So . . . a lot of what I do at work has to do with outcomes. I know that this does not make me unique, but as I study outcomes, rather than produce them, I spend a lot of time thinking about the nature of outcomes, how and why they matter, and how they are produced. For example, I teach a course in marketing that utilizes the case method, which means that each session involves making a decision regarding a marketing situation. The students are taught that there is no single right decision - you must make the best decision you can and do your best to sell the idea.
There is a flaw in this thinking that has been bothering me (clearly I focus on big-picture issues that impact the world as a whole). You can make the best decision, but it could still not work out. I do incorporate this idea in my course by emphasizing that the decision that the company in the case ultimately took (most cases are based on real-life situations) is not necessarily the best decision (even things worked out for the best). This is a difficult notion to swallow, because if things worked out well, it must have been the best decision.
I'll illustrate this with a very basic example. Let's say you have three lottery tickets for a draw two days from now. A friend has one lottery ticket for the same draw. He proposes that he swap his one lottery ticket for your three. Would you take his offer?
Most reasonable, rational people would say no - you are reducing your odds of winning by 2/3 (sorry if you said yes, but if you did, let me know, because I'll give you one ticket for three anytime). Refusing the trade is the absolutely, 100% right decision at the time you made it.
But let's say that the draw occurs and none of your three tickets win, but your friend's one ticket does. Now you feel stupid, because you could have won. It would be very difficult to convince you that you made the right decision at this point. This creates a paradox of outcomes, because the same decision is both completely right and completely wrong.
To use terminology suggested by a friend, refusing the trade was the best decision and accepting the trade proved to be the right decision. The problem occurs when we confuse the two, which is very, very frequently done. For example, we are rewarded in life for making the right decisions, not the best decisions. If you make a choice about something at work that pans out, you are a hero. If it fails, you are shunned (a bit extreme, but I'm sure there are some workplaces where shunning occurs, such as in the Amish community).
This is unfair, because whether or not you made the best decision is ignored, and only whether you made the right decision matters. Probabilistically speaking, the best decision will more often result in the right decision, but not always; in blackjack, for example, making only the best decisions gives you approximately a 50% chance of winning. So really, we are rewarding people for chance outcomes most of the time. Not only is this unfair, it is counterproductive, because you are not promoting or retaining the best decision-makers, but rather the lucky ones.
I'll have more to say on this in later posts (I'll bet you can't wait!)0, but I wanted to introduce the idea. I find it tremendously interesting, though I'm sure not everyone does. To put the paradox of outcomes in a nutshell: on one hand, the outcome of a decision has little to no bearing on the quality of the decision, but on the other hand, the outcome is the only thing that matters. Thoughts?
is this not as simple as 'hind sight is 20/20'?
ReplyDeleteas well, the lottery example is a great one, but the outcome is one event, and not generally seen along a continuum of outcomes. that is, decision a creates good outcome b, which influences situation c to create bad outcome d... which impacts... and so on.
so it becomes a case of where you draw the box around the decisions and the outcomes, and the perspective and lenses through which you view it.
winning the lottery looked like a really good outcome to Hurley before everything went to hell.
I don't know . . . I think it becomes more than hindsight being 20/20, because of the judgments that are made regarding the decision-maker.
ReplyDeleteI agree that outcomes can have unintended future consequences and that you can't isolate one behaviour as causing one outcome (except in game-like situations). But I think that this only strengthens my point, because we tend to do exactly what you say (drawing the box), even though we do it incorrectly. And so we don't only overlook the role of luck/chance, but also ignore other contributing factors.
In the lottery example, though, I suspect that the Hurley situation along with the many stories about "how the lottery ruined my life" becomes a bit of a narrative fallacy, which I'll probably cover in a blog post in the next week or so.
The best discussions I have seen (and had) recently about this idea actually followed Bill Belichick's decision to go for it on 4th-and-2 against the Colts this past November. A variety of statistical approaches suggest that it was essentially a 50-50 call (in terms of probability and expected outcome), but hindsight bias led to him being ridiculed for making the "wrong" decision.
ReplyDeleteA writer at footballoutsiders.com wrote a great column (or part of a column) explaining this idea (along the same lines as what you are discussing here), but I don't have time to dig up the link right now.