The Achilles’ Heel of military simulations

The rigging of the Millennium Challenge 2002 simulation echos events in the planning for the Vietnam War.  Chapter 21 of David Halberstam’s The Best and the Brightest describes the elaborate simulations run to prepare for our involvement in Vietnam, each involving weeks of preparation and conducted by senior military and civilian officials.  The first set did not go well for the Blue Team.  Then we see the first weakness of all analytical tools:  moral weakness of the human components, as they cheat in order to get the desired result.

The second set of war games went a little better. … There was a greater US willingness to commit more and more of its resources to the war, and corollary change among the North Vietnamese, a downplaying of their willingness to meet the larger American commitment.

As Joshua Foust says, in his forthright manner: 

… the military is not adverse to rigging games so America always wins.  This is an endemic problem: one interesting bit of information I’ve gleaned from the volumes of self-praise in Thomas Barnett’s latest book is that the DoD absolutely loves hearing how great it is.  Though understandable — who does not like having his ego stroked oh-so-lovingly? — it is a critical fault for the agency responsible for the defense of the country and the execution of its foreign policy.  Hearing “the past decade of research you’ve done into air power, robots, special forces is kind of wasted effort” doesn’t make one any friends.  Saying “here’s how you can take your tactics to the next imaginary level,” on the other hand, does.  (source)

Both the pre-Vietnam War games and Millennium Challenge 2002 also illustrate the the second weakness of any analytic system:  no matter how good the simulation, they have no utility if decision-makers choose to ignore the results.

… the real lesson of the games, and it was not a lesson they wanted to talk about, was not how vulnerable the North was to US bombing, but how invulnerable it was, how much of an American input it would require to dent the North Vietnamese will, and how even that dent was not assured, and finally, for some of the more neutral observers, the fact that the basic strategy of limited bombing already split the civilians and much of the military.

My conclusions

  1. Simulations can provide a competitive edge for conventional military organizations, especially when done with size and complexity that our non-state 4GW foes cannot duplicate.
  2. Simulations, like most tools, require moral courage in their operators (i.e., willingness to accept career risk), an organizational culture focused on external realities (such as battlefield success, not court politics or domestic pork spending),  and some intelligence in the senior decision-makers (not necessarily genius-level, but not fools either).

Military simulations, like all analytical tools, must be run correctly to produce useful results.  Their results have impact only if listened to.  So simple to say, so difficult to do.

Update:  Zenpundit‘s note on this series is (as usual) worth a look.  Esp. his comments on the nature and function of gaming, both military and in general.

Previous articles on this subject

  1. Recommended reading: an autopsy of the 2002 Millennium Challenge war games
  2. War games, the antidote to “Victory disease”
  3. Are war games a competitive edge of conventional forces vs. non-state 4GW foes?
  4. During Millenium Challenge 2002, by Ed Beakley (Project White Horse blog), posted at the DNI blog
  5. What we should have learned from MC02, by Dag von Lubitz, posted at the DNI blog

4 thoughts on “The Achilles’ Heel of military simulations

  1. There was a good milsim about US v “most militarily advanced mid-east opfor”(not specified, but Israel). In this game the opfor sent many small boats against a carrier group, and destroyed the carrier. The game was reset, and the US won. USMIL declared that the game was more to train our soldiers than to test RW mil outcomes.

    Sims are not popular when they don’t conform to conventional wisdom.

  2. Then you could do these like Monty did. He was famous as the greatest trainer the British (possibly any) Army has ever had.

    But his exercises were real (as physically possible). One example (this was in 1941) he had his men marching and fighting for 10 days across the country (in many ways a rehersal for the later El Alemain). In France, in 1940, he was asked to do one of the trickiest moves a divison can do. Did it perfectly, because he had already practised it previously with all his troops.

    He also used them ruthlessly to weed out incompetent officers. There was no bad soldiers in his view, just “useless” (quote) officers. His reputation was so bad (good?) that when it was announced that he was taking over the 8th Army, many officers packed up straight away as they knew they were going to be fired. (Yes the story about the officer who would die if he exercised was true).

    On the other side an amusing and totally unrelated story about him. The 8th Army was (in)famous for its lack of dicipline about uniforms (which he encouraged, by example, as he saw it was a Citizen’s Army). Monty only made one order about uniforms, after he saw a soldier wearing a top hat (and shorts and no top), he ordered (after laughing, in one of his orders to the troops) that “top hats are not allowed uniform for 8th Army soldiers” (in a very jocular way). And people wonder why his soldiers loved him.

    So simulations/exercises/etc are useful but they must be as real, honest and tough as war itself. And there must be casualties in the proven incompetent officers.

  3. That’s a valauble reminder about how war games have been used in the past — and their role in officer selection. Thanks!

  4. Computer simulations have their uses but you can’t beat getting out on the field and actually doing it.

    The great thing is that it is harder to cheat when you have thousands of troops (etc) out there. It either works or it doesn’t. The Van Piper example (Millenium Challange) was a classic example of the weakness of simulations. Refloating the aircraft carriers. A proper exercise (where the real carriers had to sail off as they were now marked as destroyed) would not have allowed that chicaney. Then lessons would have been really learned.

    As another aside:

    The other Monty story I love was (this was apocryphal), on his way to take over the 8th Army:

    He was getting on a plane to Cairo and mused to another officer that being a General was hard. You get a few victories and then you are feted and known and admired around the world. People start to believe that you are invincible. Then nemesis strikes, you lose a battle and then you are finished and forgottten by everyone.

    The officer turned to Monty and said (mindful that the desert had been a graveyard for British Generals) “but sir this won’t happen to you”. Monty replied” “Me? I was talking about Rommel you bl***y fool”.

    But, whatever else you may think of him, the man could train.

Leave a Reply