Balance and Bias

In a recent post, JWH of Heretical Wargaming raised the interesting question of wargamer bias, specifically solo wargamers being biased against a particular army, force or nation. This is an interesting question and got me thinking, as one or two of the comments in the sequence of posts on the Machiavelli campaign might have indicated.

I do not have a good answer to the question of bias. From a young age, for example, I was taught about Agincourt, Trafalger, and Waterloo, and hence might have imbibed some bias against the French. Similarly, my early life was surrounded by comics in which Germans soldiers shouted ‘Achtung’ or ‘Himmel’ rather a lot. As my education proceeded and the horrors of the Nazi regime before and during World War Two came into focus, bias against German armies of the period could surely have followed.

I am not sure that there is a lot I can do about my inherent biases. I can educate myself, reading about, say, French Napoleonic victories, both sea and land. I can certainly examine the achievements and limitations of both sides, or consider the relative exhaustion levels of national resources and so on. For World War Two I can look into the way that the British Empire, fighting for survival, used the colonies for resources, impoverishing the inhabitants for a war of which they knew nothing. Whether any of this would change my biases I am not sure. Given that I wargame neither Napoleonics nor World War Two I shall probably never find out.

Nevertheless, there is another issue which the playing of Machiavelli has raised. As I recall, from playing it with multiple players, everyone had a shout. We know, after a few games that in order to survive and have a chance some areas were important. Venice needed to grab the Adriatic, for example, and France the Gulf of Lyons. Other than these points the game was well balanced and every nation could potentially win.

Surprisingly, this is a bit of a problem. We are honed, perhaps as children, to seek balance. Everyone has to have a chance, at least roughly of winning. This is, often, drummed into us. As a review I turned up on the Internet remarked, possibly Machiavelli is too balanced. The author did not elaborate on the comment, so what follows is my interpretation. I think, however, it follows for wargame scenarios as well.

If a game is balanced, everyone has a chance to win. Victory and defeat, therefore, are dependent on marginal differences. In Machiavelli, for example, the ability to swiftly occupy certain sea areas makes a big difference to some states. These seemingly minor moves can have bigger effects later. If the Venetians do not seize the Adriatic early, they have to expend time and resources in capturing it later, which is awkward to do because only a limited number of fleets can be brought to bear on the problem. Thus the Venetian actions (or lack of them) in the first couple of game moves can lead to an irretrievable problem a bit later on.

The problem can, of course, be compounded by the actions of other players in a multi-player game. The Turks, for example, can set out to obstruct the Venetians; indeed, it is in their best interests so to do. The interactions are part of game, of course, but again, a turn or two delay at the start of the game can lead to problems later on. Given the game balance, these issues take on a bigger function in determining the game outcome.

Similarly, I suspect, many wargame scenarios are designed to be, in some sense, balanced. Even in a game with unequal forces, both sides have a roughly equal chance to achieve their objectives and, hence, win the game, whatever winning might be within the scenario. Given this balance of forces and outcomes, it seems to me that we are back with the Machiavelli balance problem: small mishaps early on can lead to major outcomes, even including defeat.

This is not bias, of course. In fact, unless you start to really analyse what is going on, you might well miss it. My suspicions about which areas are key in Machiavelli games comes from playing it with a bunch of physics and computing students. Analysis of what was important, which were the vital areas for each nation, came naturally to us. The thing is that this analysis and the subsequent actions of the palyers of various nations then added to the balance of the game. If I was Venice, I knew that seizing the Adriatic and, preferably, the Ionian Sea was in my best interests. So did everyone else, of course, and so the game was rebalanced in that sense.

Bias is a more systematic means of unbalancing the game. It seems to me to come in many forms, and can be really rather subtle to catch. This is further compounded by the balance of games and scenarios: if the games is so well balanced, an early marginally false step can have grave consequences later. Similarly, as the games in the Machiavelli campaign might have shown, in a well balanced game a bad (or good) dice roll can make a big difference. If the balance is so good the result can depend on a hair-trigger.

A case in point would the be heavy cavalry charges of my Italian Wars campaign. If these hit home, they are devastating. If they hesitate, and the gendarmes are counter-charged, the result is often devastating the other way. Similarly, I suspect, in World War Two games a lot depends on who reacts first. If it is you and your machine gun opens up, you win. If not, you lose. Again, this is not bias, but balance kicking in.

To start to answer JWH’s question, then, bias can be hard to distinguish from designed balance. If I favour one side by 60% to 40%, then I should see that in the game outcomes. But if my games results really depend on a single hair-trigger outcome at 50% to 50%, those results are going to be hard to distinguish unless I repeat the scenario a fair number of times. Even then, as I will have learnt about the scenario and the most optimal deployments and moves for each side, I may not be able to see if I am biased or not.

As it happens, I was concerned about bias against the French in the Machiavelli campaign. As, at present, their wargame account reads played two, won two, and they are within two turns of winning the whole campaign bias may not be a problem. However, in the wargames they have, so far, been lucky on the heavy cavalry charges. If anyone can figure out if I am biased, I would like to hear from you.