SABR
GAMES and SIMULATIONS
COMMITTEE

Evaluating Realism in Games and Simulations

I’d like to offer a few thoughts in response to the request made on the blog for input about realism in games/simulations.  Mainly, I’d like to suggest that evaluating the realism of a game really hinges on how you choose to answer some key questions about what you mean by realism.  I’d propose the following three such questions (or, perhaps more accurately, categories of questions):
(1) What’s luck, what’s skill, and how do you think it should matter in game design?
Clutch hitting is a notable example here, but there are lots of others.   If you think clutch hitting is not really a skill, do you add to realism or take away from it by designing a game to make some hitters especially effective in clutch situations?
BABIP is another example, one that is likely more far-reaching.  If you think BABIP is mainly a matter of luck, do you want a game to be built on this by, in effect, assigning the same BABIP to all pitchers and batters? If you do, it will make some players quite a bit more or less valuable in the game than they appear to be in real-life.
(2) How big a sample size is big enough?
If a player had only a handful of PAs but an outstanding OPS, do you want a game to rate him so that he’d likely have an OPS similar to what he achieved in real-life, likely with his number of PAs very limited?  If not, presumably you’d want to gradually move toward having the game replicate player performance as the number of real-life PAs increased, but how quickly?
Other questions relating to sample size may be even more difficult.  What about a batter with a fairly large number of PAs and generally little success but with a strong performance in the small percentage of his PAs vs. LH pitchers?  Would you be adding or taking away from realism by trying to have a game reflect this large difference in performance vs. RHPs and LHPs?  Or perhaps you’d regard the PAs vs. LHPs to be too small a sample?
It could be argued that even a full season for a full-time player might not really be a large enough sample size for purposes of trying to be as realistic as possible.  Consider a player who was typically somewhat of an iron man but, during one season in the middle of a long career, did miss half a season due to injury.  Is a game’s realism enhanced by treating this injury as something unlikely to happen if you replay this season?  Similarly, what about a player who had one splendid season in the course of an otherwise undistinguished career?
(3) How do you deal with real-life managerial decisions that look bad in hindsight?
Perhaps the most straightforward example here is a player who had, say, a dozen steal attempts over the course of a season but was thrown out every time.  A game could take at least 3 different approaches.  First, it could rate him so that he’d likely be thrown out nearly every time in the unlikely event that a manager chooses to send him as often as he was sent in real life.  Second, it could rate him a bit better than this with a view to encouraging the manager to send him occasionally.  Third, it could (at least partially) take the decision away from the manager by forcing the runner to attempt steals from time to time against the manager’s wishes.  All of these could be argued to best reflect realism.
However you answer the above questions, it should help you down the path of determining what game you think is most realistic.  Or, perhaps, point the way to identifying what features a new game should have to surpass those already on the market.
Hope this helps in moving the conversation forward.
Frank McDonald likes board games in general as well as baseball, and he likes to see the two of them get along well together.
Website | + posts

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ac turpis egestas sed tempus urna et. Cras pulvinar mattis nunc sed. Ipsum a arcu cursus vitae congue mauris rhoncus. Tincidunt lobortis feugiat vivamus at augue. Sit amet consectetur adipiscing elit duis tristique sollicitudin. Massa sapien faucibus et molestie ac feugiat sed lectus vestibulum. Duis at consectetur lorem donec massa sapien. Duis at tellus at urna condimentum mattis pellentesque. Id donec ultrices tincidunt arcu non sodales neque sodales. Id aliquet lectus proin nibh nisl. Rhoncus mattis rhoncus urna neque viverra.

0 0 votes
Article Rating
Subscribe
Notify of
guest

0 Comments
Inline Feedbacks
View all comments