# Wildly Different Knowledge Levels

This is a topic I’ve wanted to write about since 7 years ago but was too lazy to find a good concrete example until recently.

The gist:

1. Imagine three people of varying degrees of knowledge in a particular area: A is a random person with zero knowledge, B is someone with real amateur knowledge, and C is a professional. You pose a yes/no question. There are weird situations where A and C agree on the same answer, but B disagrees. Someone (B) who is definitively smarter than a layperson in one area might, with good reason, disbelieve what experts in the field consider the objectively right idea.
2. In the above case on social media, person A might have no good argument or makes the default argument. B thinks they are pretty smart, and posts the standard reply to A. Then C makes a much more nuanced argument why B’s argument is wrong, and A is actually right. However, B doesn’t really understand or doesn’t read C’s argument, assumes C is just another dumb A and just repeats their flawed argument of why A is wrong.
3. Also, is C actually right??? What if there is someone smarter, D, who agrees with B and figures out the nuanced response to C’s nuanced answer? Most people arguing on the internet are just A and B, and their arguments don’t even make sense compared to the real debate between C and D.

The most provable example of (1) is in chess. If you could see 3 moves ahead rather than 2, you would play a better move almost all of the time. But here is a weird exception. In the position below, should White capture the d5 pawn with the knight?

What players of increasing skill level might think:

• A: “Who is a knight? What is a knight? Why is a knight?” [No]
• B: “So knights move in an L-shape, so I can take the pawn, and taking pawns is good…” [Yes]
• C: “If I take the pawn, then Black’s knight will take back and I lose a knight for a pawn, which is bad.” [No]
• D: “Black’s knight is pinned, so if I take the pawn and Black’s knight takes back, then my bishop will capture the queen, which is really good. So that is a free pawn for me as the Black knight cannot take back.” [Yes]

In fact, winning a queen is *so good* in chess that, in almost all cases that end with one side losing a queen, it is a waste of time and mental energy to calculate any further. And yet…

• E: “The knight pin is only a relative pin. Yes if I take the knight then Black will lose a queen after capturing back the knight… but wait! After losing the queen, Black can play a bishop check (Bb4+) and win White’s queen. After all the trades, Black is up a minor piece.” [No]

(For the chess enthusiast, the line is 1. d4 d5 2. c4 e6 3. Nc3 Nf6 4. Bg5 Nbd7 5. cxd5 cxd5 {diagram} 6. Nxd5 Nxd5 7. Bxd8 Bb4+ 8. Qd2 Bxd2+ 9. Kxd2 Kxd8 and Black is up a knight for a pawn.)

The weird thing is that player C is clearly better than B, but ends up with the correct answer only because of luck. Based on their thought process, C didn’t understand what was truly going on in the position, but rather, just happened to calculate a convenient number of moves ahead and stop. Weirder, D, who calculated more steps ahead than C, would play an objectively bad move here that C would have avoided! In some sense, D is just unlucky that they stopped calculating at the wrong level!

In addition, it happens that the “number of moves to look ahead” gap between D and E is quite large. In fact, the number of moves (using “ply” or “half-moves” in the chess term) to look ahead was:

• A: N/A
• B: 1
• C: 2
• D: 3
• E: 6

We’ll come back to this later, but if this were an analogy for how society views something as knowledge increases over time, we could be at a plateau for a long time between D and E, thinking that we have the answer figured out, but in fact have the wrong answer.

Let’s replace the chess question with a more real-life one, say “Is the climate warming?

• A: “I read online that it’s true so it’s true.” [Yes]
• B: “You can’t just believe what you read on the internet. Plus it was really cold yesterday.” [No]
• C: “One data point doesn’t define a trend. If you look at long-term graphs of temperature published by X, they go up over time.” [Yes]
• D: “What is source X, is it reputable? Also, what about a long-term temperature graph going back hundreds of years–weren’t there unusually warm periods in the past as well? [No]
• E: “Yes but not as drastic as the current warming period. And source X is the vast majority of scientists…” [Yes]
• F: (If you’ve been on the internet before, you can imagine how this continues…)

If you see a twitter post where someone says a few words saying global warming is false, you often have no idea if they are person B (who might not be that smart) or person N (who is very smart but has maybe stopped at the “wrong” level).

If you see two strangers debating on the internet without any context, it might be non-obvious how far they are down this argument chain and how much they’ve thought about it. This is compounded by how most internet posts & comments so brief that you can’t really see any nuance.

Similarly, this is how popular debates can make one side look bad even when supported by all the facts. In the chess example, if C and D went on a public debate, D would win, yet C’s view is the objectively correct one. And on a larger network like Facebook or Twitter, you have people from all over the knowledge spectrum–though probably concentrated in the As and Bs–so any “debate” on such a medium is pointless. You can consider a twitter chess “debate” where E actually says the correct answer but doesn’t have room to post the full variation (or the energy to do it for the 1000th time), and then the various D’s of the world point out why E is wrong, thinking that E is just another A or C.

To pull the chess analogy even further, the knowledge gap between D and E makes this even harder. If E could teach D to think not 3 moves ahead, but 4 or 5 instead, D would still have the same wrong answer as before. They would need to think ahead 6 moves to realize E is right. Maybe once someone gets to think-5-moves-ahead, they think that’s sufficient for everything and stop calculating further.

A converse situation arises if you are a person at knowledge level E, and you run into someone who seems to disagree with you. You might be so used to teaching people to go from D to E that you assume they are arguing from the level of D. However, most people who are in D’s camp might be at knowledge level B. In chess, explaining the D-to-E step to someone at B might not make any sense. It could even make things worse, as from B’s perspective: “Someone is saying nonsense and also disagrees with me, therefore I should update my belief to be even stronger.”

Look for the context. Know what level you’re at.

If you disagree with someone, know that they might be thinking much further ahead, and you might not even know what the real debate is.

# Tabula Rasa, Extinction, and Electricity

Chess

AlphaZero was one of the bigger headlines recently. Google’s new chess AI taught itself for 4 hours starting from a blank slate—no opening or endgame tables—and crushed Stockfish, the world’s previous best computer. See chess website articles here and here, a lichess.org collection of the games here, and the original research paper here via arXiv. This obviously has lots of real-world implications.

The most interesting thing is the way it won games. Ever since the early days of chess programming, we thought that chess computers could understand basic tactics but never deep positional play. Even in the pivotal 1997 Kasparov vs Deep Blue match, the human world champion famously said that Deep Blue must have been getting help from human grandmasters as it was playing non-computer-like moves.

Watching two chess AI’s play each other is typically a boring feat. But AlphaZero plays in a very human-romantic style, at least in the games that were revealed (and there’s definitely some selection bias there). AlphaZero often gave up lots of material for tempo, and it worked. One of the most talked-about positions is the following, where AlphaZero (white) abandons the Knight on h6 and plays Re1. It went on to win the game.

There’s lots of caveats in terms of how “real” of a result this is. Namely, the example games had Stockfish set on suboptimal settings. But still, it increases my opinion of the complexity of chess. As computers have gotten better, the way they play chess became more and more boring. But maybe the curve is not monotonic and we might have a stage where the game becomes more interesting again. Though I fear that eventually it will degenerate into optimal play from move one.

Political Correctness

People have been talking about the Sam Altman blog post.

Earlier this year, I noticed something in China that really surprised me.  I realized I felt more comfortable discussing controversial ideas in Beijing than in San Francisco.  I didn’t feel completely comfortable—this was China, after all—just more comfortable than at home.

That showed me just how bad things have become, and how much things have changed since I first got started here in 2005.

It seems easier to accidentally speak heresies in San Francisco every year.  Debating a controversial idea, even if you 95% agree with the consensus side, seems ill-advised.

And:

More recently, I’ve seen credible people working on ideas like pharmaceuticals for intelligence augmentation, genetic engineering, and radical life extension leave San Francisco because they found the reaction to their work to be so toxic.  “If people live a lot longer it will be disastrous for the environment, so people working on this must be really unethical” was a memorable quote I heard this year.

I don’t have any experience with the San Francisco discussion climate, but this seems weird. The fact that someone felt the need to write this post is a sign about the culture.

I’m probably way more in favor of politically incorrect ideas, mainly since I think the world vastly overvalues traditional ideas, and ironically because there is so much that you can’t say in China. Tyler Cowen points out, “…your pent-up urges are not forbidden topics any more.  Just do be careful with your mentions of Uncle Xi, Taiwan, Tibet, Uighur terrorists, and disappearing generals.”

So Altman’s general point about politically incorrect ideas is probably correct. I don’t have any problem with discussing unpopular ideas. But I just don’t see people moving form San Francisco to China as a reasonable solution. There are certain topics that we might be overly sensitive to, but the overall level of idea tolerance would seem very tilted in favor of the US.

Human Extinction

Obligatory shout out to 80000 Hours’ extinction risk article. The idea was to discuss various sources of extinction and estimate their chances of occurring.

What’s probably more concerning is the risks we haven’t thought of yet. If you had asked people in 1900 what the greatest risks to civilisation were, they probably wouldn’t have suggested nuclear weapons, genetic engineering or artificial intelligence, since none of these were yet invented. It’s possible we’re in the same situation looking forward to the next century. Future “unknown unknowns” might pose a greater risk than the risks we know today.

Each time we discover a new technology, it’s a little like betting against a single number on a roulette wheel. Most of the time we win, and the technology is overall good. But each time there’s also a small chance the technology gives us more destructive power than we can handle, and we lose everything.

And:

An informal poll in 2008 at a conference on catastrophic risks found they believe it’s pretty likely we’ll face a catastrophe that kills over a billion people, and estimate a 19% chance of extinction before 2100.

As a trader, the first thing that comes to mind is to create some betting markets on such events happening and have a bunch of people trade, but this leads to weird selection effects and the payout is too long-term. So looking at some polls and mentally adjusting is probably right.

In addition, their ordering of what to prioritize is interesting:

1. AI safety
2. Global priorities research
3. Building the effective altruism community
4. Pandemic prevention
5. Improving institutional decision-making
6. Nuclear security

I should maybe have a recurring Twitter section. Anyway, here is a tweet by Julia Galef, and I’ve also wondered about this topic a lot.

The thought experiment I want to run is to throw together a racially diverse set of kids in a bubble, and expose the kids to roughly no knowledge of real world history or any hints of racism outside, and otherwise act like everything is normal. In this bubble world, would they start becoming racist against each other? I would guess no.

I think an underrated explanation in general of why people do something is because everyone else around them does it or that parents or teacher early on in their life do it. Social/cultural norm is a really strong incentive/disincentive for activities.

Cryptocurrencies and Electricity

There are definitely people worrying about the massive amount of world electricity consumption from bitcoin mining. Newsweek extrapolates that bitcoin will take up the world’s electric output by 2020. It’s currently at 0.15% according to some website. This is not small, giving how quickly it has been growing. Wired worries it will become the paperclip machine:

That’s bad. It means Bitcoin emits the equivalent of 17.7 million tons of carbon dioxide every year, a big middle finger to Earth’s climate and anyone who enjoys things like coastlines, forests, and not dying of mosquito-borne diseases. Refracted through a different metaphor, the Bitcoin P2P network is essentially a distributed superintelligence utterly dedicated to generating bitcoins, so of course it wants to convert all the energy (and therefore matter) in the universe into bitcoin. That is literally its job. And if it has to recruit greedy nerds by paying them phantom value, well, OK. Unleash the hypnocurrency!

I also stumbled upon a more optimistic viewpoint, claiming that bitcoin mining will trigger increased development and adoption of clean energy:

But electricity costs matter even more to a Bitcoin miner than typical heavy industry. Electricity costs can be 30-70% of their total costs of operation. Also, Bitcoin miners don’t need to worry about the geography of their customers or materials shipping routes. Bitcoins are digital, they have only two inputs (electricity and hardware) and network latency is trivial as compared with a truck full of steel. This particular miner moved an entire GPU farm across the U.S. because of cheap hydroelectric power in the Pacific Northwest and, in his words, “it’s worth it!” That’s also why we see miners in Iceland. Aside from beautiful vistas you can find abundant geothermal and hydraulic power in the land of volcanoes and waterfalls.

If Bitcoin mining really does begin to consume vast quantities of the global electricity supply it will, it follows, spur massive growth in efficient electricity production—i.e. in the green energy revolution. Moore’s Law was partially a story about incredible advances in materials science, but it was also a story about incredible demand for computing that drove those advances and made semiconductor research and development profitable. If you want to see a Moore’s-Law-like revolution in energy, then you should be rooting for, and not against, Bitcoin. The fact is that the Bitcoin protocol, right now, is providing a \$200,000 bounty every 10 minutes (the bitcoin mining reward) to the person who can find the cheapest energy on the planet. Got cheap green power? Bitcoin could make building more of it well worth your time.

It’s very unclear in bitcoin’s case how good the upside is for the world, but it doesn’t seem anywhere close to being an extinction risk.

Recommended is Tyler Cowen’s post on crytocurrencies and social value.

Progress

previously wrote that we take modern life improvements for granted and sometimes erroneously yearn for the hunter-gatherer life. Well here is a Quillette article on precisely the romanticization of that.  Here are some examples:

In his later work, Lee would acknowledge that, “Historically, the Ju/’hoansi have had a high infant mortality rate…” In a study on the life histories of the !Kung Nancy Howell found that the number of infants who died before the age of 1 was roughly 20 percent. (As high as this number is, it compares favorably with estimates from some other hunter-gatherer societies, such as among the Casiguran Agta of the Phillipines, where the rate is 34 percent.) Life expectancy for the !Kung is 36 years of age. Again, while this number is only about half the average life expectancy found among contemporary nation states, this number still compares favorably with several other hunter-gatherer populations, such as the Hiwi (27 years) and the Agta (21 years). Life expectancy across pygmy hunter-gatherer societies is even lower, ranging from about 16-24 years, although this may have as much to do with pygmy physiology as with the hunter-gatherer lifestyle.

And:

11 of these 15 societies have homicide rates higher than that of the most violent modern nation, and 14 out of the 15 have homicide rates higher than that of the United States in 2016. The one exception, the Batek of Malaysia, have a long history of being violently attacked and enslaved by neighboring groups, and developed a survival tactic of running away and studiously avoiding conflict. Yet even they recount tales of wars in the past, where their shamans would shoot enemies with blowpipes. Interestingly, Ivan Tacey & Diana Riboli have noted that “…the Batek frequently recount their nostalgic memories of British doctors, administrators and army personnel visiting their communities in helicopters to deliver medicines and other supplies,” which conflicts with the idea that hunter-gatherer societies would have no want or need of anything nation states have to offer. From 1920-1955 the !Kung had a homicide rate of 42/100,000 (about 8 times that of the US rate in 2016), however Kelly mentions that, “murders ceased after 1955 due to the presence of an outside police force.”

And:

So, what explains the popularity of this notion of an “original affluent society”? Why do people in societies with substantially greater life expectancy, reduced infant mortality, greater equality in reproductive success, and reduced rates of violence, romanticize a way of life filled with hardships they have never experienced? In wealthy, industrialized populations oriented around consumerism and occupational status, the idea that there are people out there living free of greed, in natural equality and harmony, provides an attractive alternative way of life.

I also definitely live in a bubble, as I don’t know anyone openly in favor of hunter-gatherer society.

This also reminds me of Joseph Stiglitz’s book, The Price of Inequality. Most of the book is very methodical or at least numbers-driven. Then comes this absurd passage on the Bhutanese (p. 155 of the Norton edition):

Bhutan, the remote Himalayan state to the northeast of India, for instance, is protecting its forests as part of a broader commitment to the environment. Each family is allowed to cut down a fixed number of trees for its own use. In this sparsely populated country, I asked, how could one enforce such an edict? The answer was simple and straightforward: in our jargon, social capital. The Bhutanese have internalized what is “right” when it comes to the environment. It would be wrong to cheat, and so they don’t.

I’ve been waiting for years to quote this paragraph, but here it is. There is in general some weird sacred reverence of non-Western cultures. Is this related to the Altman political correctness theme? Can I just pick a well-off small community in America and say “it would be wrong to cheat, and so they don’t”? Anyway, it’s really easy to say some society works pretty well, and then take all the modern improvements for granted.

# When Does Not Deciding Count as a Decision?

This week’s topic is whether not deciding is itself a decision. Let us start by escalating things quickly: consider the classic trolley problem.

There is a runaway trolley barreling down the railway tracks. Ahead, on the tracks, there are five people tied up and unable to move. The trolley is headed straight for them. You are standing some distance off in the train yard, next to a lever. If you pull this lever, the trolley will switch to a different set of tracks. Unfortunately, you notice that there is one person on the side track. You have two options: (1) Do nothing, and the trolley kills the five people on the main track. (2) Pull the lever, diverting the trolley onto the side track where it will kill one person. Which is the correct choice?

While there are many interesting aspects of the trolley problem, with many variants of the problem that may cause one to reconsider their views, this article is concerned with one particular question: Is actively choosing option (1), or doing nothing, equivalent to passively not making a decision? (It turns out this question has real-world consequences, as will be evident below.)

That is, is there is a difference between

• A) Considering (1) and (2), and deciding that (1) is morally superior; and
• B) Ignoring the decision, and thus passively allowing (1) to occur?

For one difference, consider the same trolley problem except that the trolley is initially headed for the 1 person, and you have to pull the lever to turn it to the 5 people. In such a case, someone using thought process (A) would STILL choose (1), to have the trolley hit 5 people, whereas someone using thought process (B) would now “choose” (2), allowing the 1 person to be killed.

In the original case, it is difficult to justify non-decision; however, one would most likely be viewed as innocent if one made no decision and allowed the trolley to kill 5. This is because the legal system generally can only punish decisions, not non-decisions. So the real question is, are the following equivalent?

• I) The trolley is already headed towards the 5 people, and you allow it to continue on course.
• II) The trolley is headed towards the 1 person, and you divert it to head towards the 5.

The outcome of both situations is the same, namely the 5 people die but the 1 person survives. However, it seems that if this were considered a wrong action, we would be able to legally punish (II), but not (I), since (I) could have been based on not deciding. However, should they be legally viewed the same? That is, should someone be accountable for not deciding?

Speed Chess

One of the interesting examples of decision vs non-decision in a non-legal, non-moral context is blitz chess. When you have only a few minutes for the whole game, you cannot afford to spend a sufficient amount of time thinking about every move. Instead, you must ration your time as a resource, and in some cases choose to not think on a particular move. Speed chess is indeed based primarily on intuition, less so on cold calculation.

Thus in speed chess, it is very feasible that not thinking about a move is itself a decision. Once you have a lot of experience, you gain the intuition of which types of positions require calculation and which do not. It becomes possible to say when it is “correct” to not decide. In this case, not deciding is clearly a decision.

Willful Ignorance

Decisions are based on available information, so a natural question relevant to whether one can be held accountable for not deciding is whether one can be held accountable for not knowing. Moreover, it is important whether someone can be held accountable for intentionally refusing to know. After all, no one would blame a child for thinking that Earth is flat. But when adults believe the world is flat, that is an entirely different issue, because most likely they have intentionally refused to hear the case of the round Earth.

The same goes for evolution, only there are significant national and state policy decisions made based on the refusal to learn about it. Of course, we wouldn’t hold a child responsible for their beliefs, but for an adult to use willful ignorance in decision making is inexcusable.

Whether willful ignorance is problematic in principle can be seen in a trolley variant. Suppose the person who has the power to pull the lever believes that the case is as in the original trolley problem. However, the side which supposedly has 1 person actually has 100.

The operator pulls the lever, diverting the trolley from the side of 5 people to the side of 100, killing all 100 people. Note that the operator cannot be blamed because of genuine ignorance.

Now consider an alternative scenario. The situation is the same as above: the operator believes that is a matter of 5 lives vs 1 life, but it is actually a matter of 5 vs 100. Before making the decision, someone else runs in, screaming that there are actually 100 people on the second track. It would be extremely easy to verify this, but instead, the operator refuses to listen to the new information and diverts the track to the 100 anyways, still clinging to the belief that there is only 1 person. In this case, the operator is being willfully ignorant.

(Can some lawyer explain if there are indeed differences in the previous situations?)

There are countless other examples where the intentional lack of information should not be a valid excuse for a bad decision. Suppose someone is about to receive the death penalty for a crime. A piece of evidence shows up that could provide reasonable doubt in the conviction. It would be absurd to refuse to see this evidence, especially because the refusal to see it would most likely be the result of the people really wanting this person to receive the death penalty, and that the extra information could disturb their beliefs.

A similar example is that some nation has borderline-quality intel justifying a war, and they decide to launch the war before they look at newer intel that could possibly negate the previous intel. Thus even if they are later found to be wrong, they would be able to use the ignorance argument by saying they didn’t know better at the time, even if they knew of the possibility of being wrong. There is a difference between genuinely believing the lack of contradictory information and the intentional refusal to look at (possibly) contradictory information.

It’s not a fine line that separates non-decision and the active decision that leads to the same result as in non-decision. Similarly, it’s not a fine line that separates genuine ignorance and willful ignorance. But even without a perfectly clear demarcation, the differences are real and these actions can and should be treated differently.

# The Swinging Pendulum: Talent vs Hard Work

Last week we had as guest speakers IS 318 chess teacher Elizabeth Spiegel and other interesting people from the documentary Brooklyn Castle. This was a highly relevant talk as we had a great number of amateur chess players in the audience, incidentally in time for the finishing days of our summer chess tournament. (In addition, Elizabeth and IS 318 students had visited JS before.) In the Q&A, there was an interesting section about the roles of talent and hard work. At dinner we discussed it more in depth, with respect to both chess and skills in general.

Elizabeth had some interesting things to say. A particular student played chess in a very creative and original fashion, a telltale sign of talent. For most players, however, hard work is far more important.

By far the most interesting point was about the amount of time dedicated to chess by some of the students: up to 20-30 hours a week. In turn, the fact that IS 318 was a relatively economically disadvantaged school was in some ways an advantage, as many of the students had nothing else to do. Thus they had an incredible amount of time to study chess. Their competitors from wealthier areas often had other extracurricular activities, and thus did not spend as much time on chess.

At one point the documentary went to the 2009 National Scholastic Grades tournament (I was there!), where IS 318 had a stellar performance:

The first place in the 8th grade section was Canyon Vista Middle School. Funny how life works, isn’t it?

It was also interesting to think about the distinction between areas where talent seems to play more of an effect. For example, child prodigies thrive in chess, math, and music, but not so much in literature, art, and finance. Perhaps the extra layers of complexity make it more difficult to do without specialized knowledge coming from long hours of study or experience.

My view was that hard work is far more significant, though I used to have a more mixed view. Last December I wrote a post on Geoff Colvin’s Talent is Overrated, a book which is in the same camp as Malcolm Gladwell’s Outliers. These have probably influenced my views greatly.

Finally, an undisclosed party member gets credit for the name of this post. He compared the general consensus on hard work vs talent to a pendulum: It used to be too far in the talent side, but now it has swung a bit too much in the hard work side. So the question now is, is talent underrated?

# Blogging, Chess, and the Sunk Cost Fallacy

Since the summer began, I have again fallen into an inconsistent posting schedule, one of the things I was trying hardest to avoid. One of the reasons is that I still have retained a perfectionist attitude, that to write something, it needs somehow to be interesting or insightful, and in addition, written well. Otherwise, I thought, someone else would have just written something that is strictly better.

But as a result, I end up scrapping many of my drafts and never following up on them, and I rarely actually publish anything. Of course, this relates to other areas of life as well. I often try very hard to avoid situations where I could make mistakes, rather than just making mistakes and learning from them.

This summer I have been getting back into chess, and a few weeks ago I noticed something that I had never noticed before. It was in a game of blitz chess, or speed chess, where the clock is as much the enemy as the person seated across is. In general I played moves fast, but the moment I made an error, I froze up and wasted a lot of time. Quite fittingly, we had earlier in the day a lecture about illusions and cognitive biases, including the sunk cost fallacy. The rational thing would have been to keep playing fast, reasonable moves to keep a time advantage. However, after making the blunder, which was losing a Knight if I recall correctly, I kept thinking about how to recover the piece instead of just playing reasonable moves. My teammate, who was also the person who gave the talk, rightfully yelled at me to keep playing quickly when I froze up.

This story is a lesson in thinking rationally even in unfamiliar or just downright messy situations. In general I catch my mistakes quickly, thus it is rarely an issue in everyday activity or even in an interview. But in chess (and in trading), there is no taking back a mistake, only continuing on making good moves even with a bad position.

Perfectionism, while sometimes useful, is something I am trying to shake off. I will post on a regular schedule (it really is like the fifth time I’ve said that), perhaps put up a few chess games, and try to make some mistakes. A weekly posting schedule, namely every Sunday, seemed to work well for a while, so I am bringing that back online. Enjoy!

(Edit: Don’t worry math people, I’ll try to resurrect the math blog too.)

# Talent Is Overrated

“Talent” is a word that is tossed around all too often, whether for top musicians or businessmen, or even just a person who creates popular Youtube videos. The idea of talent is in nearly every case taken for granted. As a young member of a very supportive family and community, I had heard the saying myself many times. But is talent a correct or even useful explanation for high-level performance?

I recently read a very intriguing book by Geoff Colvin. It was really a lucky buy—I was actually reading through reviews of Josh Waitzkin’s The Art of Learning, when the ever-so-omniscient Amazon Recommendations pointed me to a bizarre and blatantly absurd statement: Talent is Overrated.

With a plethora of examples, data, accumulation of research, and forcible writing, Colvin argues convincingly that the source of great performance in just about every field is best explained not by reference to the mysterious force known as talent, but by sheer amount and direction of deliberate practice.

My Personal Experience

First, a line from Colvin (193):

Their parents made them practice, as parents have always done, though it’s interesting to note that in these cases, when push came to shove and parents had to make a direct threat, it frequently played off the student’s intrinsic motivators. So it wasn’t “If you don’t do your piano practice we’ll cancel your allowance,” but rather “we’ll sell the piano.”  Not “If you don’t go to swimming practice you’ll be grounded Saturday night,” but rather “we’ll take you off the team.” If the child truly didn’t care about the piano or swimming, the threats wouldn’t have worked.

I was one of those kids who was, regarding the piano, totally immune to such a threat. As I wrote earlier, I absolutely dreaded playing the piano, and would have loved to see the piano disappear and find a bunch of cash in its place. But what I lacked in interest for the piano I made up for in my interest in chess. From 2003 to 2010, I competed in more than 70 rated chess tournaments. But looking back at the distribution of tournaments, I found that the majority of them occurred between 2003 and 2006, with one resurgence in 2008 [data]. It would be accurate to say that my tournament frequency was very closely correlated to how much time I spent on the game outside of tournaments in practice. As if to confirm Colvin’s thesis, here are my regular and quick rating graphs:

When the frequency of tournaments, and thus training, increased, my rating climbed. And when the frequency of tournaments and training decreased, my rating stagnated or declined. This seems to support the dedicated practice model argued in Colvin’s book. The performance in a given time period seemed to be determined by the amount of training in the same time period.

But what about compared to others? I am hardly an expert player, but my very first rating  after my first tournament, 1372, was in the 96-97th percentile of scholastic players at the time. By contrast, the current US chess champion Hikaru Nakamura, whose current USCF rating is a whopping 2834, started at a provisional rating of 684, an unimpressive statistic. However, he has played in 439 rated events over a period of 17 years, which is a hell of a lot more effort than I had ever thought about spending on the game. Thus even when you have an “advantage,” such as having a starting rating of 1372 versus 684, thinking of it in terms of talent is useless. If you do not follow it up with the necessary amount of work, the advantage will assuredly disappear.

There is a third point, to truly put the nail in the coffin of the talent model. In a two year span from 2006 to 2008, my rating stopped improving in the 1700s. Excuses aside, I simply didn’t practice the game much. But one thing I think could have happened is what Josh Waitzkin described, from Colvin (197):

The most gifted kids in chess fall apart. They are told that they are winners, and when they inevitably run into a wall, they get stuck and think they must be losers.

I don’t think it takes a gifted kid to run into the wall and get stuck (the 1372 initial rating was actually in part due to luck, as my first few tournaments were counted out of order, and a tournament that I had done really well in was incidentally the first one counted). For those two plateau years, I did feel the way that Waitzkin forewarned. I thought the high initial rating meant something special, i.e. talent, and that the 1700 plateau meant I was doomed. This thought process in terms of talent condemned me mentally to not advance. Even though I was still fairly high rated in my age group, I stopped practicing and reading as much, and as a result did not prepare myself adequately for tournament events. This caused my rating to drop.

How to Be a World-Class Performer

Colvin’s thesis works for far more than just chess. He applies it to the violin, piano, football, players, business, investment, management, art, teamwork, and just about anything, all while citing tremendous amounts of evidence for his claims. For music, the obvious counterexample is Mozart, yet early in the book Colvin disposes of this myth, as well as that of Tiger Woods. Mozart, for instance, had my years of intense, expert training starting at an early age, and Tiger Woods swung his first club at age seven… months, also trained by his father.

Another result of years of deliberate practice is the ability for an expert to see complex patterns that would completely elude an average person. A professional tennis player can return a serve of a ball traveling at a speed so high that a normal human should not even have time to react. Yet they are normal in this sense. But they don’t watch the ball, they watch their opponent’s body movements instead, and know approximately where the serve is going to land (or whether it will fault) before the racket even hits the ball. Similarly, a top stock trader can see signs that the average trader does not even consider to be relevant. A top manager sees the critical signs more so than an average one. And a master chess player can memorize an entire chess position in seconds and reproduce it perfectly, while the average person can recall only five or seven pieces. Most notably, this is not from better general memory, but by extensive training to be familiar with certain positions and patterns, so that they read a position by words instead of letters.

I would most certainly recommend this book to anyone. It breaks the shackle of “talent,” which although is a warm, comforting hope, it is no more than that, a beloved superstition with little evidence, and which discourages so many from even attempting something because they believe they “don’t have talent” or “divine spark” for it. But as it has repeatedly occurred, looking back at the backgrounds of top performers give little or no indication of any talent early on, but rather, what is common to all of them is an immense amount of training and dedicated practice. Perhaps this is the even more fascinating hope, that the world is within reach to everyone.

# How Many Moves Ahead Do You Calculate?

In the past month, I played maybe 15 casual games of chess, and from these, I discovered a few things about calculation that I had overlooked in my otherwise tournament-heavy experience. I was able to learn new things precisely because the games were casual, and thus not subject to the competitive mindset. In most of them, I talked to my opponent as I played. These were against lower level players, and to help them out, I sometimes discussed my thought process mid-game. This was also a chance to explain to non-competitive chess players what goes on through an experienced player’s mind. Here are some of the things I found myself explaining:

1. It is NOT necessary to calculate several moves ahead.

Perhaps the most common misconception about chess is that a player needs to think 3, 4, 6, or even more moves ahead to win.

In reality, for the vast majority of moves, thinking even one move ahead is sufficient. Once you play enough games, you develop an intuition that will guide you as to which pieces to move, when to move them, and where. For example, I’ve played dozens of games that opened with the Sicilian Dragon, quite a few of them in rated tournaments. When I play a new game that opens with the Dragon, I instinctively know which strategies work and which ones don’t, and my thinking time is spent on figuring out the differences and trying to exploit them. That is, I can think about refining a strategy instead of inventing one. This saves a lot of time.

Sometimes, however, you will need to calculate an exorbitant number of moves ahead. I recall a game in the 2006 National Open that ended up in a rook and pawn endgame.  On the last move of the game, I thought for perhaps 40 minutes, calculating no fewer than 10 moves ahead. When I played that move, my opponent thought for an entire hour and then resigned. What made the position so complicated? Here’s what was going through my mind: pawns dangerously close to queening, rook and pawn checks, queen checks after 2 pawns queening, mating nets, as well as king positioning. This is an extreme exception though, and most of the time I don’t calculate over two moves ahead.

2. A balance between intuition and calculation is important.

As I mentioned above, intuition plays an enormous role in chess. With it, you feel like you “know” what to do. Should I attack the queenside or the kingside? Or should I try to break open the center? Should I trade a bishop for a knight? Should I push the g-pawn or the h-pawn to start the attack? Instead of calculating such things from scratch, you can often use intuition to develop a preliminary answer, and then use calculation to confirm or deny your hunch. This is incredibly useful in a tournament setting, where your clock is ticking.

One can go the other extreme and rely too much on intuition, or rather, too much on generalizations. At some point in a game, I was winning by a knight for a pawn, a pretty huge material advantage, but had slightly less board space than my opponent did. A third-party was watching the game, and told us that he thought my opponent was winning because he had more space.

I disagreed with the third-party because, based on the current board position, the space advantage did not make up for the material deficiency. In fact, I thought it did not even compensate a loss of one pawn, let alone two pawns (the equivalent of a knight for a pawn).

3. You don’t have to calculate every possible move.

This is why humans were able to beat even the best of computers for so long. We lasted until 1996. Even though computers could calculate orders of magnitude faster than the human brain at that time, and even before, they did not do the calculations in as smart a manner. For instance, say we are in some position in the middle-game, and that I have 20 legal moves, to each of which my opponent has 20 replies. Then to calculate just one move ahead, I would have to look at 400 positions to fully cover every scenario. (In chess lingo, a “move” consists of one move by white and one move by black.) To calculate two moves ahead, it would be 400², or 160,000 positions. For a human to do this, even analyzing one position per second, it would take several hours to compute. Now suppose you want to calculate 10 moves ahead: 400^10. This is roughly 10^26 positions to analyze. Even a computer analyzing these at a billion (10^9) positions per second would require 10^19 seconds, or 300,000,000 years. It would be impossible for a human.

So how is it possible that a human can calculate 10 or more moves ahead? Well, we (unconsciously) use a technique called pruning, or ignoring certain moves. For instance, out of the original 20 moves, only 3 of them look interesting at all, and the other 17 seem either accomplish nothing or are silly moves that lose material immediately. For a full “move,” this would be 9 positions, and calculating 10 moves ahead would give 9^10 = 3.5 billion positions. This is much more reasonable, but is still an extravagant amount. What happens in real calculation is that many moves are forced, in which there is only one response. Other times, one or two of the three interesting moves degenerates into a clear position at which you can stop calculating. Thus in calculating 10 moves ahead, it is possible that you may only need to look at 15 final positions, which is much less than 3.5 billion or 10^26.

The reason computers are beating humans now isn’t because the computers think faster—it’s because they think smarter. If it can prune 20 moves into 3, then it might only need to calculate 3.5 billion positions to think 10 moves ahead. And at a billion positions per second, this would only take 3.5 seconds.

Conclusion

There were also some occasions where I played blitz chess (speed chess) and blind chess, though not concurrently. Speed chess is normal chess with strict time restrictions. This effectively limits the number of moves ahead you can calculate. It emphasizes speed over accuracy of calculation. This is also where experienced players perform really well, as intuition reduces the time spent dramatically. This is why in speed chess it can appear that the players are playing instantly and not thinking. They actually are thinking, but just in a different way.

Blind chess is also the same as normal chess, only you are not allowed to look at the board. You have to call out what move you want to make, and your opponent calls out the response, etc. The trick is about making sure you know where all your pieces are. The easiest way to do this is to keep a mental image of the board in your head, and memorize all the moves that occurred in the game. That way, if you are ever unsure of where a piece is supposed to be, you can play through the previous moves to track that piece’s location.

Overall, I treat normal chess, blitz chess, and blind chess as the same game; only a few changes in the thinking process are needed. So in response to the question “How many moves ahead do you calculate?”, my answer is one or two moves usually, more if necessary. How many moves ahead do YOU calculate?