From An apology to some min-maxers:
In fact, if anything, my sub-optimal character builds were me being lazy, and not doing the homework other players were doing to build more optimized characters.
At the same time, as GM you should min-max the opposition to the same level as you have allowed for your players. Dig into the system and look for how to optimize NPC’s and monsters. That is likely going to be more work for you, as you may need to take stock stat blocks and beef them up, but it will create challenging opponents for the players, making encounters more exciting.
This concisely captures why I gravitate away from games that support or require significant optimization. It’s just making more work for everyone involved, and not in a way that adds to the play experience. It’s work inflation: everyone needs to spend more time in order to get the same result. And the result is the same, assuming that competition has no value (which it does not, for me, in RPGs).
Compare this to working more on setting detail or even character personality. Such preparation can consume as much time as you want, but it can also qualitatively improve the play experience. Optimization and what I am going to call (for lack of a better term) preparation can thus be seen as two independent dimensions of out-of-game work required by RPGs, leading to four rough categories of game: high-op/high-prep, high-op/low-prep, and so forth.
A railroad is a way of constraining preparation requirements. There may be some overall conservation law operating such that sustainable games on average tend to be low in either or both dimensions. This may be why it is difficult for me to find an example of a game that is naturally high in both categories. Games associated with high optimization (Pathfinder, 4E D&D) can be played using sandboxes or with more extensive world building, but doing so ends up being a higher-maintenance activity.
As an observation it seems that many games in the focused design tradition, especially Forge and Story Games, seem to prize minimization of both kinds of work. Games like Lady Blackbird and Apocalypse World, for example, put few barriers before getting started (AW requires the MC to create some fronts, but the instructions seem geared to avoid potential scope creep). Minimizing out-of-game work a way to increase approachability by decreasing cost. If the reliability of play experience varies with the prep time, that is seen as a flaw in this tradition. The system is seen as “not working” if play is inconsistent, as it will surely be if preparation is required given that different groups or players will invest in different amounts of out-of-game work. Other traditions seem to focus more on the other side of the equation: increasing value (such as the intricate setting and art of Warhammer, or atmosphere of the World of Darkness, separate from amount of work required). Obviously, the two approaches are not exclusive, but there still seem to be trends toward one or the other.
And yeah I may have been working on a data analysis assignment before writing this post…
How does system mastery requirements fit into optimization? That is, how knowing the rules affects play when the rules have non-obvious penalties for certain things. For example, if attacked by a monster in Dungeon World, only the most-best fighter should actually attack the monster. So if my character has a +0 and someone else has a +3, my proper course of action is to stand behind the better character and do nothing (so as to not invoke a Move). My attacking will only result in a greater chance of complications and damage. Because of the way Dungeon World punishes characters for performing tasks they aren’t optimized for, the party should make sure that whenever possible, only the character best suited for a particular Move should invoke that move. FFG’s new Star Wars games are notoriously bad about this as often a non-optimized character will have a much greater chance of making something bad happen than with actually succeeding at the roll.
So, in effect, the best way to avoid optimization is to keep as much of the rules out of the players hands as possible. You can’t optimize what you don’t know.
Your analysis of DW optimization assumes a very mechanical approach to the game that is not recommended in the book.
For example, for your given situation of a monster attacking the party, the other party members can probably get a free damage roll if they attack it while he is occupied dealing with the fighter. It depends on how they describe their actions, the context of the scene and so on.
Clearly some characters are more optimized for certain actions, but DW does no focus enough in the strategic simulation and detailed mechanics to warrant much optimization. You can always choose the most damage output at each level up option, but it’s a trade-off for not having other cool abilities that would allow you to do fun stuff.
I’m not trying to convince you that DW is the best game ever and that “you’re playing/reading it wrong!”, just that it is a game where this very analytic/mechanical approach misses the point of the game even more than it does for other RPGs.
I think system mastery has a number of components. Some of them, such as, for example, knowledge of feats available and the synergies between them, feed into character optimization. Others, like the stat example you raise, seem more like tactical decisions that may be relevant irrespective of character build choices (though it will depend on the specifics of the game in question).
Regarding the Dungeon World example specifically, remember that “failing” such a combat roll is how you mark experience, so I think reasoning about what benefits a particular character in such a circumstance is not quite as simple as you example suggests. It’s not just about how to defeat enemies while while minimizing damage taken, at least based on my memory of the rules (it’s been a while since I have read them).