What do poker, business, and war all have in common? Escalation of commitments. There’s a lot of academic research in the fields of game theory and decision sciences that cover the concept of escalation of commitments in exhausting detail. Most of these studies pertain to decision making relative to resource allocation. So, I’d like to take a real-world approach to sharing some thoughts around this fascinating topic. First, let’s talk about no limit poker. Then, we’ll specifically hone in on business decisions. And, finally, say a couple things about war. After all, we’re in election year.
We all understand that “throwing bad money after good money” can only result in a very bad outcome. Nowhere is this more true than in no-limit hold ‘em poker. Generally, in a game of poker, a player can check, bet, raise, or fold. Each decision depends on a variety of factors. How good is my starting hand? How are my competitors playing? Aggressive? Passive? What communal cards have “flopped” and has it strengthened my hand or weakened it? What is the probability of strengthening my hand with future communal cards to come (i.e., marginal benefits against marginal costs)? What is the probability of my opponent strengthening his hand (i.e., based on my assessment of his starting hand ranges)? What is my chance of bluffing my opponents so they are forced to fold a stronger hand (i.e., what is my seat position and history of prior actions)? There’s an endless list of factors to consider.
To make things more complex, there is an inherent numbers bias against winning with regularity over the course of time (i.e., laws of variance). Beyond the math, poker involves emotion. Poker, however, is not the ultimate topic of this posting. I want to cover the concept of “pot commitment” in no-limit poker. Why? Because “pot commitment” often drives business decisions but most of us never think about it. But, we really should think about it. Pot commitment is not exactly the same thing as escalating commitment. Pot commitment is more analogous to marginal decision making.
Let’s set up the scenario. We’ll skip the basic rule of poker and jump right to the situation that forces a decision based on pot commitment. In poker, it is oftentimes a good decision to “throw good money after bad money.” It sounds irrational. But, it’s perfectly rational because of the notion of pot commitment. I’ve decided to play a particular hand. I’ve invested money. With each additional round of betting, I gain more information … but I also need to keep pumping money into the pot. Near the end, let’s say my odds of winning are reduced dramatically. My odds of winning at the beginning of the hand were favorable or even dominant. But, by the end, my odds of winning look very small. And, yet, the best decision sometimes is to continue putting more money into the pot. This happens because the marginal cost of continuing to stay in the hand is much smaller than the marginal benefit of winning the pot.
A more subtle concept that most people do not pay attention to is that the critical decision actually occurs not at the end (i.e., where the player is forced into pot commitment) but rather in the immediate prior stage. During the middle betting rounds of the hand, a good player will bet, raise, or call by factoring in whether he will end up being pot committed – as a consequence of his action - in the next (and final) round of play.
Let’s apply the notion of pot commitment to business. More specifically, let’s talk about startups. How many times have you seen VCs continue to pour money into a startup that is going nowhere? It happens all the time. Similar to the poker scenario, many of these decisions are actually (gasp) portrayed as good ones! The fine line between good and bad decisions really rests on whether a decision is based on marginal decision making or escalating commitment.
The most typical scenario where a VC continues to invest in a startup that has never tasted anything but ongoing failure is when the prior investment is considered a sunk cost and future investments are based on marginal benefits which, in turn, are assessed on new information. The two decisions are fairly detached.
I have a big problem with this line of thinking. First, the new investment assumes that the prior negative feedback (i.e., failure) is not informative because the startup has completely repositioned itself, going after a much bigger opportunity, changed its product significantly, changed its management team, replaced he CEO, etc. In truth, most scenarios involve very informative negative feedback. And, negative feedback of this sort really affects marginal decision making. Yet, many startups get funded due to escalating commitment – NOT marginal decision making. Second, sunk cost infers that the follow-on investment is being made into a new game – not the same game. But, many of these follow-on startup investments are not in a new game. Essentially, they are not fresh money being thrown into a fresh pot. Instead, they are the classic “good money being thrown at bad money.”
In business, many failures must be categorized as informative negative feedback. And, factoring in informative negative feedback means that prior investment should not all be considered sunk costs.
So, what are some suggestions for avoiding the escalating commitment and exercising marginal benefit decision making?
Rule #1 – Make sure that if something is considered a sunk cost, it is thrown away completely. Too many startups try to start over but keep the failure(s) sticking around the house. It’s the classic startup and VC head fake to deceive … itself.
Rule #2 – Do not keep calling negative feedback uninformative. Failure is a blaring alarm. Failure is almost always informative feedback. Unlike its poker counterparts, startups usually do not know precise probabilities of success and failure. When precise probabilities are not known, smart pot commitment scenarios are unlikely to occur. Therefore, startups and VCs must be weary of falling into the escalating commitment freefall or trap.
Rule #3 – Use my Test of 100x … that is, ask yourself this question, “If I had 100 times more resources, would I still have failed?” If yes, you probably are in the wrong business. If no, you’re probably in the right business but have the wrong business model and/or execution.
Rule #4 – Make sure you’re targeting a huge opportunity so that the hits from failure can be softened in context of the larger goal. In other words, make sure a $1 investment with a 1/100 chance to succeed will earn $100,000. Said differently, if you’re going to throw good money at bad money, make sure the total size of the reward has increased exponentially. Otherwise, fold your hand.
Now, I have a question for all of us to ponder regarding our war in Iraq. Does continuing our war in Iraq represent marginal benefit or escalating commitment? And, if we decide not to continue our war in Iraq and get out, would this be due to a lack of marginal benefit or refusal to fall for the trap of escalating commitment?
- John
Comments