I recently found an article entitled Mastery vs. Hoping through Cardboard Edison. I think that there are some interesting ideas in there to think about, though there are places that I disagree with it. For the folks who are allergic to clicking links it talks about the Skinner box style gambling machines and draws a parallel between those and various computer and board games. The notion is that some games enjoy a high replayability because players get hooked on the notion of playing them over and over waiting for the perfect lucky combination while believing that the outcome is impacted by their decisions.
I'd come across these ideas in studying psychology and been introduced to its application in computer games by Extra Credits, but hadn't thought about it in the context of board games before. The 'Mastery vs Hoping' article is not really about board games (though it mentions Dominion), but having come to it through a board game design resource it drove me to think about games in this way with some interesting results.
The notion is that when a player performs some simple action and is periodically rewarded, the player will persist in taking this action. The subjective experience of this is hard to describe, the player may well play on long past the point that they're actually enjoying themselves, but it continues to feel rewarding on some level. This is most effective when the reward schedule (how often the action is rewarded) is a little inconsistent. If every enemy drops the '+1 more than last time sword' players stop playing more quickly than if only some smaller fraction of them do.
For some business models this is a very powerful force in computer games. If a game relies on subscriptions or microtransactions then there's a large benefit to having people putting countless hours into the system. In board games none of these models apply, but I think that the design and testing process could easily lead to an accidental implementation of some conditioning ideas. One of the things I use as a metric of success is how often a playtester says something like "Can we play again?" or "When's the next test?" I know that I'm not alone in this and I can see how relying too heavily on that sort of feedback to guide the direction of a design could lead to creating a skinner box game that encourages players to play over and over without having much fun.
Going back to the 'Mastery vs. Hoping' article, I think that my main point of disagreement is the false dichotomy. Running with the Dominion example, the game clearly relies on a degree of mastery. Having played a few (hundred) games on isotropic, it was consistently more difficult to beat higher ranked opponents than lower ranked opponents. If the game was based principally on luck rather than mastery, I would have expected to see no relationship between previous and future performance, as this did not happen I can reasonably conclude that a player's skill has a large impact on how well they play this game. On the other hand, there is undoubtedly a gambling machine style pull towards drawing a new hand, because maybe this time everything will come off perfectly and you'll get over 100VP in a turn.
I think that this may go a long way towards explaining the success of building style games as a genre. Whether you're talking about a deckbuilder, a wargame in which you've pre-constructed an army or watching your lovingly crafted baby explode they've got a shared theme. There is a fair amount of skill in the construction part of the game, that is somewhat inconsistently rewarded once it's built, creating the 'one more turn' effect.
Outside of games with a stark contrast between a deterministic construction and uncertain resolution the mix of skill and luck can be observed. The MvH article uses RPGs as an example, citing the impact of a hit percentage chance in this regard. There's an attempt to dismiss the argument that these chances average out over time by noting that the value of two particular checks are not always equal. Colloquially, hitting with the dagger does not make up for missing with the orbital hyper death ray. It's a fair comment, but it only changes the scope of the argument.
I've written before about how people's perceptions of probability tend to be flawed, it's easy to see a single lucky or unlucky randomiser that appears to have decided the game and to conclude that the game is essentially random. This is almost never true, for three reasons. The first is that randomisers, at critical moments, will tend to balance out over time. The rarer a critical moment is, the longer the time period, but the statement remains true. The second is that in a lot of games players create the critical moments, by sound strategy putting them in a position where any one of several checks being lucky will produce a win. The third is that there are very often more critical moments in a game than players give it credit for. If an RPG fight comes down to two opponents, each one hit off death and each with a 95% chance to hit and then the first combatant misses it's easy to identify that as a critical moment in which an unlucky result changed everything. The thing is that every other strike in that fight was equally critical. If any one of the other to-hit rolls in the whole thing had been changed then either one of the combatants would already have been dead before the supposed critical juncture or one would have had two hits left and the miss would have been irrelevant.
I think that the features that the author of the MvS identifies are better understood along three axes. A 'luck vs. skill' axis that defines the extent to which victory in a game is determined by player choices as opposed to random factors. A 'random vs. fixed reinforcement axis' which defines the how much noise there is in determining when (either luck or skill) is rewarded . Finally a 'percieved luck vs. skill' axis, that determines how players experience the luck vs. skill axis, some games are extremely transparent while others disguise luck as skill or vice-versa.
Under this system Dominion might involve a lot more skill than luck, but reinforce play by having the skilled decisions rewarded inconsistently and be perceived to have a higher luck content than it really does. Is that a more fine-grained distinction? Is it more useful?
At the very least I'm finding it interesting to think about boardgames in terms of how they handle reinforcement schedules and whether what's psychologically effective is what will make the best game or not
A collection of posts by game designer Gregory Carslaw, including mirrors of all of his blogs maintained for particular projects. A complete index of posts can be found here: https://boardgamegeek.com/blogpost/58777/index
24 Jul 2013
- [+] Dice rolls