mesa-marathon.r

library(tidyverse)
library(rvest)

### EXTRACT

# marathon
url = 'https://mesamarathon.com/results?sort=&race=166724&date=&event=Marathon&gender=&division=&search=&page_167031=1&size_167031=100000&page_167067=1&size_167067=25'
page = read_html(url)
raw_marathon = page %>% html_table() %>% nth(., 2)

# half marathon
url = 'https://mesamarathon.com/results?sort=&race=166724&date=&event=Half+Marathon&gender=&division=&search=&page_167032=1&size_167032=100000&page_167068=1&size_167068=25'
page = read_html(url)
raw_half_marathon = page %>% html_table() %>% nth(., 2)

### CLEAN
hm = raw_half_marathon %>%
  rename(bib_no = `Bib #`, sex = Sex, age = Age, 
         place_overall = `Place Overall`, place_gender = `Place Gender`, place_division = `Place Div`,
         chip_time_chr = `Chip Time`, gun_time_chr = `Gun Time`, pace_chr = `Pace (min/miles)`) %>%
  select(bib_no, sex, age, place_overall, place_gender, place_division, chip_time_chr, gun_time_chr, pace_chr) %>%
  filter(chip_time_chr != '') %>% # remove runners without a finish time
  separate(chip_time_chr, sep = ':', into = c('chip_time_chr_h', 'chip_time_chr_m', 'chip_time_chr_s'), remove = FALSE) %>%
  mutate(chip_time_minutes = (as.numeric(chip_time_chr_h) * 60) + as.numeric(chip_time_chr_m) + (as.numeric(chip_time_chr_s) / 60)) %>%
  select(-chip_time_chr_h, -chip_time_chr_m, -chip_time_chr_s) %>%
  mutate(age_bucket = cut(hm$age, breaks=c(0, 20, 30, 40, 50, 60, 100), right = FALSE, labels = FALSE)) %>%
  arrange(chip_time_minutes) 

#### PLOT

# scatter
ggplot(hm, aes(x = age, y = chip_time_minutes, col = sex)) +
  geom_jitter()

# scatter by age bucket
ggplot(hm, aes(x = age, y = chip_time_minutes, col = sex)) +
  geom_jitter() +
  facet_wrap(~age_bucket, scales = 'free')

# density
hm %>%
  filter(sex %in% c('M', 'F')) %>%
  ggplot(., aes(x = chip_time_minutes, fill = sex)) +
  geom_density(alpha = .4)

# new
hm %>%
  ggplot(., aes(x = place_overall, y = chip_time_minutes)) +
  geom_line()