Animating Tracking Data

For the following vignette, you’ll need the sportyR, ggplot2, and gganimate packages loaded into your workspace.

library(sportyR)
library(ggplot2)
library(gganimate)

If this is your first experience with plotting tracking data, please check out the plotting-tracking-data vignette. Otherwise, let’s see how to make GIFs with sportyR and gganimate.

The Data

For this example, we’ll use a play from Week 15 of the 2018 NFL season between the Chicago Bears and Green Bay Packers. Data made available for the Big Data Bowl 2021 Kaggle competition.

# Load the play data
example_nfl_play <- data.table::fread(
  glue::glue(
    "https://raw.githubusercontent.com/sportsdataverse/sportyR/",
    "main/data-raw/example-pbp-data.csv"
  )
)

# Convert to data frame
example_nfl_play <- as.data.frame(example_nfl_play)

To keep things easy, let’s specify the colors for each team’s dots on the resulting GIF. We’ll make the Bears orange and the Packers yellow. The football will also need a dot to be seen; let’s make it brown.

# Prep data for plotting
example_nfl_play[example_nfl_play["team"] == "home", "color"] <- "#c83803"
example_nfl_play[example_nfl_play["team"] == "away", "color"] <- "#ffb612"
example_nfl_play[example_nfl_play["team"] == "football", "color"] <- "#624a2e"

First, let’s draw an NFL field via geom_football("nfl"). We’ll adjust the origin to be in the lower left corner of the field, as per the notes on the coordinate system on the Kaggle page describing the data.

# Create the field
nfl_field <- geom_football("nfl", x_trans = 60, y_trans = 26.6667)

# Display the field
nfl_field
Image of NFL field rendered from sportyR
Image of NFL field rendered from sportyR

Looks good! Now, let’s animate using gganimate.

# Add the points on the field
play_anim <- nfl_field +
  geom_point(
    data = example_nfl_play,
    aes(x, y),
    color = example_nfl_play$color
  ) +
  transition_time(example_nfl_play$frameId)

# Show the animation
play_anim
Gif of tracking data
Gif of tracking data

Easy peasy. As noted on the plotting-tracking-data vignette, this too works so long as the geospatial data is provided and contains a way to identify and order the frames of the resulting GIF.