The Logic of Risk Taking

https://medium.com/incerto/the-logic-of-risk-taking-107bf41029d3

by the-mitr • 9 years ago

160 74 9 years ago

medium.com

malmsteen2• 9 years ago

tl dr ?

cholmon• 9 years ago

"skin in the game"

malmsteen2• 9 years ago

<

zaptheimpaler• 9 years ago

Taleb trying to sound smart.

pierre_d528• 9 years ago

Please, could you elaborate?

tw1010• 9 years ago

He seems to have a better grasp of probability than most people (me included). Perhaps he's not the greatest at digesting his ideas into an easy-to-grok form, but he seems to be trying his best. You don't have to listen to everything he says but you also shouldn't dismiss it all because his presentation is lacking in eloquence. I think this is especially true since he has a background he shares with very few other business-esque advice-givers. When you find a person like that, someone outside the norm, you really aught to listen to the person and learn to read between the lines or reinterpret it into a form that you can understand, since finding people at that fringe is hard enough. Requiring that the person be a perfect communicator is too much to ask and you're doing yourself a disservice by ignoring them.

late2part5• 9 years ago

It's my opinion that:

1. Taleb is incredibly smart 2. He is not being arrogant or trying to impress (any more than any other writer) 3. He is writing about something that is interesting and not everyone knows

lliamander• 9 years ago

If, at a casino, there is a 1% chance of going bust, then given N people attending the casino approx 1% will go bust, no matter the size of N.

However, if a single person goes to the casino N times, their odds of going bust approach 100% as N gets larger, because as soon as you go bust you can no longer keep playing.

In any kind of iterated game, concern about tail risks is justified when the tail risk means you can no longer play the game.

Social science and economic research that attempts to analyze the "rationality" of human behavior has thus far failed to take this logic into consideration, falsely portraying people as irrationally "loss-averse". This oversight in the research community is obvious to successful investors, insurers, etc. because the latter have personal investment in the outcome of these iterated situations that the former simply do not. [This is only my attempt at a summary of his argument, not a defense]

[EDIT: spelling mistake]

sp527• 9 years ago

Loss aversion is a human bias that has fairly obvious evolutionary origins: fear is a stronger behavioral motivator than risk-seeking tendency because it confers a survival advantage (‘you only die once’).

dragonwriter4• 9 years ago

Social science and economic research that attempts to analyze the "rationality" of human behavior has thus far failed to take this logic into consideration, falsely portraying people as irrationally "loss-averse".

And here’s where he goes off the tracks: even a quick search of Google Scholar will show that it, at best, grossly overstates the case to say that social science (and particularly economics, and even more specifically the subfield of “decision theory” that he calls out) research “thus far” has neglected tail risks in addressing rationality (it's also incorrect to say that there is some kind of consensus result that people vary from rationality consistently in the direction of loss-aversity.)

RachelF• 9 years ago

Read his first non-finance book, "Fooled by Randomness".

It is actually well written, easy to read and briefly contains all the ideas in his later books.

His writing quality has decreased hugely since then, but his ego has not.

cortesoft2• 9 years ago

I feel like this idea of 'tail risk' is pretty well understood in the gambling community, especially in the poker world of no-limit holdem tournaments.

For example, everyone knows that pocket Aces are the best possible starting hand in poker, and the default strategy is to call any all-in call if you have them; your expected value will always be positive, no matter how many other people call or what they have.

However, there is a well known exception to this rule, and many poker books talk about it - if you are the small stack at the final table, it might make sense to fold your pocket aces pre-flop if more than one larger stack has already called.

The reasoning is that even if you win the hand, you aren't certain to move up in finishing position, which is what determines your winnings. On the other hand, if you lose the all in, you are CERTAIN to not move up. If multiple other people have already called the all-in, then there is a good chance that SOMEONE will be knocked out, moving you up in the finishing position without having to take a risk. Therefore, even if you have a positive expected return, it makes more sense to avoid the 'ruin' situation and fold the pocket aces.

ThrustVectoring• 9 years ago

your expected value will always be positive, no matter how many other people call or what they have.

It depends on how much of an "edge" you expect to have on future play as well, and the risk you expect of becoming unable to fund exploiting that edge in the future. Not every positive-value bet will maximize your expected overall return.

unabridged• 9 years ago

I have an idea for a poker metric that measures your risk for entire tournament. Each time you face elimination, your percent chance of surviving is multiplied together. So let's say you finish a tournament and faced all in 3 times with these survival rates:

80% * 50% * 70% = 28%

Then you would have a maximum of 28% chance of reaching that point again.

cortesoft2• 9 years ago

An interesting idea, but the number of times you face an all-in is dependent on how you do the rest of the game - if you win every hand, for example, you are never going to face an all-in because you will always be big stack.

StanislavPetrov2• 9 years ago

There's tons of information out there about this idea, which is called ICM by poker enthusiasts.

dahdum• 9 years ago

Similarly good bankroll management always accounts for risk of ruin for the stakes and game type. Limit holdem and pot limit omaha have vastly different requirements due to variance, for example. Also why the top players are so frequently staked by 3rd parties, they recognize the need to reduce their risk.

nemo44x2• 9 years ago

I believe the term is “Other People’s Money” -OPM.

hkmurakami• 9 years ago

Interesting how this calculus is completely different between tournament and cash games.

theptip2• 9 years ago

If someone used a standard cost-benefit analysis [for russian roulette], he would have claimed that one has 83.33% chance of gains, for an “expected” average return per shot of $833,333.

Forgot to consider the "cost" in the cost-benefit analysis. Most would value their life significantly higher than 6*$1m, so the expected return is negative.

late2part5• 9 years ago

Interesting that you think most people value their life over $6M USD. I bet you that across the entire world, more than 50% of adults would take a 15% chance of death to win $6M USD.

ryandrake• 9 years ago

$6M is a life changing amount of money. Tempting...

jasonmclaren• 9 years ago

There are different ways to measure how much people value their lives. Some of the most interesting involve "revealed preferences", ie. what people actually do, not what they say.

For example, how much higher wages do people demand for riskier work, controlling for other factors? Or, how much extra do they pay for, say, a car with a better safety record?

I just checked the Wikipedia article on the "value of a statistical life", and apparently most current US estimates are around $9 million. This is up from around $7 million a decade ago.

https://en.wikipedia.org/wiki/Value_of_life

ianamartin5• 9 years ago

Of course he forgot about the cost part of the equation. He’s the Deepak Chopra of the Malcolm Gladwell of the Freakonomics guys of the herbal remedy people.

A fucking clown.

goialoq2• 9 years ago

The calculation calls for "$1m", not 6x$1m

theptip2• 9 years ago

When calculating expected cost/benefit, you take the value of the outcome multiplied by the probability of that outcome. So the expected cost to play Taleb's game of Russian Roulette is 1/6 * $6m = $1m, if you value your life at $6m.

Note that's a bit higher than the expected benefit of $0.833m, I kept the numbers round for brevity. The breakeven point would be:

5/6 * C_win = 1/6 * C_lose

C_win = 1m

C_lose = 5/6 * 6/1 * 1m = $5m

My point was that Taleb omitted this calculation entirely, which makes it an incomplete cost/benefit analysis.

TailorJones• 9 years ago

Another non-closable pop-up window.

executesorder66• 9 years ago

Just use uBlock Origin.

dvt3• 9 years ago

Taleb loses me at the end (Nicomachean ethics? come on), but I love his "take risks, but don't take stupid risks" stance. A stance which I first encountered in How to Legally Own Another Person[1].

The key takeaway is twofold, in my opinion:

- People underestimate tail risk in their day-to-day: Bob smokes and drinks daily and eats red meat and doesn't exercise at all. These are "small" events independent of each other, but the aggregate will probably lead to ruin (probably a heart attack in his mid-50s).

- People overestimate tail risk (due to a lack of courage[2]): Jill says she's "risk averse" and that she won't start a business because it might ruin her -- although in the grand scheme of things, 6 months of your life and a $10,000 investment are not ruin-inducing.

[1] http://www.fooledbyrandomness.com/employee.pdf

[2] Maybe dragging ethics into this wasn't such a bad move after all.

theyregreat• 9 years ago

Exactly. I take fish oil, eat a mostly plant-based diet, don’t smoke/drink but live comfortably in a van for cheap. Why? Survival is easy. All risks are not created equal, and following the heard may result in running off a cliff, ie debt, consumption-oriented, spendy, no love/children/friends, workaholism /not enjoying vacations/travel/life and worrying about the wrong things that don’t really matter.

StanislavPetrov2• 9 years ago

You can safely calculate, from your sample, that about 1% of the gamblers will go bust. And if you keep playing and playing, you will be expected have about the same ratio, 1% of gamblers over that time window.

This whole argument is based on the false assumption that all gamblers have the same chance of winning and losing. In reality (and depending on the game), skill results in a massive disparity in outcome among gamblers. This is similarly true for the author's stock market analogy. As the weaker players are flushed out, you should expect the rate of "bust outs" to slow.

dvt3• 9 years ago

Gambling games that require skill are in the objective minority (Blackjack, Poker, and sports betting); Roulette, Craps, War, Keno, Slots, Bingo, etc., are all purely luck-based.

NB: Even in skill-based games, the odds still always (and I mean always) favor the house. Taleb is right. If you play any game for an infinite amount of time at a casino, you will go bust.

dragonwriter4• 9 years ago

Even if the odds favor you on each event, if they are < 100% success and you play long enough, you will eventually lose everything. Gambler's Ruin always wins.

gerard5• 9 years ago

NB: Even in skill-based games, the odds still always (and I mean always) favor the house.

In BlackJack, card counting can reveal plays where the odds favor the gambler. In sports betting, which can be though of as a market with high commissions, skilled players consistently make decisions with positive expectation.

robbrit• 9 years ago

Sure, the people going bust will slow and you'll end up with higher skilled folks remaining at the end, but the argument still holds. So long as your risk of ruin is non-zero you will eventually go bust.

falsedan• 9 years ago

What is this word salad

tritium• 9 years ago

  Recall from the previous chapter that 
  to do science (and other nice things) 
  requires survival t not the other way 
  around?

Uh, what?

I don't understand that sentence? Is "survival t" a typographical error?

mikeash• 9 years ago

I believe the t is meant to be a comma. The author is saying that you can potentially survive without science, but you can't do science if you're dead.

pizza3• 9 years ago

Correct

nartz• 9 years ago

After reading this and a few other sentences, i came to the conclusion that this article was machine written and stopped reading.

ianamartin5• 9 years ago

My god, this guy is so irritating. As usual, he’s not completely wrong, but he is entirely useless. It’s tautology wrapped in a bunch of garbled language and name-dropping.

Lesson #1: you don’t know where you fall in a statistical distribution.

No shit. Really? There is a 1% chance of failure in a game. If I play the game an infinite number of times, I will fail to win 1% of the time.

The alleged twist: if the nature of the game is that a single failure results in you no longer being able to participate in the game, you cannot play the game an infinite number of times.

Color my mind blown. I’m absolutely shocked that Murray Gell-Mann was able to grasp this concept so quickly. A friend of mine has a PhD in Statistics, and I remember once telling him that the probability of a fair coin tossed an infinite number of times and landing on heads approaches 50%. He grasped that concept immediately. I was impressed. We started a company to explain this.

I’m kidding. That didn’t happen because it would be stupid in every possible way.

Lesson #2: Cost benefit analysis is impossible if you don’t know where you are in the distribution.

Somehow, not knowing where you are in the range of possible outcomes means that you can’t know the range of possible outcomes.

The alleged twist: even when you know the range of possible outcomes, you apparently don’t.

I don’t even know where to start with this. I can’t even make fun of it properly.

When you’re gambling, the range of possible outcomes is well-defined for most people. You lose all your money, you break even, or you win. This model is incomplete because it doesn’t take into account human pathology. Some people cheat. Some people take out loans from shady people and lose everything and get a leg broken. Some people commit suicide because they lost a few hundred dollars.

Guess what, if you include human behavior in the model of Russian roulette, it’s also a broken model. People will cheat. People will chicken out. People will want to die and pull the trigger twice. That does not make a cost/benefit analysis “undefinable,” it just makes the range of possible outcomes larger than most people are comfortable with. That’s why most people choose to not play the game.

What Taleb does here (again) is conflate a legitimate criticism of frequentist statistical models with a critique of incomplete decision models.

But the critique of frequentist models is entirely sophomoric. It boils down to, “You can’t do something an infinite number of times. Therefore, everything we know is wrong.”

Fuck off, man. This gets addressed in every basic statistics intro. I’m not saying there aren’t problems with the idea. There are. But it’s the same problem as saying that the limit of x as y approaches infinity is 1. Is calculus fucked and completely broken because y never gets to infinity?

No. Not even a little.

Lesson #3: social sciences are broken.

Completely agree.

The alleged twist: They are broken because the basic concept of probability is broken.

The absolute comically stupid apex of all of this linguistic garbage is that Taleb uses the social sciences to prove how little his argument makes sense.

He can’t tell the difference between a bad model and a broken system for creating models. He proudly claims that probability is broken because the models that people create to estimate outcomes are bad. And he uses the worst possible collection of models for the most complex systems that we know of as proof that the system is broken.

There are absolutely legitimate criticisms not only of frequentist statistics, but of inductive logic in general. It’s only a conversation that’s been going on since, I don’t know, Pythogras? Maybe before?

There is nothing new or insightful here. Taleb is wrapping up and idea that everyone already knows in a word cluster so obtuse my 5 fifth-grade English teacher would destroy as some deep new insight.

But hey, he posted it on Medium, so it must be awesome.

Give me a break.

dsjoerg• 9 years ago

I was hoping to find a comment about how terrible the writing was, thank you. The writing is terrible.

ianamartin5• 9 years ago

I’m not completely convinced that Taleb isn’t an elaborate hoax designed to win the Turing test.

Look, I don’t like being so harsh. But the guy gets really fundamental things really really wrong. He makes me angrier than Gladwell combined with the Freakonomics guys. Which is a lot.

There’s a genre of pop-statistics that is genuinely bad for people. And it all follows the same pattern. Science says x is true, but if you think about it, surprising y thing is true. Because I’m a special snowflake, and so are you.

It’s absolute garbage.

And I don’t even know that much about stats. I dropped out of a violin performance and music theory degree. I just read a few stats textbooks and had some good professional mentors because some of this is relevant to my work as a data engineer.

This is basic level stuff that’s going wrong. The guy is clueless about everything he writes. But I’ll give it to him that he’s a genius at marketing.

He publishes books and makes money off of them. I do not. So he wins that way. But if I ever write a book, it won’t be in the fiction section where his belongs.

lz400• 9 years ago

I think Taleb likes to think of himself as a scholar so smart that has destroyed mainstream scholarship and his contributions are not being recognized by the lesser scholars. But he's smart enough to recognize that he's just not that deep of a thinker or has anything particularly insightful to say (outside of some financial modeling topics perhaps) so he uses all that pedantic prose and name dropping to cover for it. He's certainly cashing on it so it's not like he's going to stop but I can't shake the feeling that he knows, is bitter and that shows in his writing.

sn9• 9 years ago

This seems to be a chapter in his book coming out in Feb. 2018. I'll probably buy it because I usually find his books to be worth reading despite his writing style, but I dearly hope he has an editor to fix the final product from whatever this is.

fanzhang• 9 years ago

Economics and statistics understand perfectly well the distinction between gambling returns that happen in parallel versus gambling returns that happen in series.

For example, in financial economics, suppose you're investing and the returns of each stock during each year is an iid random variable R.

If you split your money over 10 stocks and 1 year, you take the arithmetic expectancy of R to calculated your expected returns. If you're investing in 1 stock over 10 years, you should take the geometric average of R to get your average yearly return.

In other words, if each stock returns 50% chance 3x and 50% chance 0x, you should definitely invest if you can split your money over a pool of 100 independent stocks over the one time period (you'll have > 99% chance of >100% gross returns!), but definitely not if you were forced to invest in one single stock for 100 time periods (you have a >99.999% of 0% gross returns).

I'm not sure why NNT feels it necessary to say the smartest Nobel Prize winning physicists are needed to under this. I'm pretty sure anyone with a rigorous high school math education can understand this. As a matter of point, the above distinction was belabored to me in undergraduate statistics class and a first year graduate economics class, both many many years ago.

qwtel2• 9 years ago

what is most fascinating about nassim's writings is the night-follows-day certainty with which they will cause a flock of bitter nerds to crawl out of their dimly-lit offices to unwittingly mutter one item or another from the "nassim taleb is wrong because he's not one of us" list, which isn't so much to proof him wrong, as it is to quickly find one another so that they can fall into each others arms, delivering reassurances that they are not wrong in opposing him, and that their distinct lack of real-world success stems from the cruelness and wrongness of the outside world, not because their supposed superior insights are not, in fact, so.

Their all-time favorite, to be repeated with ceremonial regularity, is that his insights are 'trivial'. ignoring for a moment that these people also believe it 'trival' to bring a product to market once it has been figured out (by them) in the lab (if they even lower themselves to think in terms of 'products' for the 'populace'), treating his literary writings, which are intended for the general public, which is very much in need of trivial insights, as if they were his scientific writings, then dismissing his scientific writings based on his literary writings, is simply malicious, owed to the fact that he is the wrong kind of academic, who lifts weights and calls people 'imbecile' on twitter, hence reflects poorly on their comic book-like self image of scientists as some kind of quasi-master-race, which is above cursing, emotion and whose physicality is a mere inconvenience, getting in the way of dreaming up ever more distant alternate realities, to be sold to the public as discovery.

almostarockstar• 9 years ago

You should consider using more punctuation. This entire post has only 3 sentences. It's essentially unreadable. You might have a good point, but I can't understand anything you've said.

qwtel2• 9 years ago

i know. it is sort of fun writing it like that, which is why i did it.

soVeryTired2• 9 years ago

I don't see why Taleb thinks Kelly, Shannon and co. are the only ones who understand ruin. Daniel Bernoulli formalised what is now known as the Kelly crterion in the 1700's. Kelly and Shannon's contribution was to demonstrate a connection to information theory, which was incredibly fashionable in the first half of the 20th Century.

It's really quite an old result.

The Logic of Risk Taking

Related Stories