Scribd Logo
Upload

Welcome to Scribd!

  • 42 views

    1 PB

    Uploaded by

    Yerawar Akhil surendra
    This document summarizes a journal article that uses logistic regression to predict the outcome of One Day International (ODI) cricket matches. The researchers analyzed data from 1169 ODI matches played between October 2002 and July 2011. They found that home field advantage, team strength based on ICC ratings, previous match results, whether the match was day or night, and which team batted first were significant predictors of the match outcome. The logistic regression model accurately predicted the winning team in 74.9% of matches.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    1 PB

    Uploaded by

    Yerawar Akhil surendra
    42 views3 pages
    This document summarizes a journal article that uses logistic regression to predict the outcome of One Day International (ODI) cricket matches. The researchers analyzed data from 1169 ODI matches played between October 2002 and July 2011. They found that home field advantage, team strength based on ICC ratings, previous match results, whether the match was day or night, and which team batted first were significant predictors of the match outcome. The logistic regression model accurately predicted the winning team in 74.9% of matches.

    Original Description:

    pdf

    Original Title

    19695-22229-1-PB

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document summarizes a journal article that uses logistic regression to predict the outcome of One Day International (ODI) cricket matches. The researchers analyzed data from 1169 ODI matches played between October 2002 and July 2011. They found that home field advantage, team strength based on ICC ratings, previous match results, whether the match was day or night, and which team batted first were significant predictors of the match outcome. The logistic regression model accurately predicted the winning team in 74.9% of matches.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    42 views3 pages

    1 PB

    Uploaded by

    Yerawar Akhil surendra
    This document summarizes a journal article that uses logistic regression to predict the outcome of One Day International (ODI) cricket matches. The researchers analyzed data from 1169 ODI matches played between October 2002 and July 2011. They found that home field advantage, team strength based on ICC ratings, previous match results, whether the match was day or night, and which team batted first were significant predictors of the match outcome. The logistic regression model accurately predicted the winning team in 74.9% of matches.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 3

    Journal of Tourism, Hospitality and Sports

    ISSN (Paper) 2312-5187 ISSN (Online) 2312-5179


    Vol.5, 2015

    www.iiste.org
    An International Peer-reviewed Journal

    Predicting ODI Cricket Result


    Mitesh Shah2
    Parag Shah1*
    1.H L College of Commerce, Ahmedabad 380009, Gujarat, India
    2.S V Commerce College, Ahmedabad, Gujarat, India
    * E-mail of the corresponding author: [email protected]
    Abstract
    Result of one Day International (ODI) cricket match depends on various factors related to scoring as well as the
    strengths of the two teams. While some of these factors have been well analyzed in the literature, others have yet
    to be investigated. In this paper, statistical significance for a range of variables that could explain the outcome of
    an ODI cricket match is explored. In particular, home field advantage, game plan (batting first or fielding first),
    match type (day or day & night), past performance of team will be key interests in our investigation. For
    purposes of model-building, logistic regression is applied retrospectively to data already obtained from
    previously played matches.
    Keywords: logistic regression, log likelihood, winning margin.
    1. Introduction
    Cricket began to emerge as a very popular worldwide game around the 13th century. Moreover, cricket was one
    of the first sports to use statistics as a tool for illustration and comparison. Although dating back to 13th century,
    compared to sports like baseball, there has not been much statistical modeling work done for cricket. Wood
    (1945) used the geometric distribution to model the total score, while Kimber and Hansford (1993) proposed a
    nonparametric approach based on runs scored for assessing batting performance. Chedzoy (1997) addressed the
    issue of the effect of umpiring errors in cricket. As is typical in games of sport, winning is the ultimate goal.
    Some studies, (De Silva, 2001), analyze the magnitude of the victory, but most consider the factors affecting
    winning. There are cases where the magnitude of the victory is important; and, in fact, large sums of money are
    routinely wagered when it comes to betting on the outcomes of ODI games (Bailey and Clarke, 2006).
    Clarke (1988) used a dynamic programming model to calculate the expected score for games with rain
    interruptions, so that both teams have the same chance of winning the game. Popular article, Duckworth and
    Lewis (1998), introduced a technique for revising the target for games that are shortened due to weather
    interruptions. Factors such as winning the toss and the home team advantage affecting the results of ODI games
    have been studied in the literature. In this paper, statistical significance for a range of variables that could explain
    the outcome of an ODI cricket match is explored. In particular, home field advantage, winning the toss, game
    plan (batting first or fielding first), match type (day or day & night) are key interests in our investigation.
    Logistic regression is applied to historical data for purposes of model-fitting. Our data consists of a set of the
    ODIs played between nations for the time period starting from October 2002 to July 2011. Some of the matches
    were deleted from the analysis due to certain reasons such as abundance of bad weather or when the one team
    was much superior to the other (ranked teams playing non-ranked teams). Tied games were also deleted from the
    analysis. Due to the continuous update of cricket rules, we chose, in particular, to use this most recent data and to
    collect a significant amount of information for 9 years.
    2. Data
    Variables like Pitch, Avg. First innings Score at this venue, Avg. second innings Score at this venue, ICC points,
    number of experienced players, previous match results, % of winning in last 5 matches, team batting first and
    more were considered for 1169 ODI matches from Oct 2002 to July 2011 from www.crickinfo.com.
    3. Modeling
    We model match outcome at the start of a match. We considered all possible factors affecting the outcome at the
    start of a match. Team strength, ground effect and home field advantage were found to be important (based on
    AIC and Nagelkerke R2). For team strength, we have used win percentage differences and the ICC rating
    differences and find that the rating differences have better explanatory power. This may be because the ICC
    rating takes account of result (win, draw, loss), along with the win margin, wickets and opponent rating.
    Winning the toss was also considered in the model fitting but was found to be unimportant. The playing
    conditions vary from ground to ground and country to country. For example, playing conditions in Wankheda at
    Mumbai are quite different than in Leeds at Headingley.
    Results of fitting the logistic regression model to 1169 match outcomes for various sets of predictors:
    log-likelihood, AIC and Nagelkerke R2. Covariates here are win percentage difference, the ICC rating difference,
    home factor, ground effect. We can calculate the probability of win and loss given the position at the start of a

    19

    Journal of Tourism, Hospitality and Sports


    ISSN (Paper) 2312-5187 ISSN (Online) 2312-5179
    Vol.5, 2015

    www.iiste.org
    An International Peer-reviewed Journal

    match. This will help team captains and management to consider their batting and bowling strategy for the match.
    Matches with no results or tie results were deleted. Logistic regression was used to model the outcome of the
    cricket match.
    4. Results
    Out of the total 20 variables used for modeling, variables : teams playing, ICC points of each team , previous
    match result of each team, Day or Night match, Home team or not, team batting first were significant variables
    for the model.
    Prob. =Exp(Y)/ (1+Exp(Y)) = 0.264 - 0603 *Team 1 is not a Home team
    -0.8 *Day/ Night
    + 0.851* Team 1 losing previous match
    + 0.978 *Team 1 winning previous match
    -0.54 *Team 2 losing previous match
    -0.61* Team 2 winning previous match
    - 0.764*Team 1 batting first
    +0.029* ICC points difference+ 0.101* No. of Exp. Player difference
    +0.684*(Day/Night*Team 1 batting first)
    +0.646* (Day/Night* Team 1 not a Home team * Team 1 batting first)
    Team 1
    (+) 1.063 *Aus + 0.931 *Bangladesh - 0.303*Canada + 0.966*Eng +
    0.904*Ind + 0.834*Ireland - 0.534*Kenya + 0.457*Netherlands + 1.287*NZ + 0.754*Pak + 0.917*
    Scotland+1.001*SA+0.879*SL+ 0.219*WI
    Team2
    (-) 1.015 *Aus-0.253 *Bangladesh+0.531*Canada-0.695*Eng-0.56*Ind - 0.446*Ireland
    + 0.329*Kenya - 0.19*Netherlands - 0.235*NZ -0.543*Pak + 0.206* Scotland-0.341*SA - 0.413*SL-0.594*WI
    The classification of the model is given below:
    MODEL ACCURACY
    Classification Table
    Predicted
    Winning team 1
    0
    1
    173
    360
    121
    515

    Actual Result
    Winning team 1

    0
    1

    Overall Percentage

    Correct %
    67.5
    81
    74.9

    Note:
    Model Accurately predicts the results in 74.9% of matches.
    Winning Team is predicted winning in 81% of matches
    References
    Chedzoy O.B., (1997) The effect of umpiring errors in cricket, The Statistician, 46, 4, 529-540.
    Clarke, S. R. (1998) Dynamic programming in one-day cricket-optimal scoring rates, Journal the Operational
    Research Society, 39, 331-337.
    De Silva, B. M., and Swartz, T.B. (2001) Estimation of the magnitude of the victory in one-day cricket. Austr.
    New Zeal. J. Statist., 43, 1369-1373.
    De Silva, B. M., and Swartz, T.B. (1997) Winning the coin toss and the home team advantage in one-day
    international cricket. Austr. New Zeal. J.Statist., 32, 16-22.
    Duckworth, F. and Lewis, T. (1998) A fair method for resetting the target in interrupted one-day cricket matches.
    J. Oper. Res. Soc. , 49, 22-28.
    Kimber, A.C. and Hansford A.R. (1993) A statistics analysis of batting in cricket, Journal of the Royal Statistical
    Society Series A, 156, 443-455.
    Wood, G.H. (1945), Cricket scores and geometrical progression, Journal of the Royal Statistical Society Series A,
    108, 12-22. www.cricket.org.

    20

    The IISTE is a pioneer in the Open-Access hosting service and academic event management.
    The aim of the firm is Accelerating Global Knowledge Sharing.
    More information about the firm can be found on the homepage:
    http://www.iiste.org
    CALL FOR JOURNAL PAPERS
    There are more than 30 peer-reviewed academic journals hosted under the hosting platform.
    Prospective authors of journals can find the submission instruction on the following
    page: http://www.iiste.org/journals/ All the journals articles are available online to the
    readers all over the world without financial, legal, or technical barriers other than those
    inseparable from gaining access to the internet itself. Paper version of the journals is also
    available upon request of readers and authors.

    MORE RESOURCES
    Book publication information: http://www.iiste.org/book/
    Academic conference: http://www.iiste.org/conference/upcoming-conferences-call-for-paper/

    IISTE Knowledge Sharing Partners


    EBSCO, Index Copernicus, Ulrich's Periodicals Directory, JournalTOCS, PKP Open
    Archives Harvester, Bielefeld Academic Search Engine, Elektronische Zeitschriftenbibliothek
    EZB, Open J-Gate, OCLC WorldCat, Universe Digtial Library , NewJour, Google Scholar

    You might also like