Artificial Incompetence

Or

How To Build An Idiot Commander

When Humans Just Won’t Do

Like most of the readership here, I’ve spent many decades of gaming building rule systems at the tactical and operational levels which have tried to model unit and sub-commander skills, often referred-to as CE or ‘Combat Efficiency’ in the modern era. In many ways, it’s been quite easy – in simple probability terms, a unit rated 5 is going to give serious trouble to one rated 3 (I was taught to compare the squares of each factor to see the real ratio of effective comparative power). Fair enough.

But what at the higher echelons, where the whole nature ne course of a campaign hinges on pitting a key commander – a Napoleon, Alexander or Schwartzkopf – against a trained but either useless or seriously flawed opponent? I mean, we can easily give a Napoleon player (or team, as I proposed once) so-many ‘action points’ compared to an Austrian’s 2. Leaving aside deliberate disruption (using all resource points for 1812 to re-equip the Old Guard with pantomime chicken costumes; that’s a megagame control issue), for most players – baring some classic idiots and false modesty aside – those at CLWG with even a modicum of military skills let alone ‘period knowledge’ would give it a reasonably good go. No, my problem is with their opponent, the ones who allow such genius to shine…….

There are, of course, several possible solutions:

• In megagames, this has often involved ‘casting’ – without naming Wat-I mean, names, there are a few key idiots around who can usually be depended upon to Do The Wrong Thing. Indeed, Brian used to marvel at a couple of such treasures I would regularly employ in my own games at SELWG. However, apart from one disastrous, a-historical episode at a giant WW2 megagame where the Japanese team was crippled, most such player outings involved simple ‘operational’ choices – go here/there, now/later etc. Broader command choices were either beyond them – rabbit in the headlights – or were not applicable to the game.
• There is also the use of full-on role-playing techniques. However, the problem here is that any given player has to have a very good character brief and/or existing knowledge of the idiot concerned, plus they have to be a damned good actor to carry it off! So no, this would only succeed at the cost of a rare, skilled player/umpire. And as was found at one of the ‘Master of Europe’ games, even a well-read player can go off the rails, with that particular Napoleon ignoring several increasingly heavy hints about political actions (before leaving early).
• My solo games have sometimes used a set of pre-designed ‘activity cards’ which can be used to create an automated opponent but with a realistic set of options/ideas within their historical context – and Either the cards can be named activities – advance with one corps, visit Washington for consultations, sulk in your HQ etc – or a simple points value and dealt-out to again defined pots (with plenty of blanks for ‘no action’). But while closer to my prejudiced historical guesstimate/model, this can be a bit too random, and not subject to outside stimuli/circumstance.
• But what about using players to create said incompetence as a group exercise – except they don’t know they doing so? As one of my favourite films said, “…The best spy is one who does not know they’re a spy…”. This set me thinking….

So, given Real Idiots are both too unpredictable and unskilled (as well as scarce, thank the gods), I have been playing with a systemic approach which I have dubbed ‘Artificial Incompetence’! In my never-to-appear ACW megagame of the 1862 operations in the Virginia Peninsular, one fundamental aspect to all of this is the actions of the Union C-in-C, good old General George Brinton MacClellan, the ‘Little Mac’ of (in)famy.

His really rather brilliant strategic concept was to move the bulk of the Army of the Potomac (which he had organized and rebuilt after Bull Run) by sea around the direct route to Richmond, landing at the end of the essentially unguarded James river. The normally enormous logistic problems associated with the 100,000-strong AoP would in fact now work in the Union’s favour, as the Navy could easily supply all needs, while the spring rains had turned much of Virginia into a flooded, swampy mud-bath, with the very real prospect that the Confederates would be unable to move fast enough to block him. Alas, once field operations commenced, Little Mac became increasingly reluctant…

What I have in mind, therefore, is – like Jim’s ‘cardboard hitler’ – is not to have a MacClellan player at all. In his place, the senior Union corps commanders will have a series of ‘advice to the great man’ pots in which to allocate various points each turn. The key things about this are:

• The ‘advice pots’ have a direct bearing on various aspects of their personal success These are handy for any attempt to court martial them during the game, as well as promote themselves as future contenders for CinC in the eyes of Congress, Lincoln, the Press etc.
• However, hidden from the players is exactly how the different advice pots affect MacClellan’s actions are perceived by Lincoln, Congress and the Press etc. Thus the correlation between player advice and MacClellan/Realworld effect is sort of offset. Of course, after a while they may begin to see a pattern emerging, but only in general terms – plus of course their ‘advice’ is given in secret; an erstwhile colleague may be an intense rival…
• In this game, once the ‘advice pots’ are completed, one of the Corps Cmdrs is selected to be the main ‘confidente of the day’. They will then examine the ‘operational pot’ to see how many action points are there, then issue extremely vague (but historical) orders to one or more corps for operations this time, or until orders/conditions change. The cost of such orders increases exponentially, so there will never be sufficient to simply order the AoP to roll forward and crush their opposition – as would happen in any ‘normal’ game. Instead, even if the Corps players co-operate and agree a plan of campaign, it will be very stop-go and disjointed – but in an acceptable and non-random manner of execution.
• The players will also vary their ‘advice’ according to both evolving personal and game circumstances – skewing things if a rival is doing too well, or going for that ‘final push’ if Richmond really does look like its about to fall…

So, by linking a set of sub-player choices in an offset table of historically acceptable consequences, we should have a system-driven set of command decisions, both operational and political, without recourse to a human. There is still much to do of course – even though it will never be run, I’m still determined – like the Balkans system – to complete the design. However, I am very pleased with progress so far, and have even drafted some other models (to test my assumptions) for The Sikh Wars, AWI, Franco-Prussian and early WW1 encounters (Eastern Front Austrians, of course), with interesting results.

Well, that’s it really – in every sense. If anyone wants to see the detail of ‘Little Mac’ I’ll happily send some draft docs (although they are very terse, as is my style). Don’t worry too much about responses, though, as after 46 years I won’t be renewing my CLWG membership. I wish you all well and may even see some of you at the odd megagame or toy soldier show. I hope so.