Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

https://github.com/bdilday/GeomMLBStadiums

Geoms to draw MLB stadiums in ggplot2
https://github.com/bdilday/GeomMLBStadiums

baseball ggplot2-geoms r

Last synced: 3 months ago
JSON representation

Geoms to draw MLB stadiums in ggplot2

Lists

README 

        
---

output: github_document

---



```{r, echo = FALSE}

knitr::opts_chunk$set(

  collapse = TRUE,

  comment = "#>",

#  fig.path = "README-",

  fig.height = 6,

  fig.width = 6,

  fig.units = "in"

)

```



# GeomMLBStadiums



This package defines a couple of Geoms to draw MLB stadiums in ggplot2. It also provides a Geom to draw a "spraychart" - `x` and `y` locations of batted balls with a stadium overlay. 



## Example use



### Install from github and load the necessary libraries



``` {r echo=FALSE, message=FALSE, warning=FALSE}

library(GeomMLBStadiums)

library(ggplot2)

library(dplyr)

```



``` {r eval=FALSE, message=FALSE}

devtools::install_github("bdilday/GeomMLBStadiums")

library(GeomMLBStadiums)

library(ggplot2)

library(dplyr)

```



### The stadium data



When you load the `GeomMLBStadiums` package it will attach the stadium paths as a data frame, `MLBStadiumsPathData`



``` {r}

head(MLBStadiumsPathData)

```



The data comprise the 30 current MLB stadiums, in addition to a "generic" stadium. The stadia are identified by team name, with the following conventions



``` {r}

unique(MLBStadiumsPathData$team)

```



The segments are split up into `outfield_outer`, `outfield_inner`, `infield_inner`, `infield_outer`, `foul_lines`, and `home_plate`



``` {r}

unique(MLBStadiumsPathData$segment)

```



### Coordinates



The stadium paths are in the system of the `hc_x` and `hc_y` coordinates of MLBAM. These are inverted (because they're based on a display device where `y=0` is at top, IIRC) which means by default the field gets displayed upside down. This package provides a helper function, `mlbam_xy_transformation`, that transforms these values to a system where y increases from bottom to top and home plate is located at `(0, 0)`.



``` {r}

set.seed(101)

batted_ball_data = data.frame(hc_x = rnorm(20, 125, 10), 

                              hc_y = rnorm(20, 100, 20))



head(batted_ball_data)



head(mlbam_xy_transformation(batted_ball_data))



summary(mlbam_xy_transformation(batted_ball_data))



```



### `geom_mlb_stadium`



This uses `geom_mlb_stadium`, which implicitly loads the `MLBStadiumsPathData` data, to plot the 30 current stadiums.



``` {r}

ggplot() + 

  geom_mlb_stadium(stadium_ids = "all_mlb", 

                   stadium_segments = "all") + 

  facet_wrap(~team) + 

  coord_fixed() + 

  theme_void()

```



An alternative way is to explicitly pass the data to `geom_path`.



``` {r}

MLBStadiumsPathData %>% 

  filter(team != 'generic') %>% 

  mutate(g=paste(team, segment, sep="_")) %>% 

  ggplot(aes(x, y)) + 

  geom_path(aes(group=g)) + 

  facet_wrap(~team) + 

  coord_fixed() + 

  theme_void()

```



This shows the generic stadium, which is the default,



``` {r}

ggplot() + 

  geom_mlb_stadium(stadium_segments = "all") + 

  facet_wrap(~team) + 

  coord_fixed() + 

  theme_void()

```



### `geom_spraychart`



This generates some simulated data.



``` {r}

# first generate the data

set.seed(101)

batted_ball_data = data.frame(hc_x = rnorm(20, 125, 10),

                              hc_y = rnorm(20, 100, 20))

batted_ball_data$team = rep(c("angels", "yankees"), each=10)

```



This plots the data as a spraychart. By default it uses the "generic" stadium.



``` {r}

batted_ball_data %>% 

  ggplot(aes(x=hc_x, y=hc_y)) + 

  geom_spraychart() 

```



Add some styling using `theme_void` and `coord_fixed`



``` {r}

batted_ball_data %>% 

  ggplot(aes(x=hc_x, y=hc_y)) + 

  geom_spraychart() + 

  theme_void() + 

  coord_fixed()

```



This transforms the data and the stadium before plotting, passes the team names in `stadium_ids`, draws all segments, and facets by field.



``` {r}

batted_ball_data %>% mlbam_xy_transformation() %>%  

    ggplot(aes(x=hc_x_, y=hc_y_, color=team)) + 

    geom_spraychart(stadium_ids = unique(batted_ball_data$team),

                    stadium_transform_coords = TRUE, 

                    stadium_segments = "all") + 

    theme_void() + 

    coord_fixed() + 

    facet_wrap(~team) + 

    theme(legend.position = "bottom")

```



You can make use of any of the other `ggplot2` functions, for example, contours from `stat_density2d`. The `mapping` argument for `geom_spraychart` gets passed to the underlying `geom_point`, as do any extra parameters passed into the `...` argument of `geom_spraychart`, e.g. `size=5` in the below.



``` {r}

batted_ball_data %>% mlbam_xy_transformation() %>%  

  ggplot(aes(x=hc_x_, y=hc_y_, color=team)) + 

  geom_spraychart(mapping = aes(shape=team), 

                  stadium_ids = unique(batted_ball_data$team),

                  stadium_transform_coords = TRUE, 

                  stadium_segments = "all", size=5) + 

  theme_void() + 

  coord_fixed() + 

  facet_wrap(~team) + 

  theme(legend.position = "bottom") + 

  stat_density2d(color='gray')

```