Difference between revisions of "ISSS608 2017-18 T1 Group24 Report"
Sauravj.2017 (talk | contribs) (→Teams) |
Sauravj.2017 (talk | contribs) (→Teams) |
||
Line 18: | Line 18: | ||
To understand the performance of each team, we asked ourselves a few questions and tried to get the answers to them through visual analysis of the data present at hand. | To understand the performance of each team, we asked ourselves a few questions and tried to get the answers to them through visual analysis of the data present at hand. | ||
For this, we first consolidated our data using the ‘tidyverse’ package in R. Then, with the ‘Toss win %’ of each team on the y-axis and the ‘Match wins %’ of each team on the x-axis, taken as an aggregate of all matches across all 10 seasons of IPL so far, we were able to produce the following scatterplot. Also, the color of each team’s label was an indication of the no. of times it had won the IPL trophy. | For this, we first consolidated our data using the ‘tidyverse’ package in R. Then, with the ‘Toss win %’ of each team on the y-axis and the ‘Match wins %’ of each team on the x-axis, taken as an aggregate of all matches across all 10 seasons of IPL so far, we were able to produce the following scatterplot. Also, the color of each team’s label was an indication of the no. of times it had won the IPL trophy. | ||
− | |||
[[File:TossWinVsWinPer.png|800px]] | [[File:TossWinVsWinPer.png|800px]] |
Revision as of 12:32, 2 August 2018
Abstract
Understanding team dynamics in sports is never easy. In a game of Cricket, it’s even harder because teams show so many ups and downs on the cricket field each time they come out to play, its hard to say that there is a one-stop solution that can be implemented on all players for all formats of the game. To overcome this problem, we have conducted an exploratory data analysis to identify patterns in players (batsman and bowlers) and the teams they have been a part of through a medium of visualization to better throw light on their performances over the past 10 seasons (2008-18). Through various comparisons between batsman, bowlers, and team performances, we have highlighted various anomalies, validated certain presumptions, and created a visualization dashboard that, if used, would allow team managers to buy better players for upcoming seasons of the IPL.
Introduction
The Indian Premier League is a professional Twenty20 cricket league in India contested during April and May of every year by teams representing Indian cities and some states. It is a tournament that started in 2008 and has ever since continued to be in action for the past 10 years. In the game of Twenty20 Cricket, each team gets to bat 20 overs and while one team bats, the other team will bowl and fields. The purpose of the game is for the team batting second to score one more run than the team that batted first, within the timeline of 20 overs and without losing 10 wickets. Each over consists of 6 legitimate balls thrown down at the batsman and the batsman are expected to score runs of each ball thrown at them. When a bowler is bowling at the batsman facing him, the rest of the players in the bowler’s team are expected to stop the ball and the batsman from scoring runs.
Understanding the data
To start the analysis, we selected the data via Kaggle that consisted of two files namely, ‘deliveries’ and ‘matches’. The dataset ‘deliveries’ records each delivery bowled by the bowler at the batsman and the outcome of that delivery for all matches over 10 years. The dataset ‘matches’ explains factual data like which team played against which team on which date, where, and the outcome of the match, amongst others. After going through the data, we decided to visualize the information present here by dividing it into three categories: Batsman, Bowlers, and team performances. We believe that doing this will help us understand and visualize each of the critical parameters that go into judging the performance of players on the cricket field. We hope the following analysis will add value to your life in terms of understanding the sport better, gaging our understanding of the sport from our perspective, and help bring out certain anomalies and trends about this game that you may otherwise have not heard of.
Analysis
We have divided our visualizations into different sections as per overall team performances, batsman (individuals), and overall performances of batsman and bowlers, amongst others.
Teams
To understand the performance of each team, we asked ourselves a few questions and tried to get the answers to them through visual analysis of the data present at hand.
For this, we first consolidated our data using the ‘tidyverse’ package in R. Then, with the ‘Toss win %’ of each team on the y-axis and the ‘Match wins %’ of each team on the x-axis, taken as an aggregate of all matches across all 10 seasons of IPL so far, we were able to produce the following scatterplot. Also, the color of each team’s label was an indication of the no. of times it had won the IPL trophy.