Don’t Lose Your Marbles!

Not long ago, I had the opportunity to observe participants in an estimation contest. Participants were given a problem along the lines of figuring out how many marbles are in a jar or balls in a pit. Participants then had to come up with an approximate answer based on the information that they could glean from the scenario: for example, they might be able to at least approximately measure the size of the marbles.

I was particularly impressed by one of the groups: they were very analytical. They discussed the problem in very reasoned terms. They only included very few people in their discussion. They came up with a very well-written, very logically developed answer. They were very wrong. While all the other answers clustered around the correct response, this one group had an answer that was so far out in left field that it was in some other stadium.

This, of course, is the challenge in estimation games: it’s easy to make very simple mistakes early on and then run merrily off along a completely wrong, but apparently logical, trail. In estimation games this is pretty much harmless. However, in more general areas of problem solving the same errors that can derail an estimate can also lead to much more significant problems. This isn’t necessarily all that surprising in that real-world problems are much more similar to estimation games than we might like to acknowledge: they often require us to make assumptions, act with incomplete information, make deductions about facts we cannot easily observe, and come up with a best guess at the end. Fortunately, there are some lessons we can draw from estimation games that will improve real-world problem solving, particularly when people are involved. Consider, for instance, any scenario in which you need to work with other people or cooperate with another organization and where your goals are not necessarily in complete alignment.

It’s easy, as the left-field group did, to limit participation in the discussion. In fact, this is often necessary, as too big a group can easily become unmanageable and prevent any productive discussion from taking place. However, keeping the size of the group small does not mean keeping the knowledge base available to the group equally small. Group members need to track their assumptions and the conclusions based on those assumptions. They then need to go out and verify as many of their assumptions as they can: it’s easy to find evidence to support your conclusions; the hard part is looking for evidence that will contradict them. The second is what needs to happen. Identify what information would tell you if you’re making a mistake, and then figure out to identify that information. Speculate. Play “what if?” games.

Of course, it’s also important to remember that marbles in a jar have no motivation (outside of a bad joke about method acting). Situations dealing with people have considerably more moving parts.

An important part of “people estimation” is understanding motive. When dealing with human systems, being able to think about what other people are doing and why they might be doing it is critical. Peter Ossorio, of descriptive psychology fame, makes the argument that in any given situation people will try to make the most advantageous move they can. This doesn’t mean that they’ll always get it right or that they’ll always execute even a right action correctly or effectively. However, it does mean that it can be very productive to consider how other people might view a problem or situation and consider what their likely course of action might be. Consider as well why they might doing what they are doing.

Their reasons might not be obvious, they might not be comfortable for you to think about, and they may contradict some of your basic assumptions, but those reasons exist. Figuring out what they are goes a long way to enabling you to make a better “estimate” and take actions that are more likely to get you to a result that you like. To be fair, maybe the other people involved are stupid or evil; I’ve certainly heard this given as an excuse for not considering their perspective. Ultimately, what difference does it make? They will still take actions, and your success may well depend on your ability to anticipate and work with or around those actions. Approaches which shut down speculation and exploration are most likely going to do nothing more than decrease the accuracy of your estimate.

When dealing with marbles in a jar, being in left field just means that you’ve failed to win a prize. When dealing with people problems, being in left field might just mean that you’ve lost something considerably more valuable. In this case, maybe it’s not so bad if you’ve only lost your marbles.