Print
Email
…regarding the very large risk B&N took, I’m reading Adapt now. It seems to be saying that as no one can see the future, success comes from trial and error where the basic rules have to be: variation is present and making sure the failure would not be fatal. If you’d agree with this idea, it can be applied to investing, can’t it…
Thanks,
Roy
I do agree with that idea.
Success involves much more trial and error than people think.
Specific plans made at high levels and meant to cover the distant future are often futile. This isn’t news. Business has been moving toward delegated authority for many years. A hundred years ago, people feared that workers would be reduced to fleshen machines. Unthinking cogs. Today, the big fear is exactly the opposite. Too much thinking. Too much stress. An inability to execute your work individually without worrying about what the guy next to you is doing. Your work is never fully your own. Your mind is never fully focused. You’re always hooked into the system.
We have two problems here. One is that nobody has enough knowledge in one place. It doesn’t matter if we’re talking about the future. Nobody has enough knowledge about the past, present or future to encapsulate everything the group knows.
Then, dealing with the future specifically, we can add the old saw about how “no plan survives contact with the enemy.” Well no plan can survive contact with customers, competitors or employees. From time to time, they will all surprise you with their behavior. They will all make choices you didn’t even know were options.
Great. That's companies.
But how does trial and error apply to investing?
Two ways:
- At the company level
- At the investor level
But I’ll do my best to talk about the idea of applying trial and error to your own investing.
Now, I’m a little doubtful on the usefulness of trial and error at the investor level. I mean, I’m all for experience. For practice. But the logic of variation at the investor level seems kind of wobbly.
The folks who’ve studied variation have usually done it while studying groups. Not individuals. Now, that doesn’t mean that trial and error itself isn’t an effective approach at the individual level. It just means that you might not want to assume the special virtues of variation will translate from the group to the individual.
Still, you can teach a dog by trial and error. In fact, it’s a very effective method of teaching a dog since dogs are kind of weak when it comes to aping. It’s tough to demonstrate to a dog. But it’s easy to reward or ignore behavior. They associate their action with your response. And they learn pretty fast which route to take. Then they keep taking it. That’s trial and error (from the dog’s perspective, not yours — you're playing the automaton here). And if a dog can use trial and error when interacting with a human — maybe a human can use trial and error when interacting with other humans.
In fact, I’m sure they can.
I’ve mentioned before that I was diagnosed with Asperger’s Syndrome when I was 17. Before that, I didn’t know what I had. Though obviously I knew I “had” something in the sense that I knew that while all people are different I was differenter still.
Basically, I noticed there were a bunch of things I could do that other people couldn’t do and a bunch of things other people could do that I couldn’t.
I drew like a 3-year-old. Still do. But I never needed a bookmark when reading a book. I saw words spelled out in my head as I heard them - words in conversation, music lyrics, TV, anything. Other folks didn’t seem to. But they did seem to be really good at remembering what people looked like. Which I'm not. And at ignoring the weird way some things sound, smell and feel. Of all the evils in the world, I still rank the feel of a cotton ball as the globe's greatest terror.
Now, here’s the thing, I wasn’t diagnosed with Asperger’s Syndrome until I was 17. By that time, I had adapted.
A lot.
So trial and error works. I didn’t need to know I had Asperger’s to know I wasn’t a good fit.
I needed to adapt. And I did.
But this is where I see a problem with applying the trial and error approach to your own individual behavior. Sure it selects for the best fit. But it does so by changing your routines in reaction to short-term consequences. And it does so without regard to the long-term trade-off in modifying your routine.
Let me explain with my Asperger’s example.
I didn’t know I had Asperger’s. So, I couldn’t take a systematic approach to “fitting.” What I could and did do — mostly as a child — was change bits of my behavior whenever I had a really awful experience (almost always a bad interaction with another person) that seemed to be caused by some unusual behavior of mine. Unusual here meaning usual for me — but not usual for normal folks.
So, for example, I do use bookmarks now. Actually, I read books on a Kindle. But before that I used bookmarks. Why?
Did I grow out of Asperger’s?
Did I suddenly need bookmarks?
No. I had a couple experiences where kids at school thought it was freaky that I wasn’t looking for a chapter or page number but instead just flipping through a book to find the spot where something was. I did this once in a group and they looked at me like I had three heads. So scratch that behavior.
I never looked at the teacher in school. I used to just put my head down and stare at my desk. Running stories through my head. I can still remember when a girl in my class came up to me and said that she and some other classmates had been arguing over why I always did that, and she decided to just come up and ask me herself: “Why do you just sit there staring at your desk?” Horrifying experience. Still remember it. Scratch that one too.
Some of my ad hoc modifications were more extreme. When people talked to me, they tended to think I was disagreeing with them because I didn’t look at them when they spoke and I didn’t react with my face. I didn‘t know that was the reason. But I did notice that the one time they softened was whenever I smiled. Now I wasn’t smiling to get that reaction. It’s just something that happens. A happy accident. Trial and error. You occasionally smile and if someone’s talking they tend to think you’re smiling for them. So I went with that. I started smiling and nodding like crazy. The more scared I was of being misunderstood, the more of a grinning bobblehead I became.
This one works like magic. Sitcoms don’t lie. Smiling and nodding really does solve all problems.
But…
I was only able to keep smiling and nodding by constantly agreeing with whatever the other person said. You think that’s a joke. But it’s not. Try smiling and nodding while disagreeing with someone. Don’t just mime a low key version. Actually give a full on smile and nod while disagreeing with someone. It’s harder than you think.
To give you an idea of how absurd this got, when I was in sixth grade someone asked me if I was in eighth grade. I said yes. Didn’t hesitate. Just said yes. Cheery as can be. Stuff like that happened all the time. People started thinking I was a much nicer fellow. They didn’t realize that being nice with my face required me to shut off my brain. When you’re so busy smiling and nodding that you’re agreeing to be a different age just to move the conversation along — you’ve got a problem.
And that’s why I think the trial and error approach may not work well for individuals.
In the dog example I gave, where you are holding a treat out for the dog obviously expecting something but letting him figure out what to do — the one thing you never do is punish the dog. You just ignore, ignore, ignore, reward. Ignore, ignore, ignore, reward. And then you build on that one simple step he got right and teach him how to go from “sit” to “down” to “up” or whatever else it is people teach their dogs.
But what if you punish the dog?
Here’s where I think the trial and error approach could lead to some very weird results. I got through a childhood with Asperger’s through trial and error. It certainly made me adapt. I became a better fit. But fit for what?
For being a normal kid. Not standing out.
So, how useful are the skills I learned through trial and error?
I’ll give you a hint. Remember how I started out — sweet and innocent — sitting quietly, staring at my desk, all up in my own head.
Now, picture me writing this article. What do you do when you write an article?
You spend time up in your own head.
I knew how to do that in kindergarten.
So trial and error made me a better fit in a grade school environment. It probably made me a worse fit in a writing environment.
Trial and error helps you adapt to your present environment.
In investing, that’s kind of dangerous.
The market environment changes. It rewards one behavior this year and punishes it the next. Trial and error might lead to reward, reward, reward in a bull market and punishment, punishment, punishment in a bear market. That can mess you up psychologically. It can modify behavior in short-term positive ways that are simultaneously long-term negative.
While I’m a huge believer in trial and error I’m not a big believer in trial and error when it comes to actions an individual investor should take.
Now, if you want to learn what works in investing, seeding a group of 1,000 college kids and letting them go out and do their thing in the market for a couple decades while chronicling their journey — that sounds downright informative.
But I worry about trial and error in teaching an individual human to invest. Because, to me, it seems too similar to the way I used trial and error to deal with (what I didn’t know then was) Asperger’s. Or the way a dog uses trial and error to pull out his whole bag of tricks to please you. With a dog, you — not the market — can be the patient teacher who ignores some behavior and rewards other behavior. Always thinking about the long-term. The market doesn’t do that. It’ll punish you. Hard.
And the problem with that punishment is the complexity of what you were doing when you got punished and the length of time between your error and your punishment. It’ll send you reeling.
I think the length of time between error and punishment in the stock market is a particularly dangerous problem. Reckless investors can get rewarded for a long time before they get their comeuppance. And that introduces the problem of forming meaning. Humans don’t sit idly by and take punishment. They ask: “Why me? What did I do?” They look around. And sometimes they settle on really recent, really dumb justifications for why they are being punished. Even if they made the error when they bought the stock a year ago, they’ll be convinced it was failing to sell on that last day before the bottom fell out. They’ll even swear they remember “having a bad feeling.”
Another problem is the randomness of it all. I constantly see investors learning the wrong lesson from their mistakes. Maybe I learned the wrong lesson from my Barnes & Noble debacle. Ben Graham and Warren Buffett had to learn to ignore market punishment. I mean, Graham’s principles were firmly established before 1929. If the crash taught him anything, it was not to be leveraged. A good lesson. But it also taught lesser investors to give up. It didn’t matter what you owned. Is that a good lesson?
A lot of investors take a moment to reassess their approach after a huge market downturn. Some experts even suggest this. I’d encourage the opposite. It’s better to reassess your approach in the midst of a steadily rising bull market — when nobody else is analyzing themselves — than to think that the destruction in your portfolio caused by a catastrophe which is engulfing the whole world is somehow really the result of your own personal sin.
The market can train a nation to buy and hold houses forever and let that work and work and work until it doesn’t. The market can train investors to buy an overpriced stock that’s rising higher and higher on tremendous momentum — but that’s ultimately going to crash.
That’s why I worry about applying a trial and error approach to your own investments. We all know character is behavior. Sometimes we forget behavior is character.
Change your behavior and you change yourself.
If trial and error leads you away from certain principles and towards whatever’s working in the current market trend — you’re going to be changed by that experience.
Trial and error changes groups over time. Through attrition.
Now, the obvious argument is: Why can’t you do the same thing at the individual level?
Why can’t you eliminate certain behaviors through attrition?
Just let the bad habits get killed off.
I’m not saying you won’t kill off your bad habits. You will. I killed off some of mine as a kid. I learned new ones. And we all know you can kill off a dog’s bad habits. Why not an investor’s?
I don’t think the evidence supports this approach. We know of some real strong investors who had clear principles at a very young age. With Buffett and Graham we’re talking about late teens (Buffett) and early twenties (Graham). You can say they’re flukes. I’m not sure. Finding on solid principle early on and then practicing, practicing, practicing seems to work.
I think an individual who navigates within a herd by trial and error is unlikely to outperform that herd.
At the same time, I do agree that you can’t plan everything out. No one is arguing for a plan. Principles maybe. But never a plan.
Warren Buffett once told Charlie Rose that there’s no master plan at Berkshire. Charlie said then not having a master plan is genius.
I don’t think I’ve figured this one out. I have a lot of respect for trial and error.
But I strongly believe there’s really no way to invest without a theory. There’s no way to look at data without hanging it on something.
And so I guess we end up with the wishy washy idea that you keep your principles and change everything else.
All this talk of adapting makes me wonder...
Did Ben Graham adapt?
Maybe.
In little ways, yes. Graham had a bunch of different techniques he used in his early years — like long/short — that he abandoned in later years. Ultimately, Graham decided that only net-nets, control investments, arbitrage, and related hedges (short the stock, long the convertible) were worth his time.
And, honestly, if you look at Graham’s record he may have been too rigid. He was very risk averse. He continued to diversify more widely than necessary. Which is not a big deal. But he also continued to own bonds and preferred stocks. That’s a big deal. They held back his results.
Ultimately, Graham’s record rests on control investments, net-nets, and arbitrage. In each of these areas, he added value for his investors. In other areas, he reduced volatility. But he also reduced long-term returns. If you like that trade-off you can justify what he did. I won’t.
Maybe Graham didn’t take trial and error seriously enough. Maybe he stuck too long to specific applications of his principles. He could’ve kept the same principles and operated purely in common stocks. His results would’ve been better. But also more volatile.
Graham’s students adapted where he couldn’t. They built on his legacy by seeing his blind spots.
Taking trial and error more seriously might have helped Graham’s later career. His results clearly showed that value investing was more effective when applied to common stocks than to bonds or preferred stocks.
We’re really going far afield now. The original question was about applying trial and error to investing.
Let’s bring it back.
What does all this mean for you?
My advice would be to focus more on investing with style than on investing by trial and error.
I don’t think I did a good job explaining the perils of a trial and error approach using my Asperger’s analogy.
The point I was trying to make is this:
The market is a flock of sheep. You are in that flock. Trial and error may encourage you to bah and blend in when you really ought to be standing out.
Don’t let the market train you to be short-term smart and long-term stupid.
That’s why you need principles.
For the long-term.
Follow Geoff at Gannon On Investing
About the author:
Geoff Gannon www.gurufocus.com
Subscribe RSS
