Is Wei Yi Nearing a Ratings Plateau? (Probably Not)

We have been postponing writing a proper “Prodigy Profile” on Wei Yi, simply because there is so much we’d like to say that the project is too daunting! The Chinese super-prodigy has been shattering ratings records (relative to age) for a solid year and a half, and has been on our radar as a remarkable young talent for far longer than that. We still don’t feel ready to post anything final, but as we’ve been working on the project we’ve come across a few bits and pieces we just can’t wait to share.

First of all, if you haven’t been following Wei Yi’s career, here is a quick summary of some of his key achievements:

  • Fourth youngest GM in history
  • Youngest player ever rated 2600+ (at age 14.43, over six months faster than #2 Wesley So’s 14.99)
  • Youngest player ever rated 2700+ (at age 15.76, almost a year faster than #2 Magnus Carlsen’s 16.59)
  • Currently (as of the May 2015 rating list) the highest rated player ever under the age of 17, at 2718 (note that he does not even turn 16 until June 2, 2015, so he broke this record with over a year to spare!)

So the point is that he’s pretty good at this whole “chess” thing, especially for his age. In January 2015 he had perhaps his most impressive tournament performance yet, when he won first place in the Challengers division at Tata Steel (scoring 10.5/13, for a 2793 performance rating), unless you prefer his results in April 2015 at the World Team Championships where he scored 7/9 to help China win first place overall, for a 2837 performance rating.


We can see from the graph above that for the last year and a half Wei Yi has been setting a new record for “highest rating ever achieved at age X” with each new published rating. He is now ranked 33rd in the world at 2718. And just in case you didn’t hear it clearly the first time, he is NOT YET 16 YEARS OLD!

The obvious question to ask is: how good will he get? Many of the records he is breaking formerly belonged to one Magnus Carlsen, who you may have heard of. As impressive as Wei’s rise has been so far, it’s probably premature to anoint him as the definitive successor to Carlsen’s perch atop the chess world. Still, it is of course intriguing to speculate about his future. We can’t say exactly how strong he will eventually become, or how long it will take for him to climb into the top ten in the world rankings (which isn’t guaranteed… but kind of seems inevitable.) One thing we can clearly say, though, when we look closely at his results to date, is that he doesn’t appear to be approaching any kind of ratings plateau yet.

Consider the following graph, which shows his performance rating in all 21 tournaments he’s played since January 2014 (a 16 month span in which he gained 111 rating points):


We can notice, first, that in 16 of the 21 events his performance rating exceeded his actual rating at the time. This shouldn’t come as a surprise, of course. One doesn’t gain 111 rating points without a stat like that. More interesting, however, is the trend of his performance ratings. Six of his first seven tournaments in this sample, were in the 2560 to 2650 range. This alone was enough to bring his rating up from 2607 to 2638, but just as his rating started to catch up with his apparent skill level… the performances began to improve. Since July 1, 2014, Wei has played 14 events, and only once posted a performance rating lower than 2670! Those 14 tournaments (104 games in all) have been played at a net performance rating of 2745!

The obvious conclusion is that even at 2718, Wei Yi’s rating has almost certainly not yet caught up with his “true” playing strength. He has consistently been playing at higher than a 2718 level. And perhaps even more importantly, his apparent “playing strength” has been increasing even faster than his published ratings, so there is no reason to expect his ascent to slow any time soon. It would be fair to expect his rating to climb at least into the 2740+ range before he’d be in any danger of plateauing, just based on the playing strength he has already demonstrated. And if his history is any sign, by the time his rating reaches that new height he may have grown even stronger!


6 thoughts on “Is Wei Yi Nearing a Ratings Plateau? (Probably Not)

  1. Wei Yi is a freak. He is the man to watch, (but he’s still in adolescense). He is not the Karjakin who became GM at 12yrs-7mos but plateaued upon reaching ELO 2770~ at 21 years old (Jan 2011). I am not saying he is more capable than Karjakin, but that Wei’s interest on the game has remained hawkish.

    Wei Yi is in the mold of Magnus Carlsen and Wesley So whose developments were continuous, and even fast when the atmosphere was condusive.

    Magnus developmental environment was perfect —> Euro has a bounty of strong tournaments, mobility was no problem (just a few hours moving from one Euro country to the next by fast train), and communication and IT technology was advanced.

    In contrast, Wesley was hindered after reaching 2600 due to lack of strong tournaments in Philippines, mobility and communication were extremely difficult. The only thing going for Wesley was his chess laptop (but was available only when he was 10 yrs old then) which he did a good job … until now!

    Wei Yi is fortunate that strong chess engines were available when he started woodpushing. He plays long successive tournaments which indicates to me that he has been doing this from childhood — well, Wei certainly can find somebody to play chess anytime in the overpopulated continental China, and local tournaments are aplenty. One does not earn 100+ ELO in a year without packing a lot of tournament plays.

    Magnus, Wesley and Wei have trodded different paths but theirs will eventually cross. Magnus reigns for now, and the next one to two years. Wesley will start to challenge him in two years time. Wei will join and form the triumvirate in 3-4 years from now.


    • Lots of great points there, thanks for the contribution! Wei has been remarkably active, it’s true. For well over a year now he’s been consistently playing at least one (and in many cases two) tournaments almost every single month!

      At the very least, this is an additional reason to suspect that his current rating isn’t “too high” (he didn’t just get lucky in one or two events, pop over 2700, and sit on it). Proving the converse, that his current rating is actually “too low” and likely to continue increasing is harder (sample size by itself doesn’t do it), but that was the point of this article 🙂


    • This is a minor non-chess comment, but is China overpopulated? The following countries have higher population density: UK, Italy, Switzerland, Germany, Netherlands, Nigeria, Japan, India…


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s