When You Can’t Quantify Your Risk

• Posted by kid dynamite
• on October 2nd, 2012

This is kinda a weird post topic.   I am trying to enunciate some thoughts I have about the “uncertainty of uncertainty” – by which I mean, “when you’re not confident that you even know what your risk is;”  you’re uncertain about what your uncertainty even is.  Before I get to that topic, though, I feel like I need some tangents, because it’s rare that we can ever definitively estimate what our risk actually is.   We may often be able to estimate the downside and upside scenarios with high degrees of accuracy, but estimating the probabilities of each outcome is an art as much of a science, and in the investing world we can’t ever really know for sure that our bull-case has a 65% chance (or whatever number you want to enter there) of being realized.

What makes a good investor or trader, however, is honesty and accuracy in these estimates.   This is kinda a tough concept to explain, and in general, we kinda expect “the market” to take care of it in the long run:  those who consistently mis-estimate the risks/uncertainties/probabilities that they are analyzing will lose money.

Now, having a negative result on a trade does NOT mean that your estimate was wrong – and this is where it gets even more confusing.  I think I can give an easy example via poker:  If I am playing no limit hold’em, and I manage to get all-in with my opponent preflop when I have AA and he has KK, I will win a little more than 80% of the time.   If my opponent gets lucky THIS time, it doesn’t mean I wasn’t an 80% favorite, of course.    So we return to the trading world:  if I come out on the losing end of the trade, how do I know if I mis-evaluated the scenario likelihoods, or if I just got unlucky?  Well, that’s “the art,” as opposed to “the science,” and what I meant in the previous paragraph is that if you constantly over-estimate your chances of success, writing off negative outcomes to bad luck, you’ll lose money in the long run (note: by “long run” I mean: “a large number of repeated trials,” as opposed to:  “if you hold the position long enough”).

Of course, that’s STILL a tough concept, because we don’t know what “the long run” is in investing.   In poker, especially in the age of online poker, players can play massive numbers of repetitive trials where they can accumulate statistically significant sample sizes to evaluate.  In the investing world, however, it’s unlikely that we’ll ever see a huge number of situations that are exactly the same that we can easily use to evaluate our own decision making with a high level of precision (another quick tangent:  perhaps some technical analysts try to do EXACTLY this:  find the same “pattern” over and over again, and count on the “long run” and their faith in their own models.   I don’t think this works in too many other “types” of investing).

So, another example:  I come from a merger-arbitrage background.   There are lots of trades where we think that the probability of the deal closing is in the high 90′s %s.   When I end up on the wrong side of one of these trades – the rare one that breaks -  is it because I was just unlucky and this was the one deal in every 20 that falls apart?  Or is it because I was wrong in my estimation that the deal was 95% likely to close in the first place?     In general, for good merger-arb traders, the deals tend to work out like more like the poker illustration above – where we can, with experience, estimate the probability of success with a pretty decent degree of accuracy.

I wrote a post about a year ago about a special situation in Norislk Nickel.  In that post, I touched briefly upon the issue of “unquantifiable risks,” one of which I christened, in that trade, “RRR” – “Random Russian Risk.”   That was the risk that the Russian government could just, basically, do whatever the heck they wanted.   I had NO CLUE what this risk was, and it’s important that I understood that I had no clue.  That’s Rule #1, of course: know what you don’t know.

Which brings me to the nexus of this post:  Vringo ($VRNG:  long).  I wrote this post thinking about $VRNG, because I think it illustrates this “uncertainty of uncertainty” perfectly – at least for me..   Cliff notes:  VRNG is suing $GOOG (no positions) for patent infringement.   The case is scheduled to go to trial in a few weeks.    I would estimate that VRNG’s chances of prevailing at trial are somewhere between 30% and 70%.   Note that this is a huge range:  that’s my way of saying that I have incredible uncertainty about my uncertainty here:  I can’t even tell you with any degree of confidence that my risk estimates are correct.    I could write another few paragraphs about the reason there is so much uncertainty, but I feel like it would confuse an already confusing topic.

Let me try to put it one other way: traders pride ourselves on making “tight”  markets on “results” of outcomes – that means we generally expect to be pretty confident about estimating uncertainty*.   My super wide “VRNG probability of success” market is my way of saying that this one is difficult (for me) to estimate.  What I’m leaving unsaid (or not, cause I’m about to say it?) is that anyone who tells you that they have a high degree of confidence in their ability to estimate this outcome is, in my opinion, wrong. **

In the investing and trading world, we usually don’t know exact probability outcomes for our different scenarios.   The key to success is in being able to diagnose the difference between bad luck (which happens sometimes) and over-optimism (which probably happens more) in diagnosing the likelihoods of these unknown outcomes.   This is easier said than done, and perhaps one “solution” is for traders to remember this going in:  we’re probably more uncertain about our risk than we think we are.

related:

Trading Lesson: Notional Risk/Reward Is Only Half The Equation

My Previous Vringo Posts

Norilsk Nickel – Russian Odd Lot Arbitrage

Trading Rule #1: Know What You Don’t Know

-KD

* Again,  if you’re not good at estimating uncertainty, you will, in the end, lose money.  This is true regardless of if you are making prop bets with your friends (for me, this is the easiest way to visualize this concept), or trading stocks based on your estimates of scenario outcomes.

** No, not because I think I’m The Best At Evaluating Vringo… Because I think that I have a good grasp of the risks and uncertainties here, even if all I can do with that grasp is conclude that there is significant risk and uncertainty!

Kid Dynamite is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to Amazon.com. If you click on my Amazon.com links and buy anything, even something other than the product advertised, I earn a small commission, yet you don't pay any extra. Thank you for your support.

The information in this blog post represents my own opinions and does not contain a recommendation for any particular security or investment. I or my affiliates may hold positions or other interests in securities mentioned in the Blog, please see my Disclaimer page for my full disclaimer.

blog comments powered by Disqus