Rating: 7.7/10.
Optimizing Play: Why Theorycrafting Breaks Games and How to Fix It by Christopher A. Paul
Book that studies the design of video games and sports, and how players’ attempts at theorycrafting or optimizing the game to win more often lead to negative consequences for the enjoyment of the game. There is often a conflict of interest where players are driven to play “optimal” strategies to win more, but if everybody does this, the game is no longer fun to play or watch. Therefore, developers need to make changes to avoid a stale meta or theorycrafting, or somehow encourage social norms to discourage winning at any expense. In video games, theorycrafting is generally done by a relatively small number of players who post on community forums, and the rest of the player base copies these strategies. This reduces the game from an exploratory play to a series of mechanical steps simply to be executed better than the opponent. This is an especially important consideration in multiplayer games, where there is often strong social pressure to stick to the optimal strategy.
The first example is from Overwatch, a colorful team-based shooter game by Blizzard, where the meta evolved from an exciting dive meta to a boring meta involving tanks and healers that was tedious to watch. Blizzard tried to change this by buffing and nerfing the stats, but this was not very successful. Eventually, they changed the character selection process so that this team composition was impossible, and they also added new characters more frequently so that the meta would not become stale. Other considerations include the need to keep professional play not too different from casual play, to prevent a meta at professional levels from becoming too complex to explain and understand, and to ensure the meta does not change too quickly at professional levels, as this would make the results too random.
In professional sports, baseball was the first major sport to be carefully analyzed. It is generally easier than other sports to analyze because it is based on discrete events. However, teams optimizing for wins produced less watchable content since they tried to go for the three outcomes with just the pitcher and batter, where the defense plays no role. They discovered that swapping out the pitcher many times in a game was more optimal than using a single star pitcher and also less expensive since it avoids developing an expensive star player, but this is less fun to watch.
Other sports have not been affected as much by analytics as baseball, such as football and soccer, which are more fluid and difficult to capture into structured data. In football, there are not enough games to even out the variance of a slightly more optimal approach, and the community favors how the game is meant to be played rather than deviating from it in an unusual way. Basketball meta evolved to favor three-point shots instead of mid-range two-pointers. Other strategies that evolved include intentionally losing games to get the next season’s top draft picks or intentionally fouling to stop play – these are unexciting for the game, so it led to rule changes to prevent abuse. However, the three-point meta was kept because it was exciting to watch players like Steph Curry execute it successfully. The league must prevent behaviors like intentionally fouling since if the game becomes unappealing to watch, viewers will turn to something else, and everybody loses.
Several esports games are based on real sports like soccer and basketball and feature real players with realistic stats, but they often end up with a meta that looks nothing like the real game. A few steps become critical as gamers discover overpowered strategies that cannot work in real life, like optimizing the animation to shoot from very far and developing speed to run around other players with one speedy player. Additionally, the microtransaction system discourages exploration and deviating from the optimal build, as doing so will result in losing all of your games when playing online multiplayer.
An example of a game that has avoided a stale meta is Clash Royale. This is a mobile game where players build a deck with eight cards and enter them into a MOBA arena. It has prevented a stale meta by constantly revising the stats of cards that are too high or too low every few weeks. The relatively short matches and randomization mechanics mean that many setups are viable, and the optimal ones are not significantly better than an average build. This is especially true since players cannot easily switch to a different deck that is optimal at that time because it takes time for them to learn to play it properly.
Overall, strategies that developers can use to combat a stale meta include randomization elements and banning certain characters from play. This problem can be seen as a common resource problem, so it affects multiplayer online games where individuals optimize to win, taking away from the pool of viable strategies for other players. However, it is not a problem for single-player games where you are not affected by other players’ efforts at theorycrafting.
