Life In 19x19

The world chess blitz championships have just ended. As you would probably expect, the age profile of the top players is noticeably younger than in classical (i.e.. slow) chess. But the difference is actually only about 4 years. More surprisingly, I think (if you are a go player), there are only two teenagers in the top 48 in blitz, and just one in classical - and they are way down near the bottom. But there are three over-forties in the blitz list, and the classical list has not just more but several in the top dozen.

This strikes me as a sharp and surprising contrast with the professional go scene, where the teenyboppers are numerous and successful - my sense is that at the amateur level, the proportions of young players in chess and go, and their degrees of success, are somewhat similar.

A curious state of affairs. I've been racking my brains to find an explanation, but I've only come up with partial reasons. Please suggest better reasons if you can, but a couple of those I've mulled over:

1. Chess respects classical time limits rather more than go, and has fewer blitz events. I infer from another thread that my long-standing contention that Mickey Mouse time limits are not good for go in Korea is being conceded, but China has stepped back from the brink and is re-focusing on longer events - yet teenyboppers are doing well there, so the parallel with chess still fails.

2. The style of play has changed in go. In chess there is really only one way to win a game - checkmate. In go you can win either large-scale, by killing something big or amassing a huge territory, or small-scale - eking out a small territorial win. It seems from the record number of B+R/W+R results that large-scale now dominates. Maybe this is something that favours young people. If so, it may not be something inherent. It may just be that older players trained themselves to rely on endgame techniques rather than tactical sharpness (longer time limits would explain both those factors); young players just go straight for the jugular.

I hesitate over both those explanations, though. A third possibility is that intrinsically go and chess differ in some way that favours young brains. Beats me what that might be, but speculating wildly: do young people just absorb patterns and play intuitively (i.e. rather like AlphaGo) whereas older players start to think too much about the game and so lose that intuitive edge?

John Fairbairn wrote:I hesitate over both those explanations, though. A third possibility is that intrinsically go and chess differ in some way that favours young brains. Beats me what that might be, but speculating wildly: do young people just absorb patterns and play intuitively (i.e. rather like AlphaGo) whereas older players start to think too much about the game and so lose that intuitive edge?

There is some evidence in bridge that intuition does not fade much with age. Primarily intuitive players tend to maintain their strength better with age than players who primarily rely upon calculation, at least at the top levels. We certainly know in general that conscious calculation fades with age. People slow down.

John Fairbairn wrote:I infer from another thread that my long-standing contention that Mickey Mouse time limits are not good for go in Korea is being conceded, but China has stepped back from the brink and is re-focusing on longer events - yet teenyboppers are doing well there, so the parallel with chess still fails.

Fast time limits are one of many different interpretations of reasons for the state of go in Korea. There are several other variables: AlphaGo, rise in number of strong young Chinese players, cultural and political factors, etc.

You can blame everything on "Mickey Mouse time limits", but I'd say there's at least a chance that confirmation bias is coming into play.

For fun, I looked at the top 40 live chess ratings for classical, rapid and blitz. Average age: classical 31.0, rapid 31.7, blitz 31.2. So that seems within noise.

I think chess itself has recently been moving away from classical time limits and towards rapid (which I think is more the equivalent of "blitz go" than is blitz chess, where flurries of fraction-of-a-second moves frequently take place).

I agree with Bill that an intuition-driven game favors the older player, in general. Of course, due to computers, a 25-year-old can pick up more experience to build her intuition than someone in the 20th century could have picked up in a lifetime. It will be interesting to see if the same thing happens with go. It probably already has started to, with people playing tens of thousands of games on Tygem or wherever.

I also think that chess has analogies to the two kinds of approaches that you mention in go (large-scale and small-scale). Of course you eventually win (if resignation has not come first) by checkmate, but you can do this by attacking the king early in a tactical melee, or by playing a slow positional game and trying to grind out the win in the endgame. I get the impression you are a strong enough chess player to be familiar with this, so perhaps you are getting at something else.

I do think that on a whole-board scale, chess lends itself to blitz-style intuition more than go. The board is much smaller, and the positions have more "texture" (different sorts of pieces, pawn structure, constellations of pieces and pawns that suggest strategy), making it easier to assess the nature of the entire position in a second or less than is the case in go. Of course, maybe I just think this because I'm a stronger chess player than go player.

I have to admit that the rapid and blitz tournaments made for pretty gripping viewing. I was glued to my seat every morning all week.

By the way, is there a more precise definition of "Mickey Mouse time limits" than "faster than I think appropriate"? I can't easily find a use of the term by anyone else.

With some trepidation I offer the idea that the term "Mickey Mouse" as used in these boards derives from sports, such as baseball, basketball, or American football, where the difficulty or playing situations are diluted somehow so the game is not as serious or demanding. For example, in basketball the height of the hoop is lower, in baseball there is no pitcher's mound, or in football there is no tackling. In Go the fast games might be considered less serious because there is less time to think and, therefore, the standard of play is lower.

When interviewing a golden oldie, it might be a sensible question to ask about this. Chances are, somebody was asked already. Somebody know an interview?

One difference between chess and go is the way to become professional. The harder it gets for kids to become pro, the stronger they will be.

I don't think there is the equivalent of insei classes for chess (at least in most countries).

bayu wrote:One difference between chess and go is the way to become professional. The harder it gets for kids to become pro, the stronger they will be.

I think this is backwards. The better the players are, the stronger the competition is, the harder it is for kids to become pro.

dfan wrote:By the way, is there a more precise definition of "Mickey Mouse time limits" than "faster than I think appropriate"? I can't easily find a use of the term by anyone else.

As far as I know, John coined the phrase when proposing a reason for successful performance by Korean players in international tournaments. I've explicitly stated that I dislike the term, but I think it simply resulted in repeated use. Nowadays, you can check any thread here related to time controls, and you'll probably find the term.

"Faster than appropriate" seems aligned with the usage on L19, without what I perceive to be a mocking tone.

The problem with both expressions are that they are generalized terms that gloss over the details of the actual time settings, players, and so on.

If John argued against the use of a particular time setting without adding a comical character (which I find to be mocking), I imagine that I would have much less contention toward these types of, in my opinion, disrespectful threads.

But alas, this discussion has gone on for years, and I expect it to continue for years to come (maybe until I die from high blood pressure).

Kirby wrote:But alas, this discussion has gone on for years, and I expect it to continue for years to come (maybe until I die from high blood pressure).

If you have high blood pressure I suggest you refrain from using Mickey Mouse time limits.

According to the pros themselves, younger players are much better at reading, but older players have better intuition.

idontgetit wrote:According to the pros themselves, younger players are much better at reading, but older players have better intuition.

While not trying to compare myself to a pro I think that makes a lot of sense. Being definitely an older player (64 year old, playing for 43 years) I rely almost totally on intuition and find extensive reading to be much less enjoyable these days. Of course, this leads to gross blunders sometimes.

Doesn't it simply suggest that chess is simply more complicated for humans, it takes them longer to learn all of its intricacies.

Javaness2 wrote:Doesn't it simply suggest that chess is simply more complicated for humans, it takes them longer to learn all of its intricacies.

I don't think so. I think it shows that Go is in the middle of a period of rapid change. Much of the intuition that older players have developed no longer applies at the cutting edge level of top professionals. This seems to match what we have seen in recent developments in the game. A period of rapid change this late into the history of the game is a testament to the remarkable depth of go. It may be that the game will stabilize as computers become more proficient and we will see greater parity among younger and older players. (Or perhaps there are alternate explanations, and Go will continue to be a game dominated by the young.)

One other explanation is that go simply is a much longer game and takes more endurance. Go games normally take a lot longer than Chess games, and many more moves are played. If we assume that the same amount of energy is spent per move by top professionals, whether its go or chess, then a game of go will require a lot more energy.

Perhaps for chess, older players still have enough endurance for a game. But for go, the difference in energy between younger and older players becomes more obvious.

I remember when I was in high school, a family friend who's in his 50s talk about how his strength (like phsyical strength, not go playing) isn't that much weaker than when he was young in absolute terms. It's just that he can't keep it up for nearly as long. Might apply to go as well.

John Fairbairn wrote:2. The style of play has changed in go. In chess there is really only one way to win a game - checkmate. In go you can win either large-scale, by killing something big or amassing a huge territory, or small-scale - eking out a small territorial win. It seems from the record number of B+R/W+R results that large-scale now dominates. Maybe this is something that favours young people.

I believe the record number of B+R/W+R results is due a behavioral shift in players choosing to resign in the endgame when behind by a few points rather than play it out to the end. You rarely see B/W+4.5 or higher anymore since they usually resign before it gets to that point.

I don't think a change in the style of go is responsible for either the increase in resignations or the shift in age. If you look at the distribution of ages in top 10 go players before the recent explosion of the depth of Chinese talent, it resembles that of chess. In the 70's and 80's in the Japanese era, you rarely had teenagers in the top 10. Even prodigies such as Cho Chikun and Kobayashi Koichi needed some time until their late teenage years to the early twenties to climb into the top 10. There was usually a 40 year old or two like Fujisawa Hideyuki in the 70's or Rin Kaiho or Otake Hideo later in the 80's. The age 20's and 30's usually dominated with most players generally peaking in the late 20's and typically maintaining most of their strength into their 30s.

Same pattern roughly held during the Korean era. Even when Lee Changho burst on the scene and reached #1 as a 15 year old, the rest of the top 10 was mostly guys in their 30's - Kobayashi, the two Chos, Nie Weiping, Masaki. Yoo and Ma were up and comers but already in their mid 20's by the time they established themselves in the top 10.

As recently as 2000, Cho Hunhyun was ranked #2 in the world at the age of 46. But things have changed quickly where 40 year old players just cannot compete at the highest level anymore. Cho Hunhyun was still one of the strongest in his 40's back then but in today's game Lee Changho can't even crack the top 50. Lee Sedol is the only player in his 30's left in the top 10 and even he is starting to fade. So why the sudden shift in age profile among the top go players?

I think there are two inter-connected reasons. The first is the depth of competition. When the talent pool was shallower, an exceptional talent such as the Chos or Kobayashi can still remain in the top 10 even if their reading ability declines slightly with age. In today's environment, the competition is so fierce that even the slightest slip in form will knock you out of the top 10. Even guys in their 20's who have won international titles like Kim Jiseok or Jiang Weijie have fallen out of the top 10 with just a little slump. Top 10 in the old days were fairly static because the number of people with that level of talent was far smaller. Now, the separation between 10 and 20 is almost non-existent and the pool of hungry talented players looking to leapfrog is a hundred deep.

The second reason is internet go. Back in the day, you had much fewer opportunities to play against quality opponents and hone your skills. Even a player with the best possible environment such as Lee Changho, he had the benefit that he could play his teacher Cho, but how many games would he get with Cho? Since a teacher would only give a limited number of games to a student, the quality of the rest of the games were limited by the quality of other students at the go school that was physically there and looking for a game at the same time as you. These days, a skilled young pro can basically play against top competition 24/7. A kid like Shin Jinseo has probably played more high level practice games at 16 years of age as many other Go pros did by their 20's in the pre internet days. Look at the all the tygem games between Ke Jie and Park Junghwan. That kind of play improves both their play and lifts the standard of play as a whole. Those kind of practice games between the top pros would be so much harder if they had to be physically present in the same place. I would compare it to how talented Asian footballers become so much better once they move to quality European teams and train/play with the best talent.

So these two reasons are inter-connected because it creates a feedback loop. China's exploding middle class expands the talent pool for professional go players exponentially, and the internet allows all that talent to constantly play each other and improve at a quicker pace.

Yes, it's a bit sad that we no longer get to see the wise old veterans compete on the top stage as in past eras. But that is in my opinion, a blessing for us to be able to be spectators to a historic game that is continuing to evolve and reach higher and higher levels.