Lately it seems that everyone is talking about composition of armies and composition systems at tournaments. And there are an incredible amount of different views on the subject. The two things that are clear from these discussions are that firstly most people believe that some form of composition scoring is required at tournaments and secondly that people's ideas on what constitutes a well composed army is highly subjective as are the methods to measure it.

If we step back and take a look at the concept of composition and the need for it at tournaments we can conclude that due to the amount of variables in the game of Warhammer there are inherent flaws in army books. These flaws allow combinations of troops and magical items that can give one player what is perceived to be an unfair advantage over another in a given game. Games Workshop do receive a fair amount of criticism for their army book design due to the imbalance that some combinations of troops present. However the same people that criticise for this fact are the also the ones that cry foul when troop types are removed from a latest release and/or certain rules are "dumbed down", both of which are normally motivated by the need to remove imbalances.

We all love the fact that Warhammer is such an incredibly complex game with a huge amount of scope for individuality in army design. Having this variety is what makes it so hard to not introduce flaws in army books. I think we just have to accept that the possibility of over the top armies goes hand in hand with the game's mechanics and the elements of the game that we enjoy so much. We cannot necessarily have one without the other.

Which brings us back to the need for composition scoring in a competitive environment such as tournaments. The dilemma is devising a composition scoring system that allows players to express their creative side with army design, therefore ensuring variety, whilst still curbing armies that remove the enjoyment from a game of Warhammer for whatever reason.

I am not going to go into the various systems here suffice to say that I am a believer in peer judged composition. Many people opposed to peer judged composition systems point to the fact that players have their individual subjective views on what constitutes a well designed army. I don't refute that but in the end if it is not the players themselves that determine what is a well composed army, then who is? Just because you might not agree with a player's score, does that make it wrong? I think not. If a player judges an army more harsh than others or similarly more lenient for whatever their reasons are, then that is what the composition of the army is.

The main problem that I see exists in peer judged composition is that players have a different way of interpreting the scores. For example, take a sliding scale of 1 - 5. Some players might give out 5s for a standard list and punish "harder" lists by scoring them lower. Where others might award 3s for a standard list and judge "harder' or "softer" but scoring above or below this point. Therefore the two players are awarding different scores to armies that they perceive to be "normal". We don't care whether they are awarding a "3" or  "5", we care whether they feel the army is acceptable or not. And this is the problem.

Now there a very good method that overcomes the problem detailed above with peer based composition systems. And it is not new to the Australian tournament scene but somehow it failed to stick around after a very successful implementation a few years ago. The basic idea is to calculate what the average score each player hands out over the course of the tournament, and then award or deduct points to armies they have faced based on the deviation from this average. So lets go back to our example above.

Player A awards 5 scores of 5 and 1 score of 3. Therefore the average score they have handed out is 4.67. Therefore each player that received a 5 from this person will get +0.33 to their composition score. And the player whose army was awarded a 3 will receive -1.67 to their comp score.

Player B awards 4 scores of 3, 1 of 4 and 1 of 1. Therefore their average score they have handed out is 2.83. The 4 players that received a score of 3 will receive +0.17 to their composition score, the player that received a 4 will get +1.17 and finally the player that received a 1 will get -1.83.

In both cases above, it made no difference what number each player selected, it only mattered how different this was to the average score they gave out. I have played in a few tournaments years ago that used this system and it was extremely effective at awarding composition scores to the various armies.

Well that is my take on composition, and next year at Fortunes of War I intend to implement it. Lets see how that goes.

Andrew Galea