The ITTF used the ELO rating until the end of 2017. Starting 2018, they've moved to the merit-based ranking, which is used by ATP(1973)/WTA(1975) and BWF(2007).
Timeline of the events.
Talks of the ranking system reform began as early as 2014. Even before that, there had been an ongoing dialog in which the Athletes' Commission proposed for a further reduction in points deducted for a loss. The ITTF also wanted to encourage players to participate in the World Tour, in order to attract more sponsorships. .
Very informative.
Before I criticize the current ranking system, I want to thank all sponsors and those involved with obtaining sponsorship. Even if it is 100% clear that the current ranking system is more mathematically flawed than other ranking or rating systems, if the system is responsible for increased prizes and increased popularity, then it should stay in place.
Here are some problems:
If two players in the main draw of a world tour event both win at least one match and then are eliminated in the same round, they will receive an equal number of ranking points, even if the average rankings of one players opponents was much higher than the other players. That isn't fair. The player with the harder draw should be compensated accordingly. Also, players should lose points when they lose matches.
To be fair, there is big problem with the ELO rating system. It has to do with different pools of players. As an example, I knew a man who established a 2100 USCF (United States Chess Federation) rating and kept it for nearly a decade. He played in a big city which had a large pool of players; many local and many from other parts of the country and other parts of the world. He moved to a much smaller city that had a small pool of mostly local players. He was of the strongest and most active. Within a year, he raised he rating to over 2300, without improving his skills.
The funniest example of deliberate chess ratings manipulation was from a man named Claude Bloodgood, who was incarcerated for life for having killed his mother. He apparently had some funds and registered a number of fellow prisoners for USCF memberships. He organized sanctioned tournaments and bribed some players to deliberately lose to a few select individuals, in order to artificially inflate their ratings. He then played thousands of games with the higher rated players and won them all. By 1996, Claude achieved a rating of 2759, which was the second highest in the country! He was technically eligible to play in the next U.S. Championship, but wasn't invited for obvious reasons.
I can't see too much of this nonsense occurring in the Table Tennis world, but there can be a danger of inflated ELO ratings in some regions if they have small player pools and enough tournaments are sanctioned. This could potentially adversely affect tournaments where eligibility and seeding is based on ratings.
I'm ending my post/rant with more criticism of the current ITTF ranking system: For the first two months of 2018, Ding Ning was ranked #21. That was absurd - she was clearly at least a top 5 player.