When do people score in Judo?

Below is a screen shot that shows some very basic information on the heavyweight divisions at the 2012 European Junior (under 20 years) Championships, held recently in Porec, Croatia.

EC_JUN2012
EC_JUN2012 Scores

 

As you can see, the under 20 championships duration is 4 minutes. The least amount of scoring happens in the first minute, the most in the second minute of the match.

The number of Ippon throws varies minute by minute also. With the last minute of the match being an area of interest. There is proportionately more Ippon scores during the final minute than any other period in the match. The reasons for this we do not know. And this is from a single event.

We can also see that there were 512 scores made in this incomplete dataset, of which Ippon made up approximately 22% of the scores.

As with all statistics, the size of the sample set is important. In this case it is extremely small. This is a dataset collected with a new experimental solution and was only used on one day of the event. We need to consider a much larger sample before putting much weight in these figures; I put them here purely to invite discussion.

 

The +78kg Category and Tong Wen.

On the 7th of May 2010, the qualification period for the Judo event at the London 2012 Olympic Games started. More than 4,000 athletes from 153 countries have been chasing the precious few Olympic qualification places.

On May 7th, I started my research project looking at an alternative athlete ranking system for Judo.

The IJF has a qualification system that has been in use for many years, initially just within the EJU but now for the first time as the method of qualifying for the Olympic Games. The system I have been experimenting with has been running in parallel to the IJF system quietly collecting the match results from the 75 qualification events held so far.

The two systems are different at a fundamental level. The IJF reward points to athletes based on the position they reach in the tournament. My system adds or subtracts points after each match. In the IJF system the points awarded are based on the event and the position you reach. My system awards points based on the calculated probability of a player beating their opponent. The idea being that if you beat the best in the world and come first; it should be worth more than beating weaker players and coming first.

One of the categories I have found very interesting is the +78kg category, in part because I am a fan of British fighter Karina Bryant. It is also the smallest category in elite Judo (only 146 athletes). This means it is comparatively easy to watch the category. Also, this category features the infamous Tong Wen of China. Tong Wen (佟文) did not compete for two years after running into trouble with the banned substance clenbuterol; before being cleared and re-entering competition last year at the 2011 Moscow Grand Slam.

Tong Wen is the reigning Olympic Champion and has 7 world gold medals (according to http://www.judoinside.com/judoka/view/12071/wen_tong/ ). So an athlete of this calibre re entering the category late made it very interesting.

From the perspective of my research it was equally interesting as after her first event, my system had her ranked at (if memory serves) #2 or #3 in the category. Where as the IJF system did not have her in the top ten. The reason being that the IJF system only factored in the points from winning that event. Whereas my system took into consideration the fact she beat KONITZ Franziska, POLAVDER Lucija, KOCATURK Gulsah and KIM Na-Young that day. She has fought a total of 21 times and won each one with Ippon (I am excluding the recent default to team mate QIN Qian at the 2012 IJF Masters  in Almaty. She is (I believe), the only athlete undefeated in the category.

Yet, Tong Wen is not ranked number one under the IJF system, only third, behind TACHIMOTO Megumi (who she beat at the 2011 World Championships) and QIN Qian (who she has beaten three of the three times they have actually fought). As you can probably guess, my http://rwjl.net system ranks Tong Wen #1 above TACHIMOTO Megumi (#2) and QIN Qian (#3).

I think it is fascinating to compare the systems and this category is proving very interesting to follow.

Once Qualification closes on May 1st 2012, I shall write a report on the project and I would appreciate any and all feedback on the systems and how they compare from whatever perspectives you might have. Be that a player, a coach, a statistician, and administrator or a interested fan.

Please do ask any questions or send your feedback to me at lw@judocoach.com

 

Lance

London2012 Qualification and a comment on typos.

As many know, for the past few years I have conducting research on the ranking of elite level Judo. Specifically, I have been running and experimental ranking system for Olympic qualification using the ELO ranking system and all the IJF ranking events.

To date (not including the 2011 Tokyo Grand Slam which is happening whilst I write this), there have been 17,542 fights in the qualification period. The qualification period started in the 2010 Tunis Grand Prix and there have been 64 events so far with approximately 150 nations and approximately 3,781 athletes all in the mix.

You’ll notice I said “approximately” there. That was intentional. Since the qualification started I have identified over 300 typographical mistakes in the names of athletes reported by the tournaments. Thats approximately 11% of error. Now, my data is sourced from Ippon.org and that technically is not the definitive record. However, it is generated directly from the tournament software used in the stadium, so I can imagine that the errors carry over to the official written records.

The IJF’s own ranking system is I think accurate (at least as far as I can see) this is mainly due to the diligence of the hard working lady who looks after the list. The IJF ranking list is maintained manually and it’s hard to fathom quite how much work goes into maintaining it! I know from my experiences running my experiment ( on http://www.rwjl.net by the way), that maintaining accurate information is very hard. I am lucky to have had lots of help from people, not least of all from http://www.judoinside.com

In my experimental ranking system for Judo, accuracy is vital. My system awards or deducts ranking points from athletes after every fight. Unlike the IJF ranking system, my system looks at every fight; the IJF look only at the final positions. So the IJF system requires less accuracy arguably, as you need only make sure the final results are right. After all, 50% of the names are gone after the first round at most events so need not be entered into the IJF system. On my system however, those names need to be right.

The reason they need to be rights is that points are awarded based on the predicted chance of player a beating player b in that specific match. As I calculate that chance for each fight, it matters as the ranking points are awarded after each match. Lets say you have a “Joe Bloggs” who wins points in competition A, getting a ranking of perhaps 1580. If they are mistakenly called “Joseph Bloggs” in competition B, then my system thinks they are a new person and there 1580 points are not applied and a new athlete enters the system.

This is bad as I would earn more points at competition B for beating a player with an existing rank of 1580 (Joe Bloggs) than a new athlete (Joseph Bloggs) who would start with 1500 points. In the IJF system, if Ilias Illiadis beat Joe Bloggs in round one it matters not at all if his name is wrong; because who you beat does not matter, only the position you get to. If Joe Bloggs makes it far enough to earn points, then there name being wrong would matter. However, as the IJF system is manual it’s perhaps going to get picked up at that point.

It is thanks to “encouragement” from others that I have put more and more effort into the accuracy of the names. I have about 330 corrections and as it’s automated so it should stay right. Of course, there is another problem which is one I have not yet attacked. What to do about athletes who have the same exact name! It is difficult to find workable solutions to that problem. How do you decide if two names that are identical are not the same person? How do you decide two identical names are not to different people?

As the software I have written matures I hope to find good solutions these sorts of problems. For now I have 330 specific corrections;I have found by manually going through the data and fixing mistakes people point out (please do check http://rwjl.net and see if you find any mistakes).

My future plans are to publish a final report on the experiment after the London2012 Olympic games, this will go along with the poster presentations I have given at the European and World Judo Championships. I also plan to open source the software and data, so that others can improve on what I have done and take the research further or in different directions. I have already created a data API, so that people that want it can easily obtain machine readable access to my record of the 17,542 fights that make up the qualification race for London2012. If you would like access please let me know.

Regards,

 

Lance Wicks.

An experimental relative skill based ranking system for elite level Judo.

This week I am in Istanbul at the European Judo Union Championships where, along with other things, I am attending the EJU Research Symposium.

I am presenting a poster on the research I have been conducting on a different way of ranking Judo athletes who compete on the world circuit. Below is the poster:

An experimental relative skill based ranking system for elite level Judo

The summary of the research was very clearly summarised by Dr. AnnMaria DeMars on Twitter as being:

“Getting 3rd on a bye & 1 win != beating the current world & European champs to get 3rd”

The system I am experimenting is based on the Elo system that is used in Chess and various other sports. The concept being that we award ranking points to athletes based on who they beat on a match by match basis as opposed to awarding points based on what position in the medal table they reach.

I am really hoping to get some great constructive criticism from the european Judo research community as I hope to present the research at the IAJR Symposium in Paris later this year.

 

Have the new 2010 rules affected Judo?

Nage pointThis past weekend we entered the official qualification period for the London 2012 Olympic games. In this qualification period the new rule changes are in effect (and are not supposed to change) and it is worth taking a brief look to see if the changes are having an effect on the Judo being played. In this post, we shall compare data from the 2009 Moscow Grand Prix and last weekend’s 2010 Tunis Grand Prix.

If you recall, last year we looked at the 2009 Moscow Grand Prix on this blog ( http://judometrics.com/2009/06/moscow-judo-grand-slam-2009-data/ ) using data from www.ippon.org on the scores and penalties awarded in that event. In this post we shall look at the same data, from the same source, from the 2010 Tunis Grand Prix.

In 2009 we found that Ippon made up 77% of the throwing scores in male Judo and 39% of the female throws. In 2010 we find that Ippon makes up 75% of Male throws and 55% of female throwing scores. Thats a decreaqse of 2.40% in men, but a 15.44% increase in Ippon scoring throws in womens Judo.

Within mens Judo we se a 8.10% decrease in Wazari scoring throws and a 10.49% increase in Yuko scoring throws. On the opposite side, womens Judo has decreased the number of Warazi (9.74% decrease) and Yuko (5.70% decrease) throwing scores.

Looking at the number of penalties per match, we see that in 2009 it was 1.61 penalties per match and 1.84 in 2010 for men. The women had 1.39 penalties per match in 2009 and 1.51 penalties per match in 2010. Looking at the scores per match, they have changed very little between 2009 and 2010 (.56->.59 for men, .78->.79 for women). If we look at direct Hansoku Make penalties, unsuprisingly we see the men go from 1 in 2009 to 11 in 2010. The females also increased their Hansoku Make from 0 in 2009 to 2

So to summarise, in men’s Judo we have fewer Ippon and more penalties. Women’s Judo on the otherhand has more Ippon throws and more penalties. BOth men and women have an increase in Hansoku Make, which we can presumably associate with the new leg grab rule.

This very brief look at the data gives only a snapshot of elite Judo from two events; two events in Europe. So we can’t and shouldn’t read too much into the numbers. This data does not look at the style of Judo on display. It is perfectly possible that the Judo shown in 2010 is more exciting and “pure Judo”.

However, these numbers do provide a view of Judo that we need to consider. My understanding was that the rule changes were devised to increase/promote throwing Judo. “Pure Judo” as it was phrased. My interpretation is that this has arguably been the case in Women’s Judo a 15.44% increase in the Ippon statistic is very positive.  But the general trend towards even more penalties is a negative; this might be a result of the rule changes, or different fighting styles; we can’t tell from this examination.

If we look broadly, the number of scores per match has increased in both men and women, as has  the number of scores from throws; from 1.61 to 1.86 per match for men and 1.67 to 1.86 per match for women. However, the number of scores per match from throws has only increased from .56 to .59 for men and .78 to .79 for women.

More investigation is required, an investigation into the style of Judo displayed would be interesting; as would a long-term study of the objective data over a wide range of events.

How do you feel the Judo from 2009 and 2010 has changed? How do you feel the rule changes have affected Judo? How do you think we can further explore the area?

Research Poster Presentation at the European Judo Championships in Vienna

The European Judo Union Education & Teaching Commission is pleased to announce the POSTER PRESENTATION OF RESEARCH.

The presentation will be held prior to the 2010 Individual and Team European Judo Championships in Vienna, at the 21 of April at the Ferry-Dusika Stadion on Wednesday.

This is your chance to present your programs at the Poster Presentation of research. All sessions will consist of a presentation of scholarly works related to judo. Presentations may involve any scholarly work related to any aspect of judo. Areas are not limited to topics related to the sport aspect of judo, but can also include exercise physiology, strength and conditioning, sport psychology, injury rehabilitation, rest and recovery, nutrition or other topics.

Topics may also include any area of judo history, philosophy, culture, or other values. The audience will be mixed and will include academics, coaches, officials, teachers and the broad judo family.
Presenters, please bear in mind when communicating your research.

Proposals for the Poster presentation will be considered for the program if received in time. The final schedule shall include proposals accepted for presentation as well as an invited program.

The European Judo Union is pleased to announce that it will present awards to at least three outstanding posters based on originality, methodology quality, clarity, and contribution to judo: the first classified poster will be awarded with 1.500 euro, the second with 700 euro and the third with 300 euro.

The ages of Olympic Grapplers (Judo and Freestyle Wrestling).

www.army.milIn this article we look at the ages of Judo athletes and of Freestyle Wrestlers at the Olympic level, this follows on from the article back in March (2009) on the “Ages of medalists at 2009 Judo World Cup events.” in which the age of Judo athletes competing at a high level was briefly examined.  In this article we look at ages of athletes in more detail and compare Judo against our Olympic cousins Freestyle Wrestlers.

In March the average age of a medalist was identified as being 24.95, looking at the larger pool of athletes in the 2008, 2004, 2000 and 1996 Olympic games the average age of a medalist in Judo is 25.54 for men and 24.99 for women.

Judo Gold medalists are on average 25.07. The other Judo medalists are on average 25.27. If we compare this to Freestyle Wrestling where the average age of athletes is 25.41 and gold medalists are on average 23.83 years of age.

What does this tell us?

One interpretation of this information is that grapplers at the Olympic level are in their mid twenties.

If we believe the theory that it takes 10 years to reach the elite level (as popularized by Malcom Gladwell’s Outliers book – UK/US) then we need to start working with young Judo athletes of 14/15 years. Obviously a 14 year old is not (normally) fully mature physically, so a long-term strategy needs to be implemented to bring the young athlete to the Olympic stage.

Having said that, if we look at the average age of Judo athletes over the 4 Olympic (1996-2008), we see the average age has actually crept up over this time from 24.96 to 26.14 years. The gap between gold and the average has changed too, from about three months in 1996 to 1 year in 2008. So perhaps the best players (the gold medalists) is not going up as fast as for the rest of the players? A simple examination does not make this clear, especially as the sample size of gold medalists is of course much smaller than the general population.

This is a small example of the sort of information this data can tell us. If you are a national governing body, do you invest your limited resources in athletes who will be outside of their mid-twenties? Do you target 14 year old athletes? This data can’t answer those questions, but with better research perhaps you could find the objective information to help make decisions in this area?

I shall look at the data in more detail but perhaps you’d rather do it for yourself?
I have uploaded my raw data at http://spreadsheets.google.com/ccc?key=0AqIP2spU5FrFdFR5UFdTUzdmaHVfc0FvUzZpLUR4YXc&hl=en

This  simple examination of Judo data is an example of the type of activities coaches on the EJU coaching certification courses do. It is also an examination of the type that can lead to more through research which could lead you to the International Association of Researchers.

—-

Data was collected from http://www.sports-reference.com/olympics

A case for experience… Athletes who competed in 2005 and 2007 World Judo Championships.

Within the Judo coaching community we often argue about the value of experience, about the value of attending the “big” events even if you don’t win. So this week I decided to look at the athletes who competed in the 2005 World Judo Championships in Cairo and also in the 2007 World Championships in Rio de Janeiro. It is a simple look at the -81kg category across a mere two events, so it is only a very basic examination. The results are of interest and it would be worth researching properly.

Of the 58 athletes who fought in the -81kg category in Cairo, 21 (36%) returned to the -81kg category in 2007.

Looking at the top 8 players, 3 (37.5%) returned to the category in 2007; including the Gold and one of the Bronze medalists.

Seeding:

By examining the seeding position for the 21 athletes we see that 14 (66.6%) of the returning athletes came back seeded higher than in 2005.

Position in tournament:

Of the 21 returning athletes, 15 (71%) reached a later round of the competition in 2007 than they did in 2005. Three (21%) athletes reached the same point in the event and four (19%) did worse in 2007 than in 2005.

What does this suggest?

Obviously, very little examination has been done. The point of this article is only to start a train of thought, not to prove a hypothesis, so please take the following as ideas only, not scientific proof.

What the information suggests to me that there is some truth to the idea that attending the World Champs is a sensible decision in the long term planning of a Judo athlete. It suggests to me that you need to go to the world champs more than once and that your performance should improve at the next appearance. From this data we see this is right 71% of the time.

It is perhaps interesting to consider your athletes position in the world order, what if for example they lose in the first round? Well looking at the 6 athletes (from the returning athletes) that went out in the first round in Cairo, only one failed to improve upon this in Rio. 3 made it to the second round, 2 made it to the 3rd round and 1 made it to the 4th round. So perhaps sending inexperienced players who lose in the first round is not pointless as some suggest. Especially as those who made 2nd round in 2005 also tended to improve, as did those who made the third round.

Further Research:

For performance directors and national coaches making hard decisions about who to send to the World Championships, then more research might help determine an evidence based reason for sending or not sending athletes. It would be interesting to repeat this simple examination across all the categories and see if the results remain similar. Also another extension to this would be to look across earlier World Championships.

I would very much like to know for example how many World Championships athletes attend on average. How they improve across events; can an athlete go from losing in the first round to winning the event over a number of events? Or, is there a minimum level you need to make on your first appearance to get a medal in a later event? Do nationalities affect these figures?

If we are able to identify a consistent progression, can we manage our performance/medal target “pipeline” by looking at what we get today?

Tre Torri Judo Tournament Corridonia – 2009 B-Tournament – Porto Sant’Elpidio, Italy (ITA)

In this post we shall use the JudoInside.com website again to explore a Judo competition and the data we are able to obtain from this freely available source. Today we shall look at the Tre Torri Tournament held recently in Italy, the event is a B tournament and was well attended and this event was choosen to look at as the British Team attended including many of it’s 2008 Beijing Olympic players.

We shall look at the men’s categories, not for any other reason than to save time, we shall not look at the female categories. Using JudoInside.com we are able to collect some data on the percentage of fights each athlete has won on an annual basis. This is data available on the JudoInside.com website on June 18th 2009, so is limited by the completeness of the information available on that site. There are some obvious anomalies  in the data, but for the purposes of this post we will just accept them.

So lets look at the raw data:

Raw data from Tre Torri Judo TOurnament 2009

From left to right what this data shows is the number of fights JudoInside.com had recorded in the “Head to Head” statistic (including fights prior to 199). This is followed by the athletes name and then by the past decade of results. The results consist of a percentage of fights won by the player according to the website. The final column is an average of all the percentages from 1999-2009. Below the main table is a summary of the average percentage for each of the medalist types and also the number of fights recorded for each medal type.

The summary information shows immediately an interesting statistic, the winners of the categories had many more fights on the system than the players they beat, and this was consistent between second and third place winners also. This might be used to support a hypothesis that Judo success requires a certain level of competition experience. That without having competed in enough events you will not win.

This is only one data point and should not be looked at in isolation.
We need to consider the level of this competition, it is a B tournament, not an elite level competition, yet we have players like former World Champion Craig Fallon, Olympian Eaun Burton and of course Tamerlan Tmenov. The long contest record of Tamerlan Tmenov for example affects the averages. Francesco Bruyere and Tamerlan Tmenov are the only two players to have a record in each year for the full decade.

So lets chart the data above:

Chart of winning percentages across past decade at Tre Torri 2009

This is a bit messy, but if you look carefully you can see some interesting information about the athletes. For example, the density is clearly higher in the last 4 years, perhaps indicating the length of careers of players in this event. Comparatively few athletes have careers extending over more than 4 or 5 years. This information might support a hypothesis that there is a optimum length of career, which might become part of a long term athlete development plan if proven. If we were to simplify the chart or chart each athlete separate out athletes you might be able to determine trends in victory percentages, which could be used to assess if an athlete is progressing well or slipping perhaps.

After doing some simple analysis like this in a spreadsheet, it might be interesting to look at a variety of charts and see if anything comes to eye. A simple way to do this is to import the data into Swivel and let it’s automated system create some visually pleasing charts for us. And we can easily see for example the following summary of 2008:

2008 by Athlete

Where we see that no athlete had an unbeaten year, and that the Italian player Giovanni Di Cristo was statistically at least, the best player in the bunch. We can also see pretty clearly in this next chart (below) which players are the most experienced.

Total Fights by Athlete

This is just a quick summary of the event and yet it gives some insights that might be relevant to a coach of B Tournament level players, or even of an elite level coach looking to learn more about the players in the level below him and how they progress upwards (and in this case downwards). It could be interesting to researchers looking to discover more about our sport.

The data is available on Swivel, so please do take a look and leave a comment on this site telling me what you discover or find interesting.

Moscow Judo Grand Slam 2009 data.

In this post I want to look briefly at the statistics from the Moscow Judo Grand Slam 2009. I am going to base it on the information gained from the great website www.ippon.org developed by Matthias Fischer. Ippon.org nicely summarises the information from events and gives some descriptive statistics for us to browse and interpret.

Number of Scores per Judo match:
The first area we shall look at is the number of scores per match. We can look at this and see that there is an average of 1.65 scores per Judo fight. Men averaged 1.63 scores per fight and women averaged 1.67 scores per fight.

Number of Ippon scores awarded:
This seems easy to identify, but is in fact slightly more complicated than the figures might initially suggest. Looking at the men’s categories half the scores are Ippon; however this includes penalty scores so rather than 90 Ippons, this means there are in fact only 77 Ippon throws in the men’s and 39 Ippon throws in the Women’s categories. That means only an average of approximately 36% of scores are from Ippon throws across the categories.

If we extend this we can remove the Shido from all the other scores also. This gives us for the men 77 Ippon Throws, 20 Wazari throws, 3 Yuko throws and no Koka throws with the new rules of course. This conveniently comes to a total of 100 throws, so we can see in the men 77% of throwing scores are Ippon. Which is great except that if we include penalties, Ippon throws are only 27% of the total scores, Penalties make up 65% percent of the scoring!

In the women there were almost the same number of throws (99), but the distribution is quite different. 39 Ippon Throws, 32 Wazari and 28 Yuko. There is quite a difference also in the penalty percentage in the Women’s categories, Penalties make up only 44% of the scores.

It would be interesting to examine the video footage and try and determine structure of men’s and women’s Judo at this event is quite so different. Perhaps it can be stylistic differences between nations and the sexes?

This is just a simple examination of the data but shows perhaps a little of what we can determine from the data generated by www.Ippon.org it is a site you should definitely checkout and keep an eye on.

Any comments or questions would be appreciated,

Lance

A site devoted to developing metrics for Judo by Lance Wicks.