Blog Tools
Edit your Blog
Build a Blog
RSS Feed
View Profile
« October 2008 »
1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31
You are not logged in. Log in
Entries by Topic
All topics  «
Churchill Design
Design Diary
Empire of the Sun
For The People Material
Wargame Design Musings
Wargaming For Leaders
Washington's War
CIO Insight Reference Links
Must-read Books
Leadership Blog
Buy/Order Wargaming for Leaders
Barnes & Noble
Wargaming for Leaders
Book Site
Mark Herman's Wargaming Blog
Friday, 10 October 2008
Topic: Wargaming For Leaders

I hope to begin a conversation on the application of wargames to emerging world class issues, such as National and Corporate survival. I will start by making a simple point, all decisions are based on a set of assumptions. Nothing revelatory about that, but it is this aspect of any decision that in the end determines whether it is a good or a bad one. A good decision is one whose underlying assumptions hold true even as the context under which that decision was made shifts and evolves. A bad decision is one where the shifts or the discontinuities that the decision is subjected to invalidate the underlying assumptions leading to unfortunate circumstances.  Said another way, after every disaster the post mortem reveals someone messed up.

It would be hard to have this discussion without looking at the current Wall Street meltdown. From my perspective this situation failed due to two assumptions, both philosophical. The first was the Government’s assumption that the firms on Wall Street would act in their own self-interest and remain solvent even while taking on enormous portfolio risk. The second was the financial firms belief bordering on hubris that between their professional judgment and their computer models they could shift and weave their way through whatever troubled waters they might encounter.

When I think of the issue of computer models I would make the comment that all models are wrong, some are useful. All computer models are based on their creators’ assumptions as chiseled into computer code on the underlying truths that the simulation is portraying. All computer models are limited by the limits of its programmer. A computer model cannot calculate something its programmer could not anticipate or understand. To put an edge on this point, based on work that we have done, no one that we met in financial institutions believed that the value of real estate would go down. We were successful on several occasions in running wargames where we were able to create plausible scenarios for this to occur and sufficiently suspended the disbelief of the participants so they could grapple with the implications of this alteration of reality.

The current economic crisis makes this point. In an article in the New York Times, October 3, 2008, titled “Agency’s ’04 Rule Lets Banks Pile Up New Debt And Risk, “ Stephen  Labaton wrote, “A lone voice of dissent in the 2004 proceedings came from a software consultant… who said the computer models run by the firms – which regulators would be relying on – could not anticipate moments of severe market turbulence.” Here is a case where someone, who like myself, has spent a career using computer models knows their inherent limitations. Although computer models are useful, they have their limitations. It has always been my practice to marry computer models with wargames as they are natural compliments to each other. One is a power tool that can crunch numbers until the cows come home, but knowing the correct context and circumstances that those numbers should be crunched under is the province of wargames.

I am happy to say that none of the Wall Street casualties were our clients, but we do not claim to be financial geniuses or Delphic seers. What we would claim is we are very good at forcing people to re-examine their assumptions, no matter how sacrosanct they may appear. More to follow…

Posted by markherman at 7:58 PM EDT
Updated: Friday, 10 October 2008 8:01 PM EDT
Share This Post Share This Post
Post Comment | Permalink

View Latest Entries