility-adjusted return, the strategy has been more effective when the oil sector has been rising (but oil prices have been rising even more).

But the most important takeaway is that all four flavors of the strategy have been effective.

My oversimplification was (unintentionally) spot on – regardless of whether the oil sector or oil itself has been rising or falling, the fact that the oil sector has become cheap relative to oil (in the intermediate timeframe we’re looking at here), has been bullish for oil stocks.

Thanks again to Russ for the smart comment. There is nothing better as a blogger than when readers get directly involved in pushing this collective discussion forward.

[Geek Note: There are two generally accepted ways to calculate an EMA that produce slightly different results. Here I’ve used the ((1/Period)*2) method. If your charting program uses the (2 / (Period + 1)) method, simply reduce my period by one. For example, if I’ve used a 9 period EMA, the alternate EMA would be an 8 period EMA.]

Comments

[Greg Pinelli:]

This is..to say the least..a highly overworked and largely untradeable idea.
Even a fairly sophisticated investor is going to have to wrestle very hard to squeeze simply the essence of what is going on and commit to a series of short term trades. I have a simpler idea....
Buy the commodity for the shorter.intermediate term..USO will do just fine..sell when it's 1 std dev below the 200 dMA..buy on the opposite..Not sophisticated but profitable and simple..and it keeps trading costs low...
Buy a CEF or ETF consisting of oil equities/services that pays a nice distribution and hold for the payouts..you could use the same 200dMA strategy above..it saves a lot of whip sawing..I wouldn't argue with a higher stdDev if one is more conservative....Ratios are great for spotting extremes in trends, but as daily investible indicators a very dangerous and ambiguous tool.

Jan 05 10:14 AM

[futurestrader:]

Can someone help explain the USO to me. I traded a little in the USO, but mostly in UNG in 2007, but received a K-1 statement for taxes....On that paper they had a significant loss for me in their, which helped for taxes, but I never realized any losses. I actually had significant short term capital gains, which I had to show on my regualr 1099. I switched to trading the actual contracts lat year, because I didnt want to see the inverse and actually show gains on a K-1 and have to pay taxes on something that I am not very certain how that number is derived...Do these pay distributions, and if so, is that the number that goes on the K-1? Does someone understand this, because I would like to trade USO, UNG in my brokerage account

Jan 05 11:21 AM

[1977°C:]

Any technical strategy works good when there are money on the table to gamble, when current crash cut 50% of money invested worldwide I doubt any strategy has sense.
I only believe in the daytrade as this is the only one strategy that let's trader a minimal exposure to price action moves long term.

Jan 05 11:39 AM

[Energy-Trader:]

The above analysis is far too simplistic to be effective.

For example, just look at how the major integrated oil producers like XOM, CVX, and COP bottomed in October with the S&P and then proceeded to go on a huge rally (+30%) while crude oil futures on the NYMEX were still plummeting in the 4th quarter, from $72 down to $35.

That lack of correlation will kill a "system" such as the one highlighted above.

Jan 05 07:52 PM

[Michael S:]

RE to Energy Trader: two comments: (a) readers of my actual blog not the posts that are pulled over by seeking alpha will know that I advocate a more holistic approach to the market that considers many different strategies/indicators (i.e. I don't trade any single strategy in isolation), and (b) while I appreciate your anecdotal response, the numbers are the numbers and those numbers are hard to argue with over a long trading horizon. michael

Jan 07 01:02 PM

[Michael S:]

RE to 1977C: I would argue that a solid active trading program shouldn't have suffered that sort of decline in 2008 and I could argue (based on our own audited performance) should have ended the year at the very least flat.

Because I hate snake oil salesmen and touts I should say that all of our programs are independently-audited by at least one third party, and that we are not unique - I think that there are a number of good programs that are doing very well in these markets.

michael

Jan 07 01:06 PM

[Michael S:]

RE to Greg: I'm not sure how you come to that conclusion. In my case, I put a lot of research into designing strategies, but in terms of actual trading, I'm done in about 30 minutes a day (and that's trading four separate programs). That's a benefit of mechanical trading. Where do you see this massive time drag?

Jan 07 01:09 PM