If you watched the batrep between myself and Andy (skew) on the youtube channel (and if you haven’t you can find it here) you might have noticed that during the post-match run down we had a pretty interesting (I thought) discussion about the effects of changing your list frequently in search of that winning formula.
Jokes aside about his misinterpreted comment that I am at an advantage by playing less (by which he meant I am less inclined to change things around), there is a pertinent question about when it is appropriate to tinker with your list and when it is better to stick with what you know.
In the post-game we talked about whether the fact that I rarely altered my list (with the guard mostly due to poor student budget constraints and the World Eaters a disinclination to buy more stuff as I hope to offload the army soon) meant that I have been forced to learn the intricacies of the army (or perhaps more relevantly, the units). For example, I have always played the death company with the exactly the same load out since the drop of 6th, firstly as guard allies and then later in the full BA list. I know exactly what to avoid, what to hit and how long it will take to chew through. If tomorrow, I were to start using Vanguard Vets I would have no idea what I was doing, probably make a hash of it and then decry them as useless and scrap them. Maybe they are useless, but more likely they are actually a very effective unit if employed properly.
Now, I should probably say at this point that I am not criticising Andy for not know how to use his list, this is a general point about the duality of effects when changing your list. Broadly speaking the concept is this: you change your army because you expect it will function better with a new build but in the same breath you take away some of that effectiveness through a lack of experience.
So for the mathematically minded of you, this is how it might work:
E(U(x)) = [E(F(x)) – F(y)] – G(s)
What this essentially means is that the expected benefit of changing from your old list ,y, to a new list ,x, (the left hand side term) is equal to the difference between how effective your old list was, (F(y)), and how effective you expect your new list to be, E(F(x)), on top of that you then have a term G(s) which represents how much you lose by being unfamiliar with the new list and notably increases with the size of the change.
Or that is the basic premise of the theory. With the assumption the effectiveness of your old list is fixed (i.e. you’ve played it enough times that you won’t get any better at playing it) and it more or less drops out of consideration.
The benefit you derive from changing your list therefore boils down to how much better you think your new list will be than your old one, minus the loss you incur by changing your list. For a given improvement in list(say 10%) it is clear than a smaller change in the composition is better, as you lose less experience. It is also obvious that a list should be changed when the benefits of changing a list outweigh the costs of doing so (in experience terms) and the opposite obviously applies.
We would also hope that the longer you spend with a list the better you get at playing with it. Therefore it is likely the loss may actually be a function of games played ,t, as well as size of change ,s, so the function might be modelled G(s,t).
Therefore as gamers, the question we should be asking (and now assuming that playing an extra game with a list will make you better with it), is whether the loss faced by changing a list is greater than the benefit of doing so. Then secondly, (assuming G is decreasing with an increase in t) whether I have played enough games, t, to remove the loss term G(s,t) so that the list is actually functioning at full effectiveness F(x) rather than the impaired effectiveness F(x) – G(s,t).
Overall, this is a very long winded way of saying something pretty simple. That is, whether you should change your list before you have worked out the best way of using it.