Today we’re going to talk about non-predictive control. Among strategists, I think predictive control often gets the most attention – with talk of superforecasting, prediction markets, and calibration dominating talk about decision making strategies. However, strategies for non-predictive control are just as important to an inspiring strategist. How do you operate in circumstances of extreme volatility and uncertainty? When you don’t know what’s going to happen next, how do you choose the best course of action?
Before we can meaningfully talk about how to deal with that type of uncertainty, we need to elucidate how to recognize that type of volatility, which we’ll do now. Let’s go over the three types of risks that a strategist can face.
The Three Types of Risk
The first type we will talk about is certain risk, is represented here by a clear bag with colored balls in it. Now, as long as we know how much the green, the red and the blue balls are worth, we know exactly, which ball to bet on. We can make a simple expected value calculation of the value of the balls multiplied by the probability of getting the ball and that tells us which ball to take.
A traditional business plan assumes that you’re in a certain risk environment. You know exactly what to do. You just need to create a plan that’s going to consistently pull out the right ball for you faster than anyone else can get to those balls.
The second type of risk is uncertain risk. This is represented by a a bag where we know there’s red, green and blue balls in it but we don’t know how many of each one. So, what we’re going to do is use sample the balls. We’ll start pulling a few balls out, making some small bets or just pulling without paying. As we pull more balls out, we’ll start to get a sense of the distribution and once we’ve done that we can go back to the planning strategy of the certain risk.
This strategy in the business world is analogous to lean start-up and customer development techniques where you’re running small experiments with your business. You are making small bets. You are talking to customers to get a sense of the distribution of where the expected value is and then you move forward and create your company.
Knightian Uncertain Risk
Finally, there’s Knightian uncertain risk. The idea here is that we don’t know what color balls are in the bag. We don’t even know it’s balls or hubcaps or acid, and the contents of the bag are changing and shifting moment to moment. In this case, the predictive strategies won’t work. The idea here is to use the strategies that we’re about to talk about in this post to exercise non-predictive control. There’s two complimentary strategies that I know of for how to deal with this Knightian uncertain risk, the first is called Antifragility.
The idea of antifragility comes from Nasim Taleb. The idea is that you’re going to build a system or organization that gains from the disorder inherent in Knightian uncertain risk. If something is fragile. disorder is really bad for it. If something is robust, it can withstand a lot of disorder. If something is antifragile, it is actually strengthened by disorder. For Knightian uncertain risk, antifragility is really the only one of the three that is viable long term. There are only two ways to be antifragile.
The first type of antifragility is natural selection. In order to take advantage of natural selection, our system or organization must sufficiently diversify its strategies in a real way. We can’t just diversify along one category – we have to truly branch out into other categories and completely different strategies. As long as we have a sufficient numbers of diverse variations, we’ll have a few that survive even the harshest black swans. Whatever strategies are left are the ones that can handle everything that was thrown at us. We then diversify from those strategies to continue to take advantage of natural selection.
The second type of antifragility is Hormesis. Hormesis is creating ideas or strategies that grow from adversity and stress. Our muscles are hermetic, because as we put strain on them they actually grow to handle that strain. Building systems that learn, handle adversity well, and become increasingly stronger due to stress is the second way we can become antifragile.
If you remember back to the congruency episode we talked about masculine and feminine congruency. Masculine congruency is about a system that changes the outside world in order to achieve it’s goals. Feminine congruency is a system that can maintain homeostasis even with external pressure and volatility. Anti-fragility is the femininely congruent response to Knightian uncertain risk. The masculine congruent response is effectuation. Saras Sarasvathy first discovered this process by running through thought experiments with experienced entrepreneurs and picking out patterns. The basic idea behind effectuation is that rather than coming up with a system that is fine no matter what’s in the bag, we’re going to pour balls into the bag.
The Bird in The Hand Principle
So instead of choosing a goal from all possible goals, we start out with what you already have. Who do we know? What are our strengths? What are our resources? Based on that we can come up with a number of goals that can utilize the balls we already have on our person. This is referred to by Sarasvathy as the Bird in the Hand Principle.
Affordable Loss Principle
From those, we’re only going to choose strategies that minimize the downside risk such that we can afford the loss. This will allow us to keep pouring in balls to the bag until we can finally pull on out again. This minimization of downside risk is referred to as the affordable loss principle by Sarasvathy.
Patchwork Quilt Principle
Now, once we’ve selected those goals that match with the color balls we have, we’re going to create the market that will allow us to pull the ball out. We’re going to take those people that you already know and create precommitments and partnerships with them. This has two effects on our new venture.
First, is it’s going to constrain the possibilities for our organization. The commitment will come with conditions, and those conditions will limit the total number of strategies that the organization can use and the total number of outcomes it can effect.
Second, it’s going to give us more resources and certainty for our organization. We’ll have more money, connections, and strengths that we can leverage to make our organization successful. We’re constraining what the organization is even as we are broadening what the organization can do.
Then, we start the process again making more and more commitments. Ultimately creating our new market. There’s obvious parallels here to the recursive congruency process we talked about in the congruency episode. This principle of a series of precommitments defining the organization is referred to by Sarasvathy as the Patchwork quilt principle.
In uncertain and volatile times, non-predictive control methods are critical to success. Every system should be on some continuum from Effectuative to Antifragile, depending on its own values and structure. Please play around with these processes and use them for yourself.