/EX2_ML_FOOTBALL

title author date output
Football_spLEAGUE
Adam Daniel
April 23, 2016
html_document

Reuired Libaries

require(XML)
require(RCurl)
require(plyr)
require(gtools)

Data from football statistics website of spanish league 2014-15 & 2015-16

fileURL <- "http://www.football-data.co.uk/mmz4281/1516/SP1.csv?accessType=DOWNLOAD"
download.file(fileURL, destfile = "./fbData.csv", method = "curl")
fileURLTwo <- "http://www.football-data.co.uk/mmz4281/1415/SP1.csv?accessType=DOWNLOAD"
download.file(fileURLTwo, destfile = "./fbDataTwo.csv", method = "curl")

Saving the date of the extraction date

dateDownloaded <- date()
dateDownloaded

Open the data after been saved as csv file in my working directory

footballData <- read.csv("./fbData.csv", sep = ",", header=TRUE)
footballDataTwo <- read.csv("./fbDataTwo.csv", sep = ",", header=TRUE)

Subset which cutting the unnessecery cols

footballData <- subset(footballData, select = -c(B365H : BbAvAHA))
footballDataTwo <- subset(footballDataTwo, select = -c(B365H : BbAvAHA))

Bind the data to one data frame

total <- rbind(footballData,footballDataTwo)

adding cols to list games by zero -> tie , one -> away lead , two -> home lead

listGames <- subset(total,select=c(HomeTeam,AwayTeam,HTHG,FTHG,HTAG,FTAG))
listH <- listGames$HTHG - listGames$HTAG
listF <- listGames$FTHG - listGames$FTAG
listGames <- cbind(listGames,listH,listF)
colH <- ifelse(listGames$listH<0, 1, ifelse(listGames$listH>0, 2, 0))
colF <- ifelse(listGames$listF<0, 1, ifelse(listGames$listF>0, 2, 0))
listGames <- cbind(listGames,colH,colF)

Show the density of half time score and full time score by home and away

par(mfrow=c(2,1),mar=c(4,4,2,1))
d <- density(listGames$colH)
plot(d,main = "Half Time score ")
d <- density(listGames$colF)
plot(d,main = "Full Time score ")

image:

##Conclusions

  1. The home team usually wins the game, you can see in the density graph of final results that the density on home team (2) is higher.
  2. Mostly in Halftime score , the home team leads and symmetrically have a similar chance of a tie. This implies that the chances of the away team to lead in first half are significantly lower.

Summary: Home teams have more confidence during the game. Away team getting tie result on the first half is a good result for the next half of the game.