World record runing events

analysis.R

library(ggplot2)
dat <- read.csv("world records.csv", stringsAsFactors = FALSE)

# Clean and transform the data --------------------------------------------

track <- within(dat, {
  Time <- as.numeric(Time.in.hours)
  Date <- as.Date(Date, format = "%d-%b-%y")
  Speed <- Distance / Time
  logDistance <- log10(Distance)
  Standard.Event <- Standard.event == " Yes "
  Time.in.hours <- NULL
  Standard.event <- NULL
})


# Create some plots -------------------------------------------------------

ggplot(aes(y = Speed, x = Distance), data = track) + 
  geom_point() +
  theme_bw(16) + 
  ggtitle("World record speed for mens running events")



labelx <- c(0.1, 0.2, 0.8, 1.6, 3, 10, 21.1, 42.2)
labely <- track[track$Distance %in% labelx, "Speed"]
labeltext <- c("100m", "200m", "800m", "Mile", "3,000m", "10,000m", "Half-marathon", "Marathon")
labels1 <- data.frame(Distance = labelx, Speed = labely, Text = labeltext)

labelx <- c(50, 100, 160.9)
labely <- track[track$Distance %in% labelx & track$Record == "World best", "Speed"]
labeltext <- c("50km", "100km", "100mile")
labels2 <- data.frame(Distance = labelx, Speed = labely, Text = labeltext)


ggplot(aes(y = Speed, x = Distance), data = track) + 
  # geom_path(data = data.frame(Distance = newx, Speed = newy), colour = "grey 20") +
  geom_point(aes(colour = Standard.Event), size = 3) + 
  geom_text(data = labels1, aes(label = Text), hjust = -0.2, vjust = -0.2) +
  geom_text(data = labels2, aes(label = Text), hjust = 1.2, vjust = -0) +
  scale_x_log10() + 
  theme_bw(16) + 
  ggtitle("World record speed for mens running events")



# Model the data using a segmented linear regression ----------------------

library(segmented)

# Helper function to compute the inverse of log10(x)
exp10 <- function(x)10^x

# Limit model data to distance from 200m to marathon
modeldata  <- track[track$Distance >= 0.2 & track$Distance <= 42.2, ]

# Fit linear model
lfit <- lm(Speed ~ logDistance, data = modeldata)

# Fit segmented model
sfit <- segmented(lfit, seg.Z = ~ logDistance)
exp10(sfit$psi)
summary(sfit)

bpoints <- 10^(sfit$psi)[, 2]
newx <- c(0.2, bpoints, 50)
newy <- predict(sfit, data.frame(logDistance = log10(newx)))


# Create plot with regression lines ---------------------------------------

labelx <- c(0.1, 0.2, 0.8, 1.6, 3, 10, 21.1, 42.2)
labely <- track[track$Distance %in% labelx, "Speed"]
labeltext <- c("100m", "200m", "800m", "Mile", "3,000m", "10,000m", "Half-marathon", "Marathon")
labels1 <- data.frame(Distance = labelx, Speed = labely, Text = labeltext)

labelx <- c(50, 100, 160.9)
labely <- track[track$Distance %in% labelx & track$Record == "World best", "Speed"]
labeltext <- c("50km", "100km", "100mile")
labels2 <- data.frame(Distance = labelx, Speed = labely, Text = labeltext)


ggplot(aes(y = Speed, x = Distance), data = track) + 
  geom_path(data = data.frame(Distance = newx, Speed = newy), colour = "grey 50", size = 2) +
  geom_point(aes(colour = Standard.Event), size = 3) + 
  geom_text(data = labels1, aes(label = Text), hjust = -0.2, vjust = -0.2) +
  geom_text(data = labels2, aes(label = Text), hjust = 1.2, vjust = -0) +
  scale_x_log10() + 
  theme_bw(16) + 
  ggtitle("World record speed for mens running events")

data.R

dat <- structure(list(Distance = c(0.1, 0.2, 0.4, 0.8, 1, 1.5, 1.6, 
2, 3, 5, 10, 15, 20, 21.1, 25, 30, 42.2, 100, 50, 100, 160.9), 
    Time = c("00:00:10", "00:00:19", "00:00:43", "00:01:41", 
    "00:02:12", "00:03:26", "00:03:43", "00:04:45", "00:07:21", 
    "00:12:37", "00:26:18", "00:41:13", "00:56:26", "00:58:23", 
    "01:12:25", "01:26:47", "02:02:57", "06:13:33", "02:48:06", 
    "06:10:10", "11:28:03"), Athlete = c("Usain Bolt", "Usain Bolt", 
    "Michael Johnson", "David Rudisha", "Noah Ngeny", "Hicham El Guerrouj", 
    "Hicham El Guerrouj", "Hicham El Guerrouj", "Daniel Komen", 
    "Kenenisa Bekele", "Kenenisa Bekele", "Leonard Patrick Komon", 
    "Haile Gebrselassie", "Zersenay Tadese", "Moses Mosop", "Moses Mosop", 
    "Dennis Kipruto Kimetto", "Takahiro Sunada", " Jeff Norman (GBR)", 
    " Donald Ritchie (GBR)", " Oleg Kharitonov (RUS)"), Date = c("16-Aug-09", 
    "20-Aug-09", "26-Aug-99", "09-Aug-12", "05-Sep-99", "14-Jul-98", 
    "07-Jul-99", "07-Sep-99", "01-Sep-96", "31-May-04", "26-Aug-05", 
    "26-Sep-10", "27-Jun-07", "21-Mar-10", "03-Jun-11", "03-Jun-11", 
    "28-Sep-14", "21-Jun-98", "07-Jun-80", "28-Oct-78", "20-Oct-02"
    ), Record = c("World record", "World record", "World record", 
    "World record", "World record", "World record", "World record", 
    "World record", "World record", "World record", "World record", 
    "World record", "World record", "World record", "World record", 
    "World record", "World record", "World record", "World best", 
    "World best", "World best"), Time.in.hours = c(0.0027, 0.0053, 
    0.012, 0.028, 0.0367, 0.0572, 0.062, 0.0791, 0.1224, 0.2104, 
    0.4382, 0.6869, 0.9406, 0.9731, 1.2071, 1.4465, 2.0492, 6.2258, 
    2.8017, 6.1694, 11.4675), Standard.event = c(" Yes ", " Yes ", 
    " Yes ", " Yes ", " Yes ", " No ", " Yes ", " No ", " Yes ", 
    " Yes ", " Yes ", " No ", " No ", " Yes ", " No ", " No ", 
    " Yes ", " No ", " No ", " No ", " No ")), .Names = c("Distance", 
"Time", "Athlete", "Date", "Record", "Time.in.hours", "Standard.event"
), class = "data.frame", row.names = c(NA, -21L))

Hi Andrie, I'm curious about the source of the data that you entered. From where can I get the latest data? Kindly can you help?

Wikipedia

There’s no API or data set? From wikipedia the analytics data won’t be correct?

On Sun, Jun 9, 2019 at 4:37 PM Andrie de Vries ***@***.***> wrote: Wikipedia — You are receiving this because you commented. Reply to this email directly, view it on GitHub <https://gist.github.com/832f9ab09e4a3d23d0d5?email_source=notifications&email_token=ADFBPBABA6V5G5UK56BR4N3PZTTN5A5CNFSM4HWJHH62YY3PNVWWK3TUL52HS4DFVNDWS43UINXW23LFNZ2KUY3PNVWWK3TUL5UWJTQAFTMDA#gistcomment-2938928>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ADFBPBDHSL6GCQPAZPIFGITPZTTN5ANCNFSM4HWJHH6Q> .

-- Regards Talal Masood Web: http://tal.al Cell (UAE): +971 50 945 2953 Cell (PAK): +92 300 412 3745 Sent from my iPhone