In my last entry, I briefly covered three forms of logic: Deduction, Induction and Abduction and how they relate to how we discuss Warmachine.
This next three articles form a bit of a trilogy about how to apply various forms of logic. In my first two articles, I discuss the differences in building warmachine lists based on a theory (Deduction) and building a list based on experience (Induction). In the final installment, I will explore on the ways that Abduction can be used to make a list based on your best guess plans of the meta.
It’s important to stress at this point that there is fundamentally no wrong way to go about building a list. However, it is possible build a list with an incomplete or incorrect understanding of number of factors including but not limited to: List Synergy, Faction/Caster Match-ups and Skews/Counter-Skews. For this reason in this set of articles I am encouraging you to think about process over product.
Taking the time to identify which of the three processes you already utilize, should help you improve your current approach or let you swap into a new one. This set of articles is not an instruction manual to build an über list. I do hope that by breaking down the steps you can take to build a list deductively or inductively, you will find your game play improving.
The Deductive List Building process
With Deductive List Building, you start with an idea of what your list should be doing. This will referenced as your Theory. In normal forum and list building discussions a deductive statement about a list is “I built this to be my armor cracking list” or “I built this to be my anti-infantry list.”
- P→Q (conditional statement) P Doing a large amount damage is needed to defeat High Armor
- P (hypothesis stated) P Bane Units do large amounts of Damage (and I’m taking them)
- Q (conclusion deduced) Q I should be able to defeat high armor.
(Thanks to dchadpage for the spot check)
The reason why the starting point is so important is that it sets the tone for the rest of your list. Once you have a theory, then progressing on to a hypothesis, then to observations and finally confirmation or rejection (see the image below).
Enter the Dojo: Starting with a Theory
First to definitions, what is a Theory? A Theory is best defined as a supposition or a system of ideas intended to explain something, esp. one based on general principles independent of the thing to be explained (1). Yet, you might be asking yourself, besides being an “armor cracking list” or an “infantry sweeping list” what does a theory look like in Warmachine?
If you’re a regular listener of the Muse on Minis (MoM) cast, you will be familiar with this as it often gets repeated by the MoM crew, so I’ll call it Muse’s Law (2):
All good lists are able to either:
1) Ask questions than your opponent’s list can not answer
2) Answer the important questions being asked by your opponent’s list.
At this time, for the sake of brevity, I won’t explore the DeMaris Theorem (It’s the player, not the faction), Christensen’s Axiom (I only play broken stuff), Crump’s Legit Corollary (If you have to ask, you don’t get it), the Chadwick Postulate (I don’t like it when people touch my shit), or Collin’s Lemma (Drakes!) and the Maxon Footnote (Calamity is only half as good as Crippling Grasp).
Nor is the MoM cast is the only place you can find a list building theory as Nigel’s Dojo Discussion are also quite excellent. However due to John’s influence, often MoM is the best place to hear nuggets of theory gold.
Trying to make sense of the Internets: What’s Good and What’s Not?
To build a list deductively, we rely almost entirely on the parameters and assumptions laid out by the Theory. This is where we do a review of the literature and what is the literature for Warmachine? It’s the various forums, podcasts, Facebook groups and your gaming night bull sessions with your buddies.
Why is this an important step in the list building process? This is because every faction will have different tools to answer questions and different tools to ask questions. If you are a veteran player, this step will likely seem simple to you.
However for new players, this is the part that is difficult to work through Deductively. This is because its almost impossible to build a Khador list that makes the same assumptions as a Cygnar list or Cryx list that works like a Menoth list. While there certainly are models and tactics that let lists in Cygnar act like Menoth or let Khador act like Cyrx but the bottom line is that you will not find 1:1 transfers. A Defender is not a Reckoner and Doom Reavers are not Bane Thralls. The same also holds true in reverse.
So its important to ask people what makes a good armor cracking list and what makes a good infantry clearing list.
However, not everything you read on forums, hear on podcasts, or learn from your gaming group will always be useful and many times what sounds like rock-solid gospel truth is only a well-meaning, well-informed opinion. Sometimes, many times, you will hear countervailing opinions from two somewhat reliable sources.
This is where you, the player, have to make evaluative judgments, namely do you trust the source? Do others support the claims being made?
However, if you find you self getting lost in all the noise, don’t get discouraged! You won’t be perfect the first time and likely you will never be perfect. The joy of the list building process is that you don’t have to have a definitively 100% correct answer. Just one that makes a good approximation.
Putting Models on the Table and letting the dice fly: Testing your Hypothesizing
The next step is to test how our lists to know if they fulfill Muse’s Law. Below is the most fundamental way to deductively test your lists adherence to Muse’s Law:
Ho: My list does not significantly ask or Answer questions
H1a: My list asks a significant question(s)
H1b: My List definitively answers one or more questions
Ho represents the Null Hypothesis which assumes that your list is in fact not going fulfill Muse’s Law and thus not be good. Additionally, the null Hypothesis is always assumed to be true unless you can gather enough data to reject it in favor of your test hypotheses (H1a and H1b ). However because you are a clever strategist you are attempting to build a list that is in fact good. Thus you are looking to demonstrate that your list actually accomplishes H1a or H1b .
The next step in the process is to identify which models for your faction can actually ask questions and/or answer questions, this is called operationalization. The reason we operationalize is to better understand the difference between a question asking model and a question answering model.
As an example, Arcane Tempest Gun Mages can be viewed question answering models because of the utility the models posses. In contrast, a Bronzeback can be seen as a question asking piece because if you can’t prevent it from reaching your Stormwall, you’re going to have a bad time.
The final step in this process is the best part: testing your list, which is of course to play it! During the course of your playing the list, be sure to make observations about the various models on the table. You should be aware of how they performing both internally in terms of synergies with other models as well as externally in how they ask or answer questions.
If you are able to make good observations about your lists, you should be able to identify how the various part of your lists are contributing to your attempt to asking questions, answering questions. In doing so you will be able to both build better lists as well as execute your list better on the table.
Additionally, I cannot stress how important making good observations are to determining whether you can reject the null hypothesis in favor of your test hypothesis. If you make few to zero observations during the course of your play there is no way to support your test hypothesis thus you will be unable to reject the Null Hypothesis.
In other words, even if you are winning you may not be able to verifiable demonstrate you have a good list. Now some of you might say if I’m winning then my list must be good because then why would it win? Though it is possible that since your list is winning it is good or it’s also possible that you’re suffering from undersampling and overgeneralization (inductive error).
However, Inductive processes take a different approach to list building and even if you’re making an inductive error, in the next column I introduce you to ways to avoid them.