Chapter9.gif (961 bytes) Wargames at War

Types of Wargames

While commercial wargames fall into only two types; manual map based exercises and computer versions of same, there is a far wider variety in the professional level. Models and simulations are another matter, many wargames containing a little of both. In wargame parlance a Combat Results Table (CRT) is an "attrition model" and the more elaborate CRTs are indeed models in a very real sense due to its replicability and static representation of a process. Wargames usually contain several (or many) models linked together in a system. This, of course, is the classic description of a simulation, along with a simulations ability to handle multiple scenarios in a more interactive manner. The primary difference between a wargame and a simulation is fuzzy, based on the concept that a wargame is not capable of multiple runs from which statistically significant results can be derived. For many manual games, this is generally true, in a practical sense. However, once a manual wargame is turned into a computerized version, you can let it play it self a sufficient number of times to obtain statictically significant results. The advantage of the manual wargame is that human players can obtain broader insights from it and become better able deal with the intangibles of a situation. That said, some of the wargames described below are simulations and all these wargames contain models.

The following list shows the major types of wargames and their primary characteristics (for comparative purposes).

Manual Model with Map

What is normally thought of as a "commercial manual wargame". (The original military wargames were of this type. But that was before computers and beltway bandit consultants.)

  1. Forces- Order of battle, all units involved in simulation. Must be consistent with scale of model. Optimum for playability is no more than 20 units per side.
  2. Movement- Each unit assigned a numerical value representing its ability to move across terrain.
  3. Combat- Each unit assigned a numerical value for combat ability.
  4. Map Display- Choose scale carefully. Optimal size of map is 20 by 24 inches, or the distance that players can reach units without assuming awkward position. Hex grid is used to regulate movement and combat. Each hex cell contains a discrete type of terrain which shows up on Terrain Effects Chart with its effect on movement and combat.
  5. Rules of Use- Explicitly written out procedures to operate model. This also gives insight into the underlying process that drove the situation being modeled.
  6. Easiest model to create- Best preparation is simply extensive playing of existing games.
  7. Inexpensive- Paper is most common raw material.
  8. Paper computer- System organizes processing of information in much the same way as a computer, only much more slowly.
  9. Easy to maintain- Procedures are largely self documenting because they are made obvious to the player. Otherwise, the model would be unplayable.
  10. Labor intensive to use- An average size game will take 2-4 hours to play to a decision. Larger ones take much longer.
  11. Not highly iterative- Time required for each game takes too long. Replaying individual turns has value, and proceeds much more quickly. Numerous iterations are required of a game in order for its results to have statistical significance.
  12. Precursor of Computerized version. Programmer needs a manual model to work from.
  13. Time Required- 500-2000 hours- Varies considerably with skill of creators. My personal record for a published model is 12 hours from cold start to tested prototype ("Battle for Germany" in one session from 6 PM to dawn, 1975). Another hundred hours required for testing and finishing rules. Lack of sufficient skill will make successful design impossible no matter how much time is used.

Manual Model without Map

Can be described as either a "commercial manual wargame without a map", or as a staff study with easily modified parameters.

  1. Forces- Same as model with map.
  2. Movement- Deduced as a result of force on force calculations.
  3. Combat- Adjusted force ratios compared and results computed.
  4. No Map Display- Spatial positioning is derived from player decisions and results of combat. Somewhat abstracted.
  5. Rules of Use- Same as model with map.
  6. Easiest model to create- About as easy as the one with a map.
  7. Paper computer
  8. Inexpensive
  9. Easy to maintain
  10. Labor intensive to use
  11. Not highly iterative
  12. Precursor of Computerized version
  13. Time Required- 400-2000 hours- Can be less time than map version because map does not have to be developed. Can be more difficult for the same reason if the model is complex.

Spreadsheet Combat Model

A "manual model without map" put up on a spreadsheet program.

  1. Similar to Manual Model without Map
  2. Forces- A larger Order of Battle can be handled because the computer keeps track of details rather than the player.
  3. Combat- More complex combat routines can be used, again because of the computer.
  4. No Map Display- One can be used off line for reference purposes.
  5. Rules of Use built in- Allows users to be trained much more quickly.
  6. Some graphics capability (charts)- Depends on the spreadsheet you use. Most have a graphics capability.
  7. Highly iterative- Lotus spreadsheet products have a data table feature which makes it much easier to do sensitivity analysis.
  8. Constructed on Spreadsheet Program- Recommended ones are 123, Quattro, and Excel. Improv also good.
  9. Time Required- 300-1200 hours

Cost/Benefit Model- A Spreadsheet Combat Model optimized for evaluating individual weapons (or other item) performance.

  1. Similar to Spreadsheet Combat Model
  2. Measures effectiveness of weapons systems
  3. Forces- Extensive list of weapons systems and variations can be analyzed.
  4. Combat- System on System, duel type engagements. Synergism of many systems must be abstracted.
  5. No Map Display
  6. Some graphics capability (charts)- Important for this type of analysis as a large amounts of numerical data is processed and graphics makes all this easier to absorb.
  7. Rules of Use built in- The "rules" are the formulas for processing the numbers.
  8. Constructed on Spreadsheet program- 123, Quattro or Excel are best for this because of its full array of spreadsheet commands. Also has windowing and easy to use and powerful command language. Any other spreadsheet will handle the essentials. 123 is still particularly good for these jobs, particularly because of the large number of add on programs available for handling Linear Programming, database file access and other tools.
  9. Highly iterative
  10. Time Required- 300-1500 hours- Depends on how elaborate you want to get.

Expert System Combat Model

Models of expert knowledge on various aspects of combat operations.

  1. Forces- Generally not modeled as extensively as in other simulations.
  2. Movement- Abstracted.
  3. Combat- Handled in richer detail, usually in form of interrogation.
  4. Normally No Map Display- Not needed in most cases, but can be added.
  5. Rules of Use Built In- Especially decision making rules and options.
  6. Highly Iterative
  7. Heuristic- Depends on how you set it up.
  8. Natural language interface- Most of the Expert System Shells allow user to communicate with system in plain language.
  9. Created on Expert Systems generator
  10. Can be made part of computer wargame- Works best as complement of other forms of combat models.
  11. Can be written from scratch in LISP or PROLOG- This takes much longer than using shell, although shell version can be done first as a form of prototype. This will make it cheaper to do in LISP or PROLOG as the systems analysis and design will be much more complete.
  12. Time Required- 200-2500 hours- Depends on several variables. Namely use of shell and degree of elaboration.

Computer Combat Model with Map

Computerized version of "Commercial Manual Wargame".

  1. Similar to Manual Model with Map
  2. Ideal model for senior decision maker- Powerful model with easiest user interface.
  3. Can be easiest to use (if done right)
  4. Very expensive to create and maintain
  5. Highly interactive
  6. Forces- Can handle larger number than manual model.
  7. Movement- Can be more elaborate than manual model.
  8. Combat- Can be more elaborate than manual model.
  9. Map Display- Can be more elaborate than manual model.
  10. Rules of Use Built In
  11. Time Required- 2500-12000 hours- This can be done using a workstation or high end PC based development system). If the coding is done in C, that code can be ported to other machine environments (Unix).

Computer Combat Model without Map

The infamous, traditional, "black box" model. Or something very similar to it.

  1. Similar to Manual Model without Map and Spreadsheet combat model.
  2. Very expensive to create and maintain
  3. Can be highly interactive
  4. Forces can be as large as you want.
  5. Highly iterative, unless it chews up too many cycles doing its basic calculations.
  6. Movement, but usually abstracted
  7. Combat, usually abstracted
  8. No Map Display
  9. Rules of Use Built In
  10. Time Required- 1500-12000 hours. In practice, some of these projects have gone on for years, apparently acquiring a life of their own.

A Quick Check List: Important Things to Remember When Creating Models, Simulations and Games

  1. Pay VERY close attention to the user- If they can't use the model, it may be the last one you'll do for them. (Not generally the case for Department of Defense contracts.)
  2. Don't exceed your capabilities- Try to do more than you're capable of and you'll make a mess of it.
  3. A modest success looks better than a grand failure. Know your limits. Start With simple task. This is particularly essential if it is your first effort.
  4. Use Previous Examples. No need to reinvent the wheel. Good ideas and techniques can be constantly be recycled.
  5. You can never test enough- Testing should use the same procedures applied to software. There are several levels of testing.
  1. Unit Testing-Test individual rule for soundness. Example, test map for completeness and correctness.
  2. System Testing- Test rule along with other rules that it normally operates with. Example, test map with movement and combat rules.
  3. Integration Testing-Test all rules together to see that all parts fit correctly.
  4. Validation Testing- Test entire system to see that all user requirements are met.
  5. Acceptance Testing- User tests to see that all requirements are met.
  1. Experience makes a big difference- Life experience (someone who is careful, persistent and thorough) is valuable, more valuable than a lot of experience with using games. The techniques of modeling are simple. But they must be followed carefully in order to work. Carelessness can easily be fatal.
  2. The model is never completed- You will always be coming up with new techniques. User will continually come up with new requirements. Much the same experience as with software products in general. Bugs will continue to appear, although of a less severe nature over time. User will almost always require enhancements and other modifications. Use of the model will point towards more efficient ways of doing things and will indicate that it is cost effective to make revisions.

Use of Historical Data in Models, Simulation & Gaming

It's often a less than perfect fit between historical data and modeling requirements for military professionals. There are numerous techniques that can ease the process. Here's a check list. If you run into a snag while trying to develop a game, run through this list again, and again, if need be.

  Differences Between Hobbyists and Professionals

  Why Use History?

  Table of Contents

Chapter9.gif (961 bytes)  Chapter 9 Contents