There have been a couple of articles over at seekingalpha over the last couple of weeks regarding using gasoline or crude oil ETFs to "lock in" today's relatively low gas prices.  The first author put a simple spin on the idea.  The author had no regard for tracking error and just made a simple "buy and hold case".  Not too compelling.  The next author (article here), presents an interesting twist on the original idea.  It involves a strategy that I use and demonstrate here, that is, selling puts on USO.  And then, upon assignment, sell calls.

I commented:

The problem here is tracking error. These leveraged and futures based ETFs aim to replicate the DAILY movement of the underlying. Once you start holding for longer periods, the tracking error inevitably gets you. So your "lock in" is not actually a lock. 

However, the strategy that the author presents MAY offer some relief against the error in the form of time premium. Of course, the author is at risk of owning the underlying. Due to oil's volatility, writing calls is probably not a viable long term strategy either. Imagine this scenario. You're assigned today, you write a call. No problem. Over the next month oil drops another 30%. A call above your assignment price is going to be worthless, so your only two options are to write a call that may lock in a loss or hold the underlying and pray. Naturally, you get some kick back at the pump, but I'm trying point out some nuances associated with this approach. 

But let me say, that I do like this idea as a whole.  The reason being that owning the underlying has some benefit.  It's an asset that isn't going to 0 and something that nearly everyone uses and understands.  I did want to point out that it's not quite as simple as the article states.  And as usual, it's highly recommended that one trades this on paper or in some sort of simulated account before going live with real money.  It's very important to understand how this strategy will react in the "real" market.

There are several overriding themes for trading systems.  Some are trend following trading systems, some are momentum based and some are band violation. Band violation trading systems seem to have a special place in my heart.  They are really the only systems that I've used in real time.  For whatever reason, my ideas for trend and momentum systems seem to fall flat in testing.  This doesn't mean that they're bad, it just means that I haven't enough brain power or good ideas to leverage them.  Or possibly I'm looking in the wrong places or tweaking the wrong parameters.

But back on point, an idea struck me recently as it relates to commodities.  An author to remain nameless purported that the return on commodities has been essentially 0% over long time frames.  He further argued that this is a good reason to have them in a balanced asset allocation strategy ... the zig when your portfolio is zagging so to speak. I'm not sure if I'd want something that returns 0% over time, but that's another discussion.  But it did get me thinking about a mean reversion trading system based on commodities.  Admittedly, this is not the first time this has crossed my mind, but the real problem is the intracicies of futures. Backwardation, contango, leverage, margin (calls!), rollovers, and deviation from spot are really a lot to consider when trading over longer timeframes (i.e. > a couple of months).  Some of the ETFs have attempted to remedy this, but there is still tracking error that never seems to work in the favor of the holder.

Two ETFs hold physical metal.  GLD and SLV.  I have not been successful in finding more ETFs that hold the physical commodity and can match the spot price, but I'm still digging (no pun intended).  I understand that commodities can trend for a long time in either direction, but they tend to correct a bit.  Have a look at oil.  It would've been a 4 year ride, but you'd be back at a 0% return after having had a substantial gain.

My idea is to develop a trading system using RightEdge that buys and sells short term extremes in commodity prices.  I'm going to spend some time this week polishing up some problems I have with the system.  Then I'm going to start generating signals and paper trading to see how it goes.  I can envision several iterations unless it blows up horribly or I realize that I cannot stomach the draw down.  Then I call it a failure and move on.  I'll post the signals here.  I may even setup another marketocracy fund to track the progress.  This will be highly speculative and will not conform to their rules, but should serve our purpose of testing.

And as always, I'm not using real money, so you shouldn't either.  In fact, even if I were, you shouldn't follow my advice.  My test systems usually fail a number of times and very few ever see real money traded.  Still fewer make money.

I've been talking with an evaluator of RightEdge over the last few weeks.  He recently asked some great questions that basically cuts to the chase.  My answers, unfortunately, are probably not so direct (with good reason).  The first question is:

Lets say you create something that has returns every year for the last ten years of say at least 15% on a portfolio of say 35 stocks ( including up to Nov ,2008 ) would that be good/great?

Well, that’s not telling me the whole picture.  15% is the return which seems pretty good but what’s the risk?

When developing trading system, there are a number of things to consider.  Two big ones are risk tolerance (which is a personal thing) and confidence in data. Obviously, if something doesn’t backtest well, it’s out.  Believe it or not, I use backtesting typically to disprove what I think (i.e. I have a great idea in the middle of the night, I use RightEdge to tell me that I should’ve stayed asleep).  Also, anyone that pushes a strategy, I can usually use RightEdge to pick apart the components (and usually find out it’s junk).

It’s really a matter of developing your toolbox.  Since I trade options, I use RightEdge to give me probabilities.  I also cull my list based on volatility.  I don’t think this is for everyone and certainly not the way everyone would do it.  There is no blanket system that returns 15% every year regardless of market conditions. That’s like saying “give me a stock that will always gain”.  You need to build a collection of systems and put in place risk management.  RightEdge is a piece of the solution, not the entire solution.

The bottom line is that I run several systems.  I do have one that runs against the Dow 30 and it is currently “paused” until demand for equities turns up.  However, my short put strategy is still running.  I’m also working on a system for highly liquid ETFs and after that, I want something for commodities.  To give you an idea, I expect it to take me months to get those how I like them.  It has taken me years to get my other systems together and it’s really a culmination of things learned and items I’ve put in my toolbox over the years.

This is no small investment of time or money if you want to do it right. 

The second question is:

At what point do you get excited about a system? 

This one I have a quick answer for.  I get excited about a system when I see it making gains in my account.  Until then, I remain highly skeptical.

Hope you had a great Thanksgiving.  What's a better holiday than sitting around and eating?  Oh ya, 4th of July where we get to eat a lot and blow things up.

So anyway, I caught myself on Friday feeling "good" about the market.  I know that Friday after t-day is always thin, but after the 1,000 point gain, it felt like we might almost be flirting with being solid.  I waded in with a few more buys since I was still averaging into big losses, it lowered my cost basis some more.  It wasn't exactly with the plan, but I still have more dry powder for Dow < 7,500.  Why did I buy on Friday?  I was feeling good about the current prices.  Fortunately, I knew that feelings should have nothing to do with the market, so the buying was sparse, but it goes to show that it has no place in the market at all.  This morning is proof of that.  We dropped almost 300 points at the open.  Who knows though, we could've just as easily have gone up 300 and who's to say with this whacky market that we still won't.

Another word about leveraged ETFs. 

I'm finding a number of posts regarding the 3x ETFs.  I still don't think people understand how the tracking works.  It works on a day to day basis.  The leverage is achieved through various means, all of which have spreads.  So for the fund to leverage, they also have to swallow a spread on something ... swaps, options, futures, whatever.  Please understand that this is not a free lunch.  As I've demonstrated before, many leveraged funds will actually the non-leveraged fund over time and still give you all the volatility!  The interest in these funds is apparently increasing among day traders and that's probably the best person to appreciate this tool.  Any long term holder will likely end up disappointed.