My Research for the 6th International
Science of Judo Symposium

"USE OF ATTACK RATE AS A PREDICTOR
OF VICTORY IN OLYMPIC LEVEL JUDO"

As part of my BSc in Sport Performance at the University of Bath, I completed a research project. This project consisted of an examination of the number of "Ineffective", "Effective" and "Scoring" attacks used by the athletes at the Beijing Olympic Judo Tournament 2008.
On this page I have chosen to share all the information I can about about the research under a Creative Commons license. This includes the BSc research paper, research poster from the 6th International Science of Judo Symposium. Included also is the SPSS data files and the source code for the software developed to conduct the notation for the research (The software is released under a GPL license).

The purpose of this research was to explore if attack rate could be used to predict the winner of Judo matches. I was looking to explore if attack rate is a viable statistic to use to predict who will win a Judo match. I also used the project as an opportunity to explore previous methods of notating Judo, specifically the seminal work of Sikorski et al. from 1987.

Below is the information from the project, please do read it through and criticise. Better yet, take the information and repeat it and tell me what results you get. Even better than that, lets start a conversation about the findings and the methods I used and perhaps we can move the area forward. Please email me at lw@judocoach.com.

• BSc research project
• Symposium Poster
• Data (main)
• Data (initial)
• Medalists Info
• Notation Paper Form
• Notation Software Source Code

                   DEPARTMENT OF EDUCATION
USE OF ATTACK RATE AS A PREDICTOR OF VICTORY
IN OLYMPIC LEVEL JUDO
BY
LANCE D.C. WICKS
Supervised by
Mike Callan
A Work Based Research Project submitted in partial fulfilment of a
BSc (Hons) in Sport (Sports Performance) (Work Based Learning)
2009
Acknowledgements
Lance Wicks – ED30391 - 2/39
Cheating and plagiarism statement
I confirm the following:
I have read and understood the following that explains cheating and plagiarism (a)
the University of Bath University web site, and (b) The Handbook for BSc in Sport
(Sports Performance) (Work Based Learning).
To the best of my knowledge, this work based research project does not contain
plagiarised material.
Name of Student: Lance D.C. Wicks.
Signature of student:
Lance Wicks – ED30391 - 3/39
Table of Contents
Acknowledgements.......................................................................................................2
Cheating and plagiarism statement ..............................................................................3
Table of Contents...........................................................................................................4
Chapter 1: Work Based Research Project Proposal:....................................................5
Chapter 2: Project Strategy and Plan............................................................................8
Chapter 3: Project Log.................................................................................................11
Chapter 4: Project Report :..........................................................................................14
Results................................................................................................................14
The Structure of Judo Fights..............................................................................17
Attack rate at the 2008 Beijing Olympic Judo Tournament.................................17
Medalists attack rates.........................................................................................22
Scores in Judo fights...........................................................................................24
Population and Sample.......................................................................................25
New and important aspects of the study:...........................................................25
Critical evaluation of the methods.......................................................................27
Limitations...........................................................................................................28
Future research:..................................................................................................31
References:.................................................................................................................33
Appendices:.................................................................................................................39
Lance Wicks – ED30391 - 4/39
Chapter 1: Work Based Research Project Proposal:
Student’s Name:
Lance D.C. Wicks.
Broad subject area identified as a potential work based research project:
Statistical analysis of the attack rates, scoring rates and victory in matches
involving elite Judo athletes. The purpose of this study is to determine if the
attack rate of the two players in a Judo match can be used to predict the
outcome of the match.
Specific work based issue (title) chosen to study as a work based project:
Use of Attack Rate as a Predictor of Victory in Olympic Level Judo
Justification of using qualitative / quantitative research methods:
Using quantitative research methods is justified as the nature of the research
is to examine events that occur within a Judo match and define statistical
significance.
Qualitative research methods would not be appropriate as this study does not
require understanding of subjective matters, such as athletes emotional states.
A quantitative method is practical in this study as the author is unable to attend
the tournament in person (for example to conduct interviews for qualitative
research), but will be able to use video footage from the BBC to conduct
quantitative data collection.
The proposed notation system is based on the methods used as far back as
1987 (Sikorski, Mickiewicz, Majle, & Laksa, 1987) but modified to include
attacks that do not cause a score and also do not meet the criteria used by
researchers such as Boguszewski & Boguszewska (2006a)
Lance Wicks – ED30391 - 5/39
Suggested data collection techniques:
Data collection will be done via hand notation of video footage of matches
from the 2008 Beijing Olympic tournament. If this proves to be too time
consuming a computer software would need to be developed to collect the
data.
Given the quantity potentially involved this is most likely.
Suggested sample:
The Olympic Judo tournament runs for a week, this makes collecting data on
the entire event very time consuming. It is suggested then that a sample of
matches be notated and used for the purposes of this study.
The suggested sample is of fights from the second round of matches up to but
not including quarter-final, semi-final and finals.
This sample excludes the final rounds, where it is suggested the structure of
fights changes as tactics and strategy affect performance. It also excludes the
opening round where less capable athletes are knocked out of the tournament.
References:
! Sikorski, W., Mickiewicz, G., Majle, B., & Laksa, C. (1987). Structure of the
contest and work capacity of the judoist. Proceedings of the International
Congress on Judo ÒContemporary Problems of Training and Judo Contest, 9-
11.
! Sterkowicz, S., Lech, G., & Almansba, R. (2007). The course of fight and the
level of sports achievements in judo. Archives of Budo, 3, 72-81.
! Hughes, M., & Franks, I. M. (2004). Notational Analysis of Sport: Systems for
Better Coaching and Performance in Sport. Routledge.
! Thomas A. Zak, Cliff J. Huang, & John J. Siegfried. (2008, January 21).
Production Efficiency: The Case of Professional Basketball. research-article, .
Retrieved May 12, 2008, from http://www.jstor.org/pss/2352368.
! Boguszewski, D., & Boguszewska, K. (2006). Dynamics of judo contests
performed by finalists of European Championships (Rotterdam 2005).
Archives of Budo, 2, 40-44.
Lance Wicks – ED30391 - 6/39
! Degoutte, F., Jouanel, P., & Filaire, E. (2003). Energy demands during a judo
match and recovery. Br J Sports Med, 37(3), 245-249. doi:
10.1136/bjsm.37.3.245.
! Kalina, R. M., Kulesza, A., Myslowski, B., Wolkowicz, B., Jagiello, W., Gabrys,
T., et al. (2004). Dynamics of Judo, Boxing and Taekwon-do Contests
Performed by Finalists of Olympic Games in Sydney. Sport Training in
Interdisciplinary Scientific Researches. Czestochowa.
Signed (student): _____________________
Lance Wicks – ED30391 - 7/39
Chapter 2: Project Strategy and Plan
Outside the sport of Judo, considerable research has been conducted into the
efficacy of attacking actions. For example the 2004 European Cup Soccer
Tournament was analysed in detail with the focus being on descriptive statistics and
efficacy of aspects of team play (Carmichael & Thomas, 2005). Football and Baseball
have had considerable research into this area conducted (Act, 2004;
Papahristodoulou, 2006, 2007; Schwarz, 2005).
In sports that have gambling opportunities, more research has been put into the
study of predictive probabilities than in minority sports such as Judo. Tennis for
example has had examinations into predicting results (Forrest, 2007). Research into
greyhound and horse racing has also been conducted (Clarke, Bailey, & Yelas, 2008;
Hausch, Ziemba, & Rubinstein, 1981; Snyder, 1978; View, 1994) which extends firmly
into predicting outcomes based on a variety of performance indicators.
The works of Papahristodoulou (2006, 2007) are interesting in relation to this study
as his conclusions suggest a relationship between “shots at goal” and “victory”; as
this study proposes a relationship between “attacks” and “victory” in Judo.
Heinisch presented research at the 5th International Judo Federation World
Research Symposium (Emerson Franchini & Boscolo Del Vecchio, 2007) that
supports the argument that attacking more results in victory.
The influential Judo researcher Stanislaw Sterkowicz examined activity rate and
compared medallist and non-medallists (Sterkowicz, Lech, & Almansba, 2007). This
study suggests that higher attack rates result in victory.
Within Judo there is not a standard set of performance indicators that can be
analysed statistically to determine probabilities. By following the examples from
Basketball (Kubatko, Oliver, Pelton, & Rosenbaum, 2007), Rugby (Bracewell,
2002) and American Football (Harville, 1980), and from research in other sports, this
project aims to determine if the elements being analysed are useful performance
indicators within a Judo match.
Lance Wicks – ED30391 - 8/39
Unlike previous studies, this project aims not to describe Judo matches, rather decide
if the data being analysed is able to predict the result of the match. This may, as with
Bailey & Clarke (2006), allow coaches and team managers, etc. to predict the winner
of matches whilst the match is in progress.
A paper-based hand notation form, based on the work of Hughes & Frankes
(2004) and the lectures by Simon Hicks (2005), was designed and will be used in
this research. One form per match will be used to record the attacks, scores and
eventual winner of the match.
During the research process, the hand notation system was replaced by a softwarebased
notation system that was found to be both more accurate and quicker to use.
This is discussed in the project report later in this document. For the purposes of this
project strategy the methodology used on the paper based forms is described. This
notation method is replicated in the software version, with key presses replacing
hand written notations (source code for the software is included in the appendices).
In reviewing the design of previous research (Boguszewski, 2006; Boguszewski &
Boguszewska, 2006b; Sterkowicz et al., 2007; Sterkowicz & Maslej, 1999) it became
apparent that the earlier designs had solved the design problem of deciding what
was or was not an attack. These earlier studies either ignored all non-scoring attacks
or found objective definitions of a non-scoring attack. For example, deciding an
effective (though non-scoring) attack was any occasion where the opponent touched
the ground with any part of the body except the feet.
Within Judo there are non-scoring attacks that do not force the opponent to touch the
floor which could be considered valid attacks. Excluding these attacks from an
examination of Judo could lead to inaccuracies that would be expected to affect the
inferences possible from the data collected (Mike Hughes & Ian M. Franks, 2004).
To address this design limitation, the author developed a method of notation that
includes all three types of attack. These are explained and detailed on the following
pages.
Lance Wicks – ED30391 - 9/39
On the paper based hand notation form each attack will be recorded as a vertical line
“|” if the opponent does not touch the floor with another part of the body other than
the feet. These attacks include any attacking action where the attacker turns their
body past 45 degrees and any attacking movement where they “grab” the opponent
whilst facing forward. It also includes any attacking movement where the players feet
touch the opponent.
These attacks are recorded as a “|” on the notation forms.
A “+” will be recorded if the attack causes the opponent to touch the floor with any
part of their body other than their feet, but no score is given.
If a score is given, no “|” or “+” is marked, rather a letter indicating the score is written
instead. These are K for Koka, Y for Yuko, W for Wazari, I for Ippon and P for a
penalty. Penalties are recorded against the player being penalised, not the player
benefiting.
When Matte is called, the recorder moves down one line on the notation form. If no
attack has been made a horizontal line is recorded “-”, if neither player has attacked,
then both columns have a “-” recorded.
The final score and duration of the match shall be recorded on the form from the
official results (“The official website of the BEIJING 2008 Olympic Games,” n.d.).
Inter and Intra Operator reliability is considered important (Mike Hughes & Ian M.
Franks, 2004; A.M. Nevill, G. Atkinson, M.D. Hughes, & S.M. Cooper, 2002; M.
Hughes, S. M. Cooper, A. Nevill, & S. Brown, 2003; G Atkinson & A M Nevill, 1998) ,
so within this study both types of reliability shall be tested. This will be done with
multiple notations of a set of fights by the author and also by teaching the notation
methodology to a group of students who will notate the same fight.
Lance Wicks – ED30391 - 10/39
Chapter 3: Project Log
Data collection phasing
When comparing the available video footage (8 X 8 hour DVDs) and the resources
available (1 X Person) consideration as to scheduling of the data collection was
important. A second factor was the author's ability to collect data accurately during
extended notation periods. The author found that when notating on paper based
forms only two Judo matches could be recorded accurately before serious errors in
notation started to occur. With the later software-based notation system, six Judo
fights could be recorded accurately before serious errors started to occur.
Also the authors work and family commitments needed to be factored into the
sequencing of the data collection. This scheduling revolved around a timetable of 1
hour 4 times per week.
One hour allowed the notation of an average of 5 Judo fights. This equalling 4 hours
of footage per week, 20 fights. The notation period was scheduled to be from the first
week in September 2008 to the second week of March 2009, 20 weeks in total
factoring in holidays, etc. This was estimated to allow data collection from the entire
footage available.
Unfortunately this was not the case in practise. Time management was an issue with
notation sessions being less frequent than planned.
The video footage used in this study was from the BBC Interactive service, and as
such some fights were not included in the footage. This was because the BBC
footage covered one mat of the two in operation, and some fights were also excluded
if the footage was incomplete or action was missed due to video replays etc.
To explore the reliability and accuracy of the data being collected, extra steps were
introduced. Before, during and after the data was collected from the DVD footage,
intra-operator reliability testing was completed. The author notated three fights, three
times at the start and end of the project.
An inter-operator test was conducted at the University of Bath in April 2009. The
author taught nine students in the first year of the FdSc Sport, who were all Judo
Brown belts and above using the paper-based notation system. These students then
Lance Wicks – ED30391 - 11/39
notated a fight individually and the notation sheets were collected by the author and
used to determine reliability levels. The fight notated was the same as used in the
intra-operator testing.
Process of data collection
The data was initially collected using a paper based form that was designed by the
author for a pilot study (Wicks, 2006). This was however very time consuming and
error prone, limiting the amount of data able to be collected.
An electronic notation system was written by the author in the Perl programming
language, and used for a majority of the data collection (70% of Judo fights
recorded), the source for this software is included in the appendix. This data was
then entered into a spreadsheet (OpenOffice.org, n.d.) and at completion of the data
collection phase this data was then exported into PAWS Statistics 17.0 statistical
analysis package.
The notation form included recording the official scores in each fight, this was
sourced from the official site (“The official website of the BEIJING 2008 Olympic
Games,” n.d.). Disk number, fight number (per disk), and time stamps for start and
finish of fights was also recorded on the notation form to ensure locating fights again
was possible if required.
Once collected the data was analysed for reliability, this is covered later in this
document.
Skill sets used evaluation
During the planning and execution of this research project a number of existing and
new skills were required. Time management skills were highlighted and the process
of completing the research has improved the authors skills in this area .
The necessity to decrease the time and complexity of notating fights, necessitated
the development of a computer software to achieve this task. The author although
having programming skills had never written software of this type before.
The time constraints also forced the author to make decisions regarding the size and
consistency of the sample group used in this study. This involved evaluation of the
implications of not being able to collect data from an entire population.
Lance Wicks – ED30391 - 12/39
In writing the software required, the author developed new skills in software design,
software programming and software development “best practise”(Conway, 2005;
Gunderloy, 2004). This included the operation of an Integrated Development
Environment (Coda by Panic, n.d.) and the use of revision control software (SVN,
n.d.).
Developing these skills has increased the author's versatility as a person working in
the Information Technology (IT) sector. The software developed could with more
development, be sold to other researchers, becoming a business opportunity for the
author that previously did not exist.
Developing a more advanced notational software package could have benefits to the
author and to the wider Judo community by making the process of collecting
information easier and more standardised. This could lead to the collection of
normative data which could be used by researchers.
In working on this project, the author has become aware of the importance of
accuracy and reliability testing in a research project. This has become part of the
research project itself and extended to the authors work with interpreting and
presenting Judo data on his www.JudoMetrics.com website.
In analysing the data collected, the PASW Statistics software (formerly known as
SPSS Statistics) was used. The author during this process developed skills both in
using the software and also developed a better knowledge of inferential statistics. It
has proven difficult to present the data without a stronger knowledge of statistics,
future research by the author would be assisted by attendance in statistics training
Alternatively (or additionally) consultation with an experienced statistician could be
beneficial.
Lance Wicks – ED30391 - 13/39
Chapter 4: Project Report :
Results
In this study the author aims to prove or disprove the hypothesis that the player who
attacks more in a Judo fight will win the Judo fight. The following tables suggest that
this hypothesis was true at the 2008 Beijing Olympic Judo Tournament.
Table 1.
Summary Showing Number of Fights Won By Player Who Attacks Most.
TOTAL FIGHTS: 58
Population size: 386
BLUE WINS WITH MORE ATTACKS 15
WHITE WINS WITH MORE ATTACKS 17
Total fights won by player attack most: 32
% won by player who attacks most 55%
This simple table (table 1) shows that based on the 2008 Beijing Olympic Judo
Tournament, the hypothesis is true, that the player who attacks more will win a Judo
fight a majority of the time. This concurs with the pilot study done in 2006 at
Commonwealth level (Wicks, 2006) and the work of Heinisch shown at the 2008 Judo
World Championships and International Judo Research Symposium (Emerson
Franchini & Boscolo Del Vecchio, 2007).
This table uses percentages which is not ideal given the small sample size; they have
been used to simplify presentation of the information only. The data also includes 7
matches where the number of attacks was equal and this affects the results.
Lance Wicks – ED30391 - 14/39
The basic structure of Judo matches at this elite tournament can be suggested by the
data in table 2 below:
Table 2.
Structure of Judo fights at the Beijing 2008 Olympic Judo Tournament.
Average duration of fight: 00:03:52
Average Segments per fights: 13.38
Average duration of segment: 00:00:17
Average scores per fight 2.97
Average Scores per fight (excluding penalties) 1.36
Average attacks per fight 21.03
Total Ippon 26 15%
Total Wazari 21 12%
Total Yuko 27 16%
Total Koka 5 3%
Total Penalty 93 54%
Ippon (% of total excluding penalties) 33%
Wazari (% of total excluding penalties) 27%
Yuko (% of total excluding penalties) 34%
Koka (% of total excluding penalties) 6%
Mean number of attacks per match 21.03
Mean number of segments per match 13.38
Mean number of attacks per segment 1.57
The above table indicates that the average fight in Olympic Judo is 3:52 minutes long
and consists of 13 segments of action; with each segment being 17 seconds in
duration.
Over half of all scores at this level are penalties. Scores earned by the players are
almost evenly distributed between Ippon, Wazari and Yuko; with Koka scores being
infrequently scored.
Lance Wicks – ED30391 - 15/39
Through observation of these basic descriptive statistics, we can suggest that the
hypothesis of this study is not proven; only 55% of fights being won by the player who
attacked the most in each fight. This is a majority of fights being one by the player
who attacks most often, however but cannot be considered statistically significant as
is discussed later in this report.
The data relating to the structure of a Judo fight at the Beijing Olympic Judo
Tournament is interesting as it repeats aspects of the work of Sikorski et al. (1987).
The structure of Judo fights identified in this study and in the 1987 study is nearly
identical in terms of duration of overall fights, segments of activity and penalties being
the most frequent score. This too is discussed in more detail in the later discussion
section of this report.
Inter and Intra Operator testing was conducted during this study and identified that
the methodology needs considerable time to learn. The hand notation system used
originally was slow and inaccurate compared to the computerised system used later,
this concurs with the views of Mike Hughes & I. M. Franks (2004).
The most prevalent score in the Beijing 2008 Olympic Judo Tournament is penalties
at over 54% of all scores awarded. The next highest score is Yuko scored throws at
just under 16%.
Lance Wicks – ED30391 - 16/39
Discussion
The Structure of Judo Fights
As described earlier in this study, a seminal piece of research in Judo is the work of
Sikorski et al. (1987), in which the structure of a Judo match is defined. The 1987
study has been the basis for many other research projects (Boguszewski &
Boguszewska, 2006a; Degoutte, Jouanel, & Filaire, 2003; Sterkowicz et al., 2007)
and is an important influence on this study of the Beijing Olympic Judo Tournament.
There has been 20 years of development and rule changes in the sport of Judo since
the 1987 research was published(“International Judo Federation Rules,” 2008), and
comparing the structure of Judo fights from both studies shows some similarities, as
shown in table 3 below.
Table 3.
Structure of Judo fights in 2008 and 1987.
1987 2008
Average Fight Duration 00:03:56 00:03:52
Average segment of action 11 to 20 seconds* 00:17.35
* Sikorski et al. (1987) provide a range rather than an average.
The table above suggests that the average duration of a Judo fight has not changed
considerably in the past 20 years. The duration of a segment of action (the period of
time between the “Hajime” and “Matte” calls of the referee) in 2008 also matches the
range described in 1987.
Attack rate at the 2008 Beijing Olympic Judo Tournament.
Sikorski et al. (1987, p. 62) suggest that the medalists in their study scored on
average twice per fight and gained the equivalent of a Yuko or Wazari score (they
used a numeric scoring system). The following pages examine the 2008 data and
find similarities between the 1987 and 2008 data that could suggest that the structure
of Judo has not changed considerably over the two decades between events, despite
the changes in weight categories and rules (International Judo Federation, 2009;
“International Judo Federation - Weights,” n.d.; “International Judo Federation Rules,”
2008; Villamon, D. Brown, Espartero, & Gutierrez, 2004).
Lance Wicks – ED30391 - 17/39
In this study 55% of all fights were won by the player who attacked the most in the
fight. This can be used to support the hypothesis that a player who attacks more shall
win more, if we consider the prior works in the area such as Sikorski et al. (1987) and
Heinisch (Emerson Franchini & Boscolo Del Vecchio, 2007).
The table below shows the descriptive statistics relating to the Winning Players total
attack rate (WTA) and losing players total attack rate (LTA). This table shows that the
winning players attack only slightly more than the losing players. The mode is
interesting in this table as it is the widest difference between the two sets of figures.
Table 4.
Winning and Losing players total attack rates (including scoring and non-scoring
attacks).
WTA LTA
N Valid 58 58
Mean 11.0517 10.0517
Std. Error of Mean .90686 1.01454
Median 9.0000 9.0000
Mode 7.00a 1.00
Std. Deviation 6.90645 7.72651
Variance 47.699 59.699
Skewness 1.138 1.233
Std. Error of Skewness .314 .314
Kurtosis 1.480 2.256
Std. Error of Kurtosis .618 .618
Range 34.00 38.00
Minimum 1.00 .00
Maximum 35.00 38.00
a. Multiple modes exist. The smallest value is shown
Lance Wicks – ED30391 - 18/39
The table above is based on total attacks including attacks that resulted in a score.
This data clearly shows that the winning player does, on average, attack 1 more time
per match than the losing player, however the standard error of the means overlap
meaning we do not consider these results statistically significant.
The next step was to then look at the attacks that did not result in a score. In this
table WTANS relates to the winning player's total attacks not including scoring
techniques; LTANS refers to the losing players equivalent rate. In this table, the
losing player is shown to make slightly more attacks than the winning player.
Table 5.
Winning and Losing players total attack rates ( NOT including scoring attacks).
WTANS LTANS
N Valid 58 58
Mean 9.8276 9.9138
Std. Error of Mean .91209 1.01241
Median 8.0000 9.0000
Mode 6.00 1.00
Std. Deviation 6.94625 7.71029
Variance 48.250 59.449
Skewness 1.148 1.258
Std. Error of Skewness .314 .314
Kurtosis 1.846 2.319
Std. Error of Kurtosis .618 .618
Range 35.00 38.00
Minimum .00 .00
Maximum 35.00 38
Lance Wicks – ED30391 - 19/39
Separating the data further, we are able to examine the scoring attack frequencies for
winning and losing players in the table below that shows that winning players have a
higher mean for scoring attacks.
Table 6.
Winning and Losing player's total attack rates (including scoring attacks ONLY).
WTAS LTAS
N Valid 58 58
Mean 1.2241 .1379
Std. Error of Mean .12060 .04567
Median 1.0000 .0000
Mode 1.00 .00
Std. Deviation .91849 .34784
Variance .844 .121
Skewness .657 2.156
Std. Error of Skewness .314 .314
Kurtosis .430 2.742
Std. Error of Kurtosis .618 .618
Range 4.00 1.00
Minimum .00 .00
Maximum 4.00 1.00
The average number of attacks that score is shown in this table to be less than once
per match, suggesting that losing players are not frequently able to score against
their opponents.
Lance Wicks – ED30391 - 20/39
The final statistic to consider from this area of the study is the players' tendency to
receive penalties from the referee. This statistic again favours the winning player with
the losing players being penalised on average once per fight, which is the equivalent
of having a Koka scored against them by the winning player.
Table 7.
Winning and Losing player's penalty rates.
WP LP
N Valid 58 58
Mean .60 1.00
Std. Error of Mean .110 .139
Median .00 1.00
Mode 0 0
Std. Deviation .836 1.060
Variance .700 1.123
Skewness 1.803 .641
Std. Error of Skewness .314 .314
Kurtosis 4.243 -.505
Std. Error of Kurtosis .618 .618
Range 4 4
Minimum 0 0
Maximum 4 4
Looking at these four sets of statistics we can suggest that it is the number of attacks
that result in a score, and the penalty rate, that result in victory and not the nonscoring
attack rate.
Lance Wicks – ED30391 - 21/39
Medalists attack rates.
Examination of the tendencies of medalists winning matches, from within the sample
is interesting and provides an interesting comparison to summary statistics, despite
the sample being very small.
Table 8.
Medal winners winning against non-medalist
Fights medalist beats non-medalist: 21
% Where medalist attacked more than non-medalist: 67%
Attack rate of medalists: 11.9524
Attack rate of non-medalists: 10.0000
This table suggests that the medalists held a higher attack rate than that of their
opponents. When compared to the 55% average from the entire sample this provides
more support to the hypothesis that winning in Judo involves attacking more than
your opponent.
The attack rate for medalists is also higher than that for winning players generally
across the sample. Medalists are attacking almost 2 more times than their opponents
and almost 1 more time than winning players generally.
The attack rate of non-medalists is very close to the mean for losing players
generally, suggesting that it is the medalists that have a different behaviour to the rest
of the players.
Examination of the penalty rate of this sub-sample group shows little difference from
that of the general sample group. Table 9 below, when compared to table 7, suggests
that the penalty rate is not what differentiates the medalists from other winning
players.
Lance Wicks – ED30391 - 22/39
Table 9.
Penalty rates of Medalists and Non-Medalists in fights where Medalists beat non
medalists.
Penalty Rate Std. Error of Mean
Medalists: 0.67 0.22
Non-medalists: 1.000 0.249
This examination of the Medalists when compared against the general sample of
winning players does suggest that the hypothesis is correct that the player that
attacks more wins the fight. We need to consider the sample size and we cannot
consider the results statistically significant, but they do give a perspective of the data
that might otherwise be missed.
Lance Wicks – ED30391 - 23/39
Scores in Judo fights.
Within this study an examination of the scores awarded was included. This data is
presented in the table below which shows the dominance of the Penalty score within
the sport of Judo in 2008.
Table 10.
Total number of scores analysis
TSI TSW TSY TSK TSP
N Valid 58 58 58 58 58
Mean .4483 .3621 .4655 .0862 1.6034
Std. Error of Mean .06587 .07660 .08941 .03718 .22400
Median .0000 .0000 .0000 .0000 1.0000
Mode .00 .00 .00 .00 .00
Std. Deviation .50166 .58334 .68096 .28312 1.70592
Variance .252 .340 .464 .080 2.910
Skewness .214 1.389 1.510 3.027 1.112
Std. Error of
Skewness
.314 .314 .314 .314 .314
Kurtosis -2.025 1.004 2.370 7.420 1.768
Std. Error of
Kurtosis
.618 .618 .618 .618 .618
Range 1.00 2.00 3.00 1.00 8.00
Minimum .00 .00 .00 .00 .00
Maximum 1.00 2.00 3.00 1.00 8.00
Sum 26.00 21.00 27.00 5.00 93.00
TSI = Total Scores Ippon, TSW = Total Scores Wazari, TSY = Total Scores Yuko,
TSK = Total Scores Koka, TSP = Total Scores Penalty.
Lance Wicks – ED30391 - 24/39
Population and Sample
The total population for this study is 386 fights (Fischer, 2008), the sample size of 58
fights is therefore 15% of the total population. The sample is taken from six of 14
weight categories. Within these categories the sample represents between 14% and
17% of the population of that weight category's fights.
The total video footage observed was 75 fights, however 17 fights were rejected as
the footage was either incomplete, interrupted or periods of action were not visible
due to video replays, etc.
The sample is an opportunistic sample and as such consideration must be made of
the fights notated and their position within the structure of the tournament. The
sample does not include any medal fights, with a majority (55%) of notated fights
being sourced from the second round of fights at which point half the competitors
have already been eliminated.
The sample is also taken from medium weight categories, not included are the light
weight or heavy weight categories. This could affect the relevance of the results and
how they should be interpreted.
When compared to the sample used by Sikorski et al. (1987) it should be observed
that the 1987 study looked only at male competitors whereas this study looks at male
and female Judo players. The sample in this study is also from a single event.
New and important aspects of the study:
This is an area that has not been explored explicitly before in Judo research.
Researchers such as Heinisch (Emerson Franchini & Boscolo Del Vecchio, 2007)
and Sikorski et al. (1987) have explored it in passing, but this is the first time the
attack rate to victory relationship has been explored as a prime objective of research.
The hypothesis tested in this study is not considered proven, due to this marginal
nature of the results. This is unexpected and may indicate too small a sample size or
perhaps that Olympic Judo (or this specific event) is in some way different to the
events studied previously. However, as the following findings show, we can suggest
that the Judo at this event and from earlier studies are similar, lending more weight to
the suggestion that the sample size for this study is the reason for the marginality of
Lance Wicks – ED30391 - 25/39
results. Also the examination of medalists data more strongly supports the
hypothesis, suggesting systemic errors in the methodology.
This study has found that the structure of Judo fights has not changed since 1987,
despite a variety of rule changes in the sport. This is an area that has not been
directly confirmed since the 1987 study and as such is of relevance to Judo
researchers.
The findings of this study suggest that the hypothesis is likely to be, true, although
unfortunately not proven as not a statistical significant difference; attacking more in
Judo will result in victory. The study suggests that over 55% of fights are won by the
player who attacks the most. Deeper examination of the statistics suggest that
scoring attacks are arguably the key variable that relates to winning the match.
The study also confirms the findings of Sikorski et al. (1987) in relation to the average
duration of fights and duration of segments of action in Judo fights. In the 1987 study,
the most common scoring event was identified as penalties, this matches with this
study where penalties make up over 50% of all scores in Beijing.
In this study segments of action were found to be on average 17 seconds in duration.
In this study the duration of breaks between pauses were not observed, however the
number of segments of action was observed to be on average 13. Sikorski et al.
(1987) do not describe the number of segments per fight.
We can suggest, given the close similarities between this study and the 1987
research, that the number of segments in 1987 would be similar to the 17 figure of
2007. We can also suggest that the breaks between segments in 2007 would be
similar to those identified in 1987; approximately 10 seconds.
Lance Wicks – ED30391 - 26/39
Implications:
This study has confirmed the finding of the 1987 study by Sikorski et al. The Sikorski
research included investigation into blood lactate levels in athletes and has served as
the basis for various studies looking at the physiology of Judo (E. Franchini, Nunes,
Moraes, & Del Vecchio, 2007; Ribeiro, Tierra-Criollo, & Martins, 2006).
If the structure of Judo has not changed since the 1987 research as suggested by
this study, then existing training methodologies based on the 1987 research remain
valid. This means that existing strength and conditioning programmes would not need
to be re-designed for a different physiological requirement.
The suggestion that attack rate has a direct result on the outcome of a Judo fight
could be used to propose different behaviour during matches by Judo players.
Players might be coached to attack more frequently.
Critical evaluation of the methods
The methodology used in this study has several issues that must be considered. The
methodology requires considerable training to produce consistent results. This was
proven in the inter and intra operator testing. The sample size can also be considered
an issue.
The identification of non scoring ineffective attacks is subjective and contributes to
the inter-operator variations observed. It also makes repeatability of the research
difficult. Future research could either choose to ignore these ineffective attacks or
develop a more objective method of identifying these attacks.
Use of BBC television footage proved troublesome, resulting in a large number of
fights having to be excluded. It would have aided the methodology to use footage
filmed specifically for the study, without the interference of action replays and
onscreen graphics etc.
The notation software developed assisted considerably with the speed, ease and
accuracy of data collection. Specific improvements to the software would include a
live onscreen display of variables and the ability to reverse a key press. It may also
be worthwhile considering using generic video analysis software such as Dartfish or
Sportscode.
Lance Wicks – ED30391 - 27/39
The data, it can be argued, does not accurately represent the effectiveness of
attacks. The selection of throwing techniques can affect the statistics. For example,
Uchi Mata attacks were observed to be effective (although non-scoring) from a Judo
perspective, in that the opponents balance was broken demonstrating good “kuzushi”
(Kano, 1994; Almansba et al., 2007; Sugai, 1992).but were notated as ineffective in
this study as the opponent remained standing, despite being very close to being
scored against.
The opposite effect was observed of players using “drop” techniques such as Drop
Seoi Nage or Drop Kata Guruma (Inman, 2005; Nakanishi, 1998). Attacks that Judo
terms would be considered ineffective were notated as effective as the opponent
often used standard defensive tactics that involved dropping onto their knees, often
to move into an attacking position in ground fighting.
Limitations
This study has limitations that need to be considered when assessing the findings.
The main limitations found in this study relate to the complexity of predicting Judo
fight outcomes and the issues related to collecting data to make predictions.
This study is drawn from a small sample of a much larger population. It does not
include all fights from the tournament and is based on only one tournament. As such,
any inferences drawn can only be confidently said to relate to the single event and
the relevance across the sport of Judo generally is questionable.
The rules of Judo are also a limitation to this study. Five months after the event, the
International Judo Federation (the governing body for Judo) changed the rules to
exclude the Koka score (“International Judo Federation Rules,” 2008). This means
that the repeatability of this study at other Judo tournaments is affected. All results
looking at similar data will be considerably different as the Koka score will not be part
of the study. Also excluded is the first penalty awarded, this as of 1 January 2009 is a
warning only, and will affect the results of any future study.
Lance Wicks – ED30391 - 28/39
A further limitation of this study is the reliability of the data collected, the inter and
intra-operator reliability testing suggest that the reliability of the data needs to be
considered carefully.
Table 11.
Inter-Operator Reliability Testing
Student Segments Blue
Ineffective
Attacks
Blue
Effective
Attacks
Blue Koka Blue Yuko Blue
Wazari
Blue Ippon
R1 26 11 8 0 0 0 0
R2 15 12 3 0 0 0 0
R3 20 11 7 0 0 0 0
R4 19 4 1 0 0 0 0
R5 20 9 4 0 1 0 0
R6 10 11 7 0 0 0 0
R7 25 16 5 0 0 0 0
The table above shows clearly the large differences in data collected by the students
all received 15 minutes of explanation of the notation method at the same time during
a lecture by the author.
This suggests that the level of training was not sufficient and that some of the
students may not have understood the task properly. There may also have been
language difficulties as some students were not native English speakers. Finally, we
can consider also that the students were of differing technical levels in the sport of
Judo and this may have had an effect on their ability to notate the fight.
Intra-operator reliability was also tested, the author notated the same fight three
times at the start of the data collection process and again at the end of the data
collection phase, the table below shows the results of that testing.
Lance Wicks – ED30391 - 29/39
Table 12.
Intra-Operator Reliability Testing
Segments Blue
Ineffective
Attacks
Blue Effective
Attacks
Blue Koka Blue Yuko Blue Wazari Blue Ippon
O1 14 17 4 0 0 0 0
O2 15 15 5 0 0 0 0
O3 15 15 5 0 0 0 0
O4 15 16 5 0 0 0 0
O5 15 15 4 0 0 0 0
O6 14 16 5 0 0 0 0
Observations O1, O2 and O3 were conducted at the start of the data collection. O4,
O5 and O6 were collected at the end of the data collection phase. Comparing the
data presented in both the tables it is clear that there are reliability and accuracy
issues to be considered in the data notation phase of the methodology. Intra-operator
reliability over time was substantially superior to inter-operator reliability.
Lance Wicks – ED30391 - 30/39
Future research:
This study has identified that attack rate can be used to make gross predictions about
the outcome of a Judo fight, further research is required to confirm this. Further
research should also include examination of attack rate over time and the relationship
to tactics and strategy in Judo fights. Specifically, researchis indicated that examines
how attack rates alter based on which player leads on points during a match.
Research needs to identify if a Judo player once establishing a lead on points has a
change in attack rate and if their opponent increases or decreases their attack rate.
Related research is also needed to examine the relationship between the awarding of
penalties by the referee and the effect on attack rates of both players.
Given the importance of penalties in Judo fights, research could be conducted to
determine what percentage of Judo fights are won on penalties and not by positive
player action.
Research into the development of a standard software system to notate Judo needs
to be conducted. The existing work of Hughes, McDonald and Michel (1988; 2006;
2008) could along with the software created for this study be examined to define the
important features required.
Perhaps the most important area of research required is in the area of performance
indicators for Judo. Outside of the sport of Judo (as discussed in the literature review
of this study), examination of probabilities and predicting results is more common and
predictions are becoming more accurate (Clarke et al., 2008; Gray & Le, 2002;
Newton & Keller, 2005; Winkler, 1971; Yelas & Clarke, 2004).
To create an accurate model for predicting the outcome of Judo fights, research is
required into the performance indicators that predict victory in Judo. This study
suggests that attack rate is a gross indicator, further research looking at Judo more
finely may provide more fine predictions. For example, McDonald (2006) interviewed
a number of world leading Judo coaches and identified a number of elements of the
Judo fight that they believed were important (for example who establishes the
Lance Wicks – ED30391 - 31/39
dominant grip). Research into quantifying dominance of grip could be used to then
study Judo fights and look for a correlation between grip dominance and victory (or
grip dominance and attack rate).
Research into the relationship between attack rate and awarding of penalties should
be conducted. This research could be used to test the hypothesis that penalties are
being awarded to players that do not attack frequently enough (and what “enough” is
statistically). It might also be used to explore the theory that by increasing attack rate,
players can prevent penalties and win more fights.
Further research should be conducted to establish valid criteria for defining an
“effective non-scoring attack” and “ineffective non-scoring attack”. This would reduce
the amount of subjectivity involved in collecting data in this area.
This study suggests that players who attack most in the earlier rounds may be more
likely to go on to be medallists; this hypothesis could be explored further in future
research.
Lance Wicks – ED30391 - 32/39
References:
Act, S. A. (2004). May the Best Team Win: Baseball Economics and Public Policy, by
Andrew Zimbalist (Brookings Institution Press, Washington DC, 2003), pp.
xviii+ 198. Baseball is an American obsession. It is widely. THE ECONOMIC
RECORD, 80(249), 258-267.
Almansba, R., Franchini, E., Sterkowicz, S., Imamura, R., Calmet, M., & Ahmaidi, S.
(2007). A comparative study of speed expressed by the number of throws
between heavier and lighter categories in judo. Science & Sports, In Press,
Corrected Proof. doi: 10.1016/j.scispo.2007.10.014.
Atkinson, G., & Nevill, A. M. (1998). Statistical methods for assessing measurement
error (reliability) in variables relevant to sports medicine. Sports medicine
(Auckland, N.Z.), 26(4), 217-38.
Bailey, M., & Clarke, S. (2006). PREDICTING THE MATCH OUTCOME IN ONE DAY
INTERNATIONAL CRICKET MATCHES, WHILE THE GAME IS IN
PROGRESS. Journal of Sports Science and Medicine. Retrieved from
http://jssm.org/vol5/n4/2/v5n4-2text.php.
Boguszewski, D. (2006). Fight Dynamics of the Double Olympic Champion in Judo
(1988, 1992). Journal of Human Kinetics volume, 16, 97-106.
Boguszewski, D., & Boguszewska, K. (2006a). Dynamics of judo contests performed
by finalists of European Championships (Rotterdam 2005). Archives of Budo,
2, 40-44.
Boguszewski, D., & Boguszewska, K. (2006b). Dynamics of judo contests performed
by finalists of European Championships (Rotterdam 2005). Archives of Budo,
2, 40-44.
Bracewell, P. (2002). Implementing Statistics in a Diagnostic Coaching Structure for
Lance Wicks – ED30391 - 33/39
Rugby. Research Letters in the Information and Mathematical Sciences, 3,
79–84.
Carmichael, F., & Thomas, D. (2005). Does the Best Team Win? An Analysis of Team
Performances at EURO 2004. football-research.bbk.ac.uk. Retrieved from
http://www.football-research.bbk.ac.uk/docs/bestteamwin.pdf.
Clarke, S. R., Bailey, M., & Yelas, S. (2008). Successful applications of statistical
modeling to betting markets. MATHEMATICS TODAY-SOUTHEND ON SEA-,
44(1), 38.
Coda by Panic. (n.d.). Coda. Panic. Retrieved from http://www.panic.com/coda/.
Conway, D. (2005). Perl Best Practices. O'Reilly Media, Inc.
Degoutte, F., Jouanel, P., & Filaire, E. (2003). Energy demands during a judo match
and recovery. Br J Sports Med, 37(3), 245-249. doi: 10.1136/bjsm.37.3.245.
Fischer, M. (2008 8). Olympic Games 2008 Overview. Retrieved May 8, 2009, from
http://www.judo-world.info/world/olympics/2008_beijing/tta.php?
tta_mode=&aktion=live&sprache=english.
Forrest, D. (2007). Anyone for Tennis (Betting)? The European Journal of Finance,
13(8), 751-768.
Franchini, E., Nunes, A. V., Moraes, J. M., & Del Vecchio, F. B. (2007). Physical
Fitness and Anthropometrical Profile of the Brazilian Male Judo Team. Journal
of PHYSIOLOGICAL ANTHROPOLOGY, 26(2), 59-67.
Franchini, E., & Boscolo Del Vecchio, F. (2007). Annals of the 5th International Judo
Federation World Research Symposium. Retrieved from
http://www.judoresearch.org/Rio2007/5Annals-final2.pdf.
Gray, S., & Le, T. (2002). How to fix a one-day international cricket match. In In
Cohen, G. (ed.), University of Technology, Sydney, Proceedings of the Sixth
Lance Wicks – ED30391 - 34/39
Australian Conference on Mathematics and Computers in Sport, Sydney,
University of Technology, 2002, p.14-31. Australia. Retrieved from
http://search.ebscohost.com/login.aspx?direct=true&db=s3h&AN=SPHS-
840934&site=ehost-live.
Gunderloy, M. (2004). Coder to Developer: Tools and Strategies for Delivering Your
Software. John Wiley & Sons.
Harville, D. (1980). Predictions for National Football League Games Via Linear-Model
Methodology. Journal of the American Statistical Association, 75(371), 516-
524.
Hausch, D. B., Ziemba, W. T., & Rubinstein, M. (1981). Efficiency of the market for
racetrack betting. Management Science, 27(12), 1435-1452.
Hicks, S. (2005). Lecture. University of Bath.
Hughes, M., Cooper, S. M., Nevill, A., & Brown, S. (2003). An example of reliability
testing and establishing performance profiles for non-parametric data from
performance analysis. eIJCSS International Journal of Computers in Sport
Science (Electronic), 2(1), 34-56.
Hughes, M. (1988). Computerized notation analysis in field games. Ergonomics,
31(11), 1585. doi: 10.1080/00140138808966808.
Hughes, M., & Franks, I. M. (2004). Notational Analysis of Sport: Systems for Better
Coaching and Performance in Sport. Routledge.
Hughes, M., & Franks, I. M. (2004). Notational Analysis of Sport: Systems for Better
Coaching and Performance in (p. 304). Routledge.
Inman, R. (2005). The Judo Handbook. Silverdale Books.
International Judo Federation. (2009 3). MODIFICATION OF OLYMPIC
QUALIFICATION SYSTEM FOR JUDO. Retrieved April 5, 2009, from
Lance Wicks – ED30391 - 35/39
http://www.intjudo.eu/?
Menu=News&Action=Full&type_id=194&id=194&lang_id=2&mid=4.
International Judo Federation - Weights. (n.d.). . Retrieved July 25, 2008, from http://
www.intjudo.eu/?
Menu=Static_Page&Action=List&m_static_id=45&lang_id=2&mid=7&main=12
#5.
International Judo Federation Rules. (2008). . Retrieved May 8, 2008, from
http://www.intjudo.eu/?
Menu=Static_Page&Action=List&m_static_id=42&lang_id=2&mid=2&main=12
#26.
Kano, J. (1994). Kodokan Judo (New edition.). New York: Kodansha International
Ltd.
Kubatko, J., Oliver, D., Pelton, K., & Rosenbaum, D. (2007). A Starting Point for
Analyzing Basketball Statistics. Journal of Quantitative Analysis in Sports,
3(3). doi: 10.2202/1559-0410.1070.
McDonald, K. (2006). Thesis. University of Bath.
michel. (2008, 02T15:18:18Z 6). FRAMI - SOFTWARE FOR ANALYSIS OF
TECHNIQUES AND TACTICS IN JUDO
(Bianca MIARKA, Michel CALMET, Emerson FRANCHINI). Retrieved April 18,
2009, from http://www.jorrescam.fr/spip.php?article40#.
Nakanishi, H. (1998). Seoi-nage. Chiron.
Nevill, A., Atkinson, G., Hughes, M., & Cooper, S. (2002). Statistical methods for
analysing discrete and categorical data recorded in performance analysis. /
Methodes statistiques pour l ' analyse discrete et par categories des donnees
Lance Wicks – ED30391 - 36/39
rassemblees pour l ' analyse de la performance. Journal of Sports Sciences,
20(10), 829-844.
Newton, P. K., & Keller, J. B. (2005). Probability of Winning at Tennis I. Theory and
Data. Studies in Applied Mathematics, 114(3), 241-269.
OpenOffice.org. (n.d.) . OpenOffice. OpenOffice.org. Retrieved from
http://www.openoffice.org.
Papahristodoulou, C. (2006). Team Performance in UEFA Champions League 2005-
06.
Papahristodoulou, C. (2007). An analysis of Champions League match statistics.
Working Paper.
Ribeiro, S. R., Tierra-Criollo, C. J., & Martins, R. (2006). Effects of different strengths
in the judo fights, muscular electrical activity and biomechanical parameters in
elite athletes. Revista Brasileira de Medicina do Esporte, 12, 27-32.
Schwarz, A. (2005). The Numbers Game: Baseball's Lifelong Fascination With
Statistics. Thomas Dunne Books.
Sikorski, W., Mickiewicz, G., Majle, B., & Laksa, C. (1987). Structure of the contest
and work capacity of the judoist. Proceedings of the International Congress on
Judo ÒContemporary Problems of Training and Judo Contest, 9-11.
Snyder, W. W. (1978). Horse racing: testing the efficient markets model. Journal of
Finance, 33(4), 1109-1118.
Sterkowicz, S., Lech, G., & Almansba, R. (2007). The course of fight and the level of
sports achievements in judo. Archives of Budo, 3, 72-81.
Sterkowicz, S., & Maslej, P. (1999). An evaluation of the technical and tactical
aspects of judo matches at the senior level.
Sugai, H. (1992). Uchimata. Ippon Books.
Lance Wicks – ED30391 - 37/39
SVN. (n.d.). SVN. Tigris. Retrieved from http://subversion.tigris.org/.
The official website of the BEIJING 2008 Olympic Games. (n.d.). . Retrieved May 4,
2009, from http://results.beijing2008.cn/WRM/ENG/Schedule/JU.shtml.
View, T. O. C. (1994). Expert prediction, symbolic learning, and neural networks.
Anexperiment on greyhound racing. Expert, IEEE [see also IEEE Intelligent
Systems and Their Applications], 9(6), 21-27.
Villamon, M., Brown, D., Espartero, J., & Gutierrez, C. (2004). Reflexive
Modernization and the Disembedding of Judo from 1946 to the 2000 Sydney
Olympics. International Review for the Sociology of Sport, 39(2), 139-156. doi:
10.1177/1012690204043458.
Wicks, L. (2006). Statistical Summary of Judo at 2006 Commonwealth Judo
Tournament. Retrieved from http://www.judocoach.com/analysis/2006-
Summary.pdf.
Winkler, R. L. (1971). Probabilistic Prediction: Some Experimental Results. Journal of
the American Statistical Association, 66(336), 675-685.
Yelas, S., & Clarke, S. (2004). Forecasting the 2003 Rugby World Cup. In In Morton,
R.H. (ed.), Massey University, Proceedings of the Seventh Australasian
Conference on Mathematics and Computers in Sport, Palmerston North, N.Z.,
Massey University, 2004, p.270-277. Australia. Retrieved from
http://search.ebscohost.com/login.aspx?direct=true&db=s3h&AN=SPHS-
965917&site=ehost-live.
Lance Wicks – ED30391 - 38/39
Appendices:
Lance Wicks – ED30391 - 39/39
notator.pl
Date: Wednesday, May 6, 2009 Page 1 of 4
use strict;
use warnings;
use English qw( -no_match_vars );
use version;
use Term::ReadKey;
ReadMode 4;
local $OUTPUT_AUTOFLUSH = 1;
# $Id$
our $VERSION = qv('0.0.1');
# ---------------------------------------------------------
#
# This file created by Lance Wicks.
# Last Modified, 18 April 2009.
#
#
# notator.pl - Notation software for BSc. project.
# Copyright (C) 2009 Lance Wicks
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as
# published by the Free Software Foundation, either version 3 of the
# License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see .
#
# -----------------------------------------------------------
# ---------------------------------------------
# Sub routine stubs
# ---------------------------------------------
sub DisplayWelcome;
sub ResetCounters;
sub PrintResults;
# -----------------------------------------------
# Global Variables
# -----------------------------------------------
my $key;
my $segments = 1;
my $events = 0;
my $blue_Attack = 0;
my $blue_EffAttack = 0;
my $blue_Koka = 0;
my $blue_Yuko = 0;
my $blue_Wazari = 0;
my $blue_Ippon = 0;
my $blue_Penalty = 0;
my $white_Attack = 0;
my $white_EffAttack = 0;
my $white_Koka = 0;
my $white_Yuko = 0;
my $white_Wazari = 0;
my $white_Ippon = 0;
my $white_Penalty = 0;
# -----------------------------------------------
# MAIN LOOP
# -----------------------------------------------
DisplayWelcome();
my $run_flag = 1;
while ($run_flag) {
while (not defined ($key = ReadKey(-1))) {
# No key yet
}
if ($key eq "q" || $key eq "Q") {
notator.pl
Date: Wednesday, May 6, 2009 Page 2 of 4
$run_flag = 0;
}
if ($key eq "f" || $key eq "F") {
$blue_Attack++;
$events++;
}
if ($key eq "d" || $key eq "D") {
$blue_EffAttack++;
$events++;
}
if ($key eq "v" || $key eq "V") {
$blue_Koka++;
$events++;
}
if ($key eq "c" || $key eq "C") {
$blue_Yuko++;
$events++;
}
if ($key eq "x" || $key eq "X") {
$blue_Wazari++;
$events++;
}
if ($key eq "z" || $key eq "Z") {
$blue_Ippon++;
$events++;
}
if ($key eq "t" || $key eq "T") {
$blue_Penalty++;
$events++;
}
if ($key eq "j" || $key eq "J") {
$white_Attack++;
$events++;
}
if ($key eq "k" || $key eq "K") {
$white_EffAttack++;
$events++;
}
if ($key eq "n" || $key eq "N") {
$white_Koka++;
$events++;
}
if ($key eq "m" || $key eq "M") {
$white_Yuko++;
$events++;
}
if ($key eq "," || $key eq "<") {
$white_Wazari++;
$events++;
}
if ($key eq "." || $key eq ">") {
$white_Ippon++;
$events++;
}
if ($key eq "u" || $key eq "U") {
$white_Penalty++;
$events++;
}
if ($key eq " ") {
$segments++;
$events++;
}
notator.pl
Date: Wednesday, May 6, 2009 Page 3 of 4
if ($events >= 10 ){
print "\b\b";
} else {
print "\b";
}
print "$events";
}
ReadMode 0;
PrintResults();
# -----------------------------------------------
# Sub Routines
# -----------------------------------------------
sub DisplayWelcome {
print "\n\nJudo Notator ($VERSION)\n\n";
print "-------------------------------------------------------------------------------\n";
print "| BLUE | WHITE |\n";
print "| F = Attack | J = Attack |\n";
print "| D = Effective Attack | J = Effective Attack |\n";
print "| | |\n";
print "| V = Koka | N = Koka |\n";
print "| C = Yoka | M = Yoka |\n";
print "| X = Wazari | < = Wazari |\n";
print "| Z = Ippon | > = Ippon |\n";
print "| T = Receive Penalty | U = Receive Penalty |\n";
print "| | |\n";
print "| SPACE = MATTE |\n";
print "| | |\n";
print "| Q = SOREMADE |\n";
print "-------------------------------------------------------------------------------\n";
print " Events: ";
}
sub ResetCounters{
$segments = 1;
$blue_Attack = 0;
$blue_EffAttack = 0;
$blue_Koka = 0;
$blue_Yuko = 0;
$blue_Wazari = 0;
$blue_Ippon = 0;
$blue_Penalty = 0;
$white_Attack = 0;
$white_EffAttack = 0;
$white_Koka = 0;
$white_Yuko = 0;
$white_Wazari = 0;
$white_Ippon = 0;
$white_Penalty = 0;
$events = 0;
}
sub PrintResults {
print "\n\n";
my @months = qw(Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec);
my @weekDays = qw(Sun Mon Tue Wed Thu Fri Sat Sun);
my ($second, $minute, $hour, $dayOfMonth, $month, $yearOffset, $dayOfWeek, $dayOfYear,
$daylightSavings) = localtime();
my $year = 1900 + $yearOffset;
my $theTime = "$hour:$minute:$second, $weekDays[$dayOfWeek] $months[$month] $dayOfMonth,
$year";
print "Notation Time and Date\n$theTime\n\n";
print "Segments: $segments\n";
print "\nBLUE\n";
print "Ineffective Attacks: $blue_Attack\n";
print "Effective Attacks: $blue_EffAttack\n";
print "Koka: $blue_Koka\n";
print "Yuka: $blue_Yuko\n";
notator.pl
Date: Wednesday, May 6, 2009 Page 4 of 4
print "Wazari: $blue_Wazari\n";
print "Ippon: $blue_Ippon\n";
print "Penalty: $blue_Penalty\n";
print "\nWHITE\n";
print "Ineffective Attacks: $white_Attack\n";
print "Effective Attacks: $white_EffAttack\n";
print "Koka: $white_Koka\n";
print "Yuka: $white_Yuko\n";
print "Wazari: $white_Wazari\n";
print "Ippon: $white_Ippon\n";
print "Penalty: $white_Penalty\n\n\n";
}