s2-exercises.Rmd

---
title: "Session 2 Exercises"
author: "Ben Raymond, Adrien Ickowicz"
output:
  html_document:
    toc: false
    theme: flatly
    highlight: zenburn
    css: "extra/exercises.css"
---

```{r chunkopts, eval = TRUE, echo = FALSE}
knitr::opts_chunk$set(message = FALSE, warning = FALSE)
options(width=80)
options(tibble.width = 80, tibble.print_max = 10, tibble.print_min = 10)

```

# Preparation

Load some packages:

```{r setup}
library(tidyverse)
library(datavolley)
library(ovlytics)

```

Read in our data

```{r data}
f <- dir("example_data/DE Men 2019", full.names = TRUE, pattern = "\\.dvw$")
x <- lapply(f, dv_read, skill_evaluation_decode = "german")

## concatentate the play-by-play data into one big data frame
px <- bind_rows(lapply(x, plays))

```

Check for inconsistent team names / IDs:

```{r check1}
px %>% count(team, team_id) %>% arrange(team)

x <- remap_team_names(x, remap = tibble(from = "BERLIN", to = "BERLIN RECYCLING Volleys", to_team_id = "5413"))

px <- bind_rows(lapply(x, plays))
px %>% count(team, team_id) %>% arrange(team)

check_player_names(x)

```

Add some extra columns:

```{r extracols}
px <- ov_augment_plays(px, to_add = "all")

```


# Exercises

1. What is the mean serve error rate?

2. By team?

3. By team, but only for Herrsching and Giesen? (First find their full team names in the data set)

4. By player?

5. Produce a table (data frame) of mean serve error rate and mean ace rate by team

6. Use that to plot mean serve error rate vs mean ace rate

7. What is the average sideout rate by pass quality (i.e. sideout rate on perfect passes, sideout rate on good passes, etc)

8. Calculate the first-ball attack kill rate (i.e. kill rate on attacks made on serve reception)

9. 9. Calculate the breakpoint rate (fraction of points won while serving) by rotation (use the `setter_position` column), for all teams combined


<div id="footer"></div>