EmRatThich Table Tennis World Ranking System

ITTF rating - world ranking table tennis by EmRatThich

Why we need another table tennis ranking system? Why don’t you just use the ITTF ratings?

Because the current ITTF World Ranking has many flaws. It’s not correct!

The purpose of rating system

The rating system is a method for calculating the relative skill levels of players in competitor-versus-competitor games such as sports, chess, table tennis, tennis, baskets, computer games, personnel management, and more.

So, “Rating = Skill Level”

How to measure “skills” in table tennis

According to the “Rating Central System” of Dr. David J. Marcus, Ph.D., the skills are considered as “playing strength”.

The higher ranked player will have a higher chance to win the competition.

typical laws of rating system: Better player has a higher chance to win (from Rating Central)
typical laws of rating system: Better player has a higher chance to win (from Rating Central)

The upset can also be determined. The better ranking system will predict correctly the probability of the upset.

So the laws is:

“There is a 95% probability that the player’s playing strength is between the mean minus twice the standard deviation and the mean plus twice the standard deviation.”

“The probability that a match will be an upset is determined solely by the difference in playing strengths of the two players. The larger the difference in the playing strengths, the more likely it is that the stronger player will win.

playing strength rating system: Upset and the probability of winning
playing strength rating system: Upset and the probability of winning

“The purpose of rating system is to seed the better player (higher skilled player)”

So in table tennis, the rating system helps you to determine who will have higher chance to win. The best player has the highest ranking.

The problem of the new ITTF rating system

Basic principles of the New Ranking System (Taken directly from ITTF.com):

– The new ranking is based on the final positions reached at the tournaments only (regardless the relative strength of the opponent player).
Only bonus points are awarded (no rating points are given), which are valid for 12 months, except the World Championships.
NO loss of points in case of losing a match.

This new rating system is totally ridiculous!

The new world ranking data of ITTF only consider the recent results, and with the wrong multiplying weightings factor. The new ITTF Table Tennis ranking introduces many problems:

  • Many real top players will drop out of the top 100 best table tennis players. For example, many top players dropped out of the current Men Ranking (like Zhang Jike, Jun Mizutani, etc).
  • Some weaker players now jump into the top ranking due to the recent activities. For example, with the New World Ranking System makes Kanak Jha (USA Young Table Tennis Player) into Top 100.
  • This shows who’s the most active on the pro tour instead of who’s best. The official ITTF ranking system will force players to play more, pay more, but less quality. The seeding is a mess. The best player will be ranked lower.
current ITTF world ranking 01 2018
current ITTF world ranking 01 2018

Look at this recent official table tennis Men World Ranking.

  • Ovtcharov is ranked number 1, with 16 000 points. Which is 66% higher than Fang Bo (only 9000 points). Based on the Distribution Laws, Ovtcharov will have about 85% probability to win Fang Bo. Which is false! Ask Ovtcharov himself!
  • “How do you feel the confidence to win Fang Bo”? Of course, it’s not 85% to win!
  • Ma Long has only 13000 points, and ranked 7. So his winning probability to Ovtcharov, Fan Zhendong is only about 15%. Ridiculous! The rating system of ITTF is false!
  • Fang Bo is ranked only 21, but look at the real results! He can easily win many higher ranked players. He has won, without any loss to many higher ranked players like: Assar Omar (ranked 16), Quadri, Yuya, Hugo, Simon, Kenta.

Using this ranking system is dangerous

Why this ranking system is dangerous. Because, for the next event like Olympics, or World Championships, the top players are seedings. But the seed won’t reflect the “real best players”. Many players will try to participate many tournaments to get the “fake ranking”. There is the mess in the seedings. And it’s not fun at all. Top players need to be super-active or they are kicked out of the ranking.

With the new system: Best players is ranked very low (Timo Boll)
With the new system: Best players are ranked very low (Timo Boll)

What does the ITTF world ranking stand for?

The so-called ITTF World Ranking is not the real World Ranking. You should understand the Official ITTF ranking as:

“ITTF Ranking = The ranking for the most active player recently

It’s NOT the ranking of the top table tennis players in the world.

EmRatThich Table Tennis World Ranking

The Motivation – Why I need to spend time to make my own ranking

Because you can no longer use the official ITTF rating system to evaluate who is the best table tennis player. I decided to make my own ranking system.

I searched on the Internet, and found that the most accurate “Rating System” which has been developed by David J. Marcus, Ph.D., and Sean P. O’Neill (Rating Centrals). Unfortunately, the author didn’t update the latest results due to some difficulties.

Rating Central Ranking - Latest Result only for June 2017
Rating Central Ranking – Latest Result only for June 2017

So I decided to make my own ranking system with the latest data from ITTF.

There are many advantages:

1. Self-correcting rating system, based on ELO rating system

ELO rating system is widely used in the world, not only for chess players, but also for many sports competition. The current ITTF rating use “NO loss of points in case of losing a match” which is completely wrong! So the points will go up constantly during the time, and it’s called as “rating inflation”.

elo rating system
elo rating system

No losing-point is a very bad idea. Because with this, the ITTF rating system is not the “self-correcting” system. You just play as many tournaments as possible. No matter your skills, your winning percentage, just play and win points. Even if you lose all of your matches to the lower ranked players, don’t worry! Just abuse and play!

Meanwhile, if you are the best players! If you don’t play enough. You are out of the game.

With ITTF system, it’s impossible to compare the rating between the current players and the player in the history. It’s the very bad design. While in every rating system in another sport, they try to fight and balance the “Ratings inflation and deflation”.

rating inflation
rating inflation

“For example, if there is inflation, a modern rating of 2500 means less than a historical rating of 2500, while the reverse is true if there is deflation. Using ratings to compare players between different eras is made more difficult when inflation or deflation are present.”

old ITTF rating system
old ITTF rating system

The old ITTF rating system is better. The points reflect the true player skills.

ZoomTT.com has published an article to compare the old and the new ITTF ranking system. Obviously, the new ITTF rating system failed to predict the results. 66% vs 75%.

Prediction between Old and New ITTF rating system
Prediction between Old and New ITTF rating system

2. Most accurate distribution model

ITTF Rating system didn’t use the correct distribution model. That’s why in the current ranking, Ovtcharov has a very high point due to his recent activities, meanwhile, the same level players have such a lower points.

Many other rating systems used the distribution model to distribute the reward points. My rating system used the “logistic distribution” which is considered the most accurate model.

Most accurate distribution model
Most accurate distribution model

Logistic distribution model, which provide a better fit for the actual results achieved, especially for lower ranked players.

Top 100 Rating Distribution from OLD ITTF Rating
Top 100 Rating Distribution from OLD ITTF Rating

This is the rating distribution of the top 100 Men table tennis player in the world. (Rating from the OLD ITTF rating system November 2017). The rating is from 2200 (Omar Assar) to 3300 (Ma Long). The X-axis is: Ranking position from 1 to 100. The Y-axis is the correspondant rating points.

 logistic distribution calculated by EmRatThich
logistic distribution calculated by EmRatThich

This is an example of the Logistic Distribution which reflects better the world ranking system. The adjusted parameter can be used to fit the real ranking system.

ranking distribution in Chess
ranking distribution in Chess

This is the ranking distribution for the Chess player. You can observe the same tendance. Which shows that my ranking system uses the better distribution than the new ITTF rating system.

logistic distribution in ELO
logistic distribution is also used in ELO Ranking system

3. Self-adjusted model

Attribute the correct rating points to the player requires a lot of work. It’s not as simple to “accumulate the recent points” like in ITTF model.

I tend to build the system which can self-adjust to fit the player performance. My system is designed to correctly reflect the players “playing skills” and “performance”.

performance rating
performance rating

I’ve implemented the adjusted distribution method to evaluate the performance ratings.

performance rating in real situation (Personal Management)
performance rating in the real situation (Personnel Management)

This kind of distribution model is very effective, which is used in Personnel Management in the big company.

4. Different weighting factors

Why we need the Weighting factors for the tournament?

To attribute correctly the rating points, you need to consider several weighting factors. The first one is the “Tournament Factor”.

Just because winning a match in the Olympics is much more important than winning a match in the local, or the Open tournament.

Remember: To become a grand slam in table tennis, you need to win “Grand Slam” = Olympic Games + World Cup + World Championships for Table Tennis Tournament!

Weighting factors for the tournament have been used widely, like in ATP (Tennis World Ranking) and other sports. It’s called as “Point Distribution for Event”

Point Distribution for ATP (Tennis) More reward for more important event
Point Distribution for ATP (Tennis) More reward for more important event

So here, I divide the tournament into several tiers: (here is the code, which will be updated constantly)

Olympics Games versus another Tournament
Olympics Games versus another Tournament

Tier 1: The most important tournament for a table tennis career

  • Olympic Games: The most important tournament in the history
  • World Table Tennis Championships
  • Men’s World Cup

Tier 2: The elite prestigious tournaments,

  • ITTF World Tour (including Grand Finals)
  • Challenge Series
  • ITTF World Junior Circuit (Junior & Cadet events including WJC Final)
  • ITTF World Cadet Challenge
  • Asian Cup, Asian Championship, European Championship

Tier 3: Regional and Open Competition, less difficult level

  • Regional Events
  • Multi-Sport Competitions
  • Challenge Series
  • ITTF Show-case events
  • Continental Cups
  • Africa Cup, Panam Cup, Oceania Cup

The top players will play his best at the most important event. So the more important event, the score reflects better the true level skills.

Ma Long won’t lose to any foreigner at the Olympics Game. But he can lose the same player in the Open Tournament.

5. Lucky and True Skill in table tennis

The second weighting factors should be considered is the “score factors”. EmRatThich World Table Tennis Ranking is currently the only one which considers the “luckiness” and the “skill factors” into the rating systems.

ITTF current rating system only considers the match results, for example, player A win so he gets the points. For ITTF, player A won player B 4-0 or 4-1 or 4-3, all are the same. ITTF only counts the “match results”.

Remember: Winning 4-0 and 4-3 is not the same!

But for me, winning 4-0 and winning 4-3 is totally different. By winning 4-0, you dominate the game, and your skill factors increase. If you win 4-3, the “lucky factor” will play an important role in the match results.

So I have implemented the “Lucky factor” into my rating system:

score is not the same
Match Score is not the same

Note that: This true-score-factor will be updated to reflect the real world ranking system. This is my first attempt, and it works very well.

World Table Tennis Ranking January 2018

So what’s is the result? Who is the top 20 table tennis Men players in 2018?

By analyzing:

  • 43 735 table tennis matches played in 2017 (in official ITTF events)
  • nearly 100 table tennis international tournaments during 2017
43 thousands table tennis matches played in ITTF tournaments
43 thousand table tennis matches played in ITTF tournaments
Nearly 100 tournaments organized by ITTF in 2017
Nearly 100 tournaments organized by ITTF in 2017

And by applying and fitting the most accurate distribution model

ELO distribution
ELO distribution

Here is the Current Table Tennis Men’s World Ranking, as of January 2018 by EmRatThich

The recent world ranking of top 20 table tennis players in 2018, rating calculated by EmRatThich.

1FAN Zhendong (CHN)3182.47
2MA Long (CHN)3158.41
3OVTCHAROV Dimitrij (GER)3044.81
4BOLL Timo (GER)2945.38
5LIN Gaoyuan (CHN)2720.67
6XU Xin (CHN)2661.3
7HARIMOTO Tomokazu (JPN)2507.19
8MIZUTANI Jun (JPN)2428.72
9YAN An (CHN)2398.9
10FANG Bo (CHN)2385.39
11NIWA Koki (JPN)2361.72
12FREITAS Marcos (POR)2361.01
13MATSUDAIRA Kenta (JPN)2313.47
14LEE Sangsu (KOR)2293.41
15GAUZY Simon (FRA)2288.53
16SAMSONOV Vladimir (BLR)2263.98
17WONG Chun Ting (HKG)2193.25
18ARUNA Quadri (NGR)2157.05
19FILUS Ruwen (GER)2100.82
20CALDERANO Hugo (BRA)2092.93
2018 01 emratthich ranking
2018 01 most accurate world table tennis ranking by EmRatThich

Please share and rating this article. 1 Star2 Stars3 Stars4 Stars5 Stars (2 votes, average: 5.00 out of 5)

Loading...

Spread the love
  • 54
  •  
  •  
  •  
  •  
    54
    Shares

1 Comment

Leave a Reply

Your email address will not be published.