#MLB Showdown
"MLB Showdown, meet Sabermetrics"
This project makes MLB Showdown cards that are realistic representations of players. Season metrics are transformed into statistically accurate cards that play as true-to-life as possible.
Status
- Card charts (distributing the 1-20 values): Implemented, but still under development.
- Points: TBD
- Speed: TBD
- Fielding: TBD
- IP: TBD
Usage
php formula.php u p db Rpath [ dist [ num_opp ] ]
u = MySQL username
p = MySQL password
db = database name
Rpath = path to R executable
dist = Method to generate random opponents. Default (and only recommended option) is 'discrete'. Options:'discrete','continuous','single'.
num_opp = Number of opponents of each kind to generate. Default is 200. If 'single' is chosen num_opp is always 1.
- Data is found here, specifically
cards-tables.sql
for card data and[yyyy]_all.sql
for MLB season data. - R and required libraries (truncnorm, rjson, RMySQL) used in the default case. If
single
is used fornum_opp
, R is not needed so pass a dummy argument. In this case a distribution of opponents won't be available, as there is only one defined opponent card representing a 'standard' batter/pitcher.
Premises
- Game design is enhanced with increasingly-lifelike players.
- Statistical accuracy leads to innovative card sets.
- There is an elegant beauty in producing cards from a well-designed algorithm.
- One of the most difficult concepts in chart creation is 'cutting small differences'. Many cards look very close. The differences should be defined by a technical process rather than the human eye and gut feelings.
Related
Developed alongside this project are a card data visualization and a game statistics simulator. Additionally, card data was compiled in an easy-to-use format.
About MLB Showdown
Play baseball with trading cards, where 20 sided dice are rolled to determine the result of an at-bat. See the rules of the game. Example pitcher and position player cards.