Tuesday, April 28, 2009

Predicting things is hard

My friend The Wobbler recently twondered (that's tweeting and pondering, or possibly wondering) on the value of long range forecasts. I believe any of us who have ever been involved in long range forecasting and lived long enough to see what happens can agree that any value it may hold has little or nothing to do with accuracy.

As an example, I once worked for a consulting firm here in Louisiana that specializes in disaster and anti-terrorism planning. In the mid-1990's, since it was obvious that no one was interested in planning for terrorism, and emergency managers had the lowest budgets anywhere, we made plans to become an information technology company. Then 9/11 happened, so we focused on homeland security, as it was now called. IT was still important. Disaster management was relegated to the background. Then came a few active hurricane seasons in a row, capped by Katrina and Rita. In terms of mission and capability, the company is now back to almost exactly where they started. They are just bigger.

So forecasting is a waste of time, right? Maybe not. An old professor of mine used to tell a long and extremely uninteresting story about his early days in scientific computing. He was assigned to create a particularly difficult calculation, and every time he needed a value for the calculation, he was told to estimate it. Or extrapolate it, interpolate it or some other fancy term for guessing. In the end, he found that the experience, while frustrating and not providing a very trustworthy answer, allowed him to make progress he would not have been able to make otherwise. As he said, "Sometimes any answer is better than no answer at all."

So maybe there is value in prediction. After all, the focus on IT made our company better prepared when the unpredictable did happen. New capabilities were created, and success followed. But perhaps an understanding of what a forecast really gives us should inform our choices about how much effort to put into it, and how much to rely on what it says. Perhaps if investment bankers and hedge fund managers had a better understanding of this issue, we would all still have the money that was in our 401K's a couple of years ago.

Friday, April 24, 2009

So, it turns out that conferences kind of suck if you:

1. Don't know anyone.
2. Have never been before.
3. Aren't really interested in the subject matter.
4. Are too old to drink yourself stupid every night.

I've been to two of these in the last two weeks, and I could seriously stand not to go to another for a while. Not that I don't like HBO as much as the next guy, but in spite of (4), my natural inclination when lonely, bored and slightly anxious is to hit the bar and have that "just one more" that turns into a nagging headache way too early in the morning.

I have noticed that conferences seem to be having a hard time attracting anywhere near the number of participants they are used to having. I'm sure the economy is part of it, but it feels deeper, somehow. Like there is some fundamental need that used to be met by these gatherings that is either no longer needed or is being met another way. Sort of like the way the internet porn has made Penthouse and Playboy seem irrelevant. I don't know if it's a temporary thing, or if the industry conference is destined to go the way of the three-martini lunch.