July 18, 2007
Balance of Metabolic Pathways, CrossFit Certification Seminar - video [wmv] [mov]
"I don't know" - Nassim Nicholas Taleb and The Black Swan
Amazon book page
Author's home page
Taleb on Black Swans (podcast)
Shattering the bell curve
The axeman cometh
Post thoughts on power laws, predictions, randomness and unknown unknowns
Posted by lauren at July 18, 2007 8:11 PM
I've been looking for an excuse to do some serious power lifts and back flips and unorthodox training methods in typical slow paced body building style gyms for years. Now i finally have it. Thank you CrossFit for your ideology and supporting similar beliefs as myself of your focus on compound movements as the base of your training. I wish more people would understand how much time and energy they are wasting on ineffective, low-impact single joint movements, especially as novices. Keep rockin' out CF'ers
Is anyone training in/around Miami?
30M / 155 / CFT 695
Guess its time to attend a cert.
they'e all gonna laugh at you!
Art De Vany has covered the topic of predictability of complex systems with a fair bit of detail on his blog: http://www.arthurdevany.com/ (mostly wfs)
These seminar clips are awesome. Especially for those of us outside the U.S. who can't make it over in person.
Thanks as always. Looking forward to seeing the series continue.
I agree! These cert clips are fascinating but they always end way too soon!! Attending a cert is not something I plan to do but I'd certainly consider purchasing a DVD of coach Glassman explaining the philosophy behind Crossfit. The "What is Fitness" and "Foundations" publications are great but nothing like hearing it straight from the source.
I thought the review by Mr. Gilder was very sensible. The author's home page and "notebook" struck me as a bit presumptuous. I'm afraid I don't have too much in the way of constructive criticism on the subject. Although we may live in a culture of predominantly gullible and non-observant passengers, I didn't find anything in Talebs observations (although admittedly revealing and understandably addicting) that showed me new ways of conducting business. Perhaps the new book offers more substantial operating applications, but I think I'll save my money for now.
I will say that I do believe that we lend far too much credence to the old bell curve predictions. But I don't see how this necessarily means that our notional, ill-conceived predictions lead us to more error than if we were standing around waiting for the next catestrophic event. Taleb's work offers a broader, more enlightened way to observe our journey but I don't see the tangible techniques that would aid in the "coping" with our world's uncertainties.
I'm sorry everyone is now stupider for having read my post...lol. I'll wait for a more substantive post before I attempt another.
Best seminar clip seen to date. Very helpful for those of us striving to break the fitness comas most are living in. Thanks Coach.
Well, obviously he has some disagreements with previous thinkers on the subject, and thtas ok, there is room for all. Me? I've always believed that there are three types of people in the world, those that can count and those that can't.
Thanks for the face time! And big thanks to Franklin for coaching me through Hellish Helen.
Had a great time at the CFSC Cert. Trainers & Coaches... you all rock! Thanks for the inspiration.
I agree, those videos are fascinating. I literally hang off ever word and yes, they always end to quickly. I too would definitely buy a DVD not being in the US.
Thank god for a rest day.. my lower back and glutes are still sore after Diane...
Just getting started - but the clips and "What is Fitness?" resonate with me. I'm reminded of Heinlein's philosophy:
A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects.
-Robert A. Heinlein
Hey! Pleased to see that I was able to foreshadow the appearance of Nassim Taleb by a day..
As a finance professional, I can say without a doubt that NNT's writings have influenced the way I think about investments and risk. Yes, his writing can be a little arch, but the lessons are extraordinarily valuable. I wish I knew more about derivatives so I could understand his first book on options.
I fully agree with NNT that the stock market does not behave like a normal system with a gaussian distribution. Extreme events happen with remarkable frequency. I think this is because at the end of the day, it's a bunch of emotional people making decisions rather than cold logical machines. And you thought we had an efficient market?
CrossFit Black Swan: Murph two days in a row.
A HUGE thank you out to Mount for introducing me to Crossfit. Been lifting for 15 years and only on CF for 7 weeks, but best overall shape I've been in for a VERY long time!
Got my neighbor and his wife and my wife involved with CF also. Wife is an aerobics instructor and this pushes her limits as well!
Thanks again for turning my weights vs. aerobics discussion to an overall win with CF!!
One thing i never understood was this on the CF FAQ:
"10. The world's most successful athletes and coaches rely on exercise science the way deer hunters rely on the accordion.""
Is this to say that CF doesnt rely on exercise science or that even the best ignore what they shouldn't? I mean...it seems like CF is crawling with exercise science and that statement on the FAQ is what detered me from CF because I am a scientist and amatuer exercise physiologist and I found that statement appauling -- but after CFing for a while u see that CF is based totally on science....so I always wanted clarification on that
i'm definitely not a mathematician, but i understand the power law idea - it's becoming increasingly prevalent in our society as concepts get cattle-rushed by huge numbers of people.
for example, marketing types talk about the "long tail", a power-law concept that describes the long integral area under the low-magnitude part of the curve, useful for describing the products in the market that are not the big-name sellers. if you take a look at amazon's marketing system, they use a "customers who viewed this item might like" system that picks up the long tail to increase profits and individual consumer spending. fascinating stuff.
on a different note: anyone have good advice on what to buy for a doorway chinning bar? i'm over 200lbs, so i need something sturdy, but i'm also poor...
#18 -- you can get a good chin-up bar from www.beachbody.com. its around 50-60$ w/ shipping and handling. its not made for kipping, though.
I've been waiting for The Black Swan to be a topic here... and, I'm so busy today that I won't have a chance to participate until tomorrow.
...wish me luck. Today, I interview with the Reno Fire Department.
If you flip a coin 99 times and get "heads" 99 times, (standard coin, standard toss) then what are the odds of "heads" on the 100th toss? Answer: 50/50.
The Black Swan is right, true predictability is a coin toss. Play the odds all you want, but there is no reason to look surprised when Mr. Grenade goes boom.
I think it is safe to say that in a world of bell-curve "athletes", CrossFitters are the true Black Swans!
#2 ryan. HI i am a girl and very new on this , waiting for get my certification ,maybe next time tthey come to floridA,but starting doing the WOD,week ago,and love it .i dont have all the equipment, but any way.And i a'm in miami too!
#17 - I would think that statement means CF uses what works. I have yet to use a stability ball in CrossFit and I can tell you my core is stronger then it ever has been. I used to have one and gave it away. The 6' $2 piece of PVC pipe I have though is a much better tool.
I have hunted deer for over 20 years and have yet to use the accordion. But I have used scents, camouflage, a bow, a rifle, good camp gear, wind charts, daylight charts and the like. All have helped in some way to hunt deer.
An example of "Exercise science" in the case of this statement is what you find in the popular magazines to make the masses think they are doing something. See the last CFJ for the SB article by Mark Rippetoe, it has a lot of his view on popular "exercise science".
#22, I believe the coin has an equal amount of probability of landing on its edge and rolling off the table into a circulation vent.
I only state this because I have seen it happen, and lost a quarter I was going to use.
on the topic-
"Kid, I've flown from one side of this galaxy to the other. I've seen a lot of strange stuff, but I've never seen anything to make me believe there's one all-powerful Force controlling everything. There's no mystical energy field controls my destiny. IT IS ALL A LOT OF SIMPLE TRICKS AND NONSENSE!"
Some of my engineering work has recently pushed me in the area of emergent behaviors. When a complex system is created from many different smaller systems working in concert, sometimes unintended results emerge. Some of these are minor (like your rearview mirror rattling at a certain RPM in your car) but others can be colossal and dangerous. Thanks to Murphy it seems that the more complex and, as a result unpredictable and undetectable, these emergent characteristics are, the more catastrophic they become. Their at times unpredictability also seems to lend themselves very nicely to Taleb's power-law. His work looks, to me, to be even more interesting to systems engineers than to economists. Given the time, I'll have to look into it a little more, but on the surface it seems very interesting.
If you flipped a coin 99 times and got 99 heads, you should be suspect that it's either not a "standard coin" or a "standard toss."
Consider the probability of getting a specific pattern (of fair coins) in 99 (fair) tosses is 1 in 2^99 or about 1.5x10^(-30). If you consider that the universe is only about 3x10^27 seconds old, and if you flipped a coin once a second, it would take about two trillion times the age of the universe for 99 straight heads to come up. If I were a betting man, I'd say the game was rigged.
Is the guy talking on the video today Coach?
Taleb’s first (non-technical) book is both shorter and more accessible to the masses than black swan. Check out ‘Fooled by Randomness: The Hidden Role of Chance in the Markets and Life.’ Also check out Benoit Mandelbrot, his writings on non-gausian distributions and fractals are great. My favorite book of his is ‘The (Mis) Behavior of Markets: A Fractal View of Risk, Ruin And Reward’ which happens to deal with my career but he has other writings that address other areas of interest. Another great introduction to non-gausian modeling is Garnett Williams’ ‘Chaos Theory tamed.’
On Taleb, he is pompous and his writing can be strained but he knows what he is doing and reminds us that models, for all their complexity, are at best an approximation of reality and at worst are ruinously wrong. I think that people read his books expecting a 12 step plan or a system (like a late night real estate info-mecial) that they can use.
The trouble is life is not formulaic in nature. Sure we can observe the past and make a formula fit it but we cannot with any degree of certainty make predictions. If we could there would be no financial markets because all securities would be priced correctly, we would not have to bother with the Olympics because we would already know the winners in each event and we would know with certainty that we are destroying the planet with carbon emissions. The truth is we don’t know any of those things. People buy and sell stocks, the US basketball team lost, and know one really knows what the weather is going to be like in 50 years. We have to actually live our lives, make the best decisions we can, let time pass and observe the outcome. People are not rational, assumptions are wrong, correlations change and the world isn’t really ruled by a bell curve.
i would propose that any system in which the possible outcomes are determined in part by the consequences of human decisions will be susceptible to fat tails on one or both ends of the distribution curve
in other words, you better have a contingency plan if the unthinkable happens, because it most likely will
About CF not utilizing exercise science...
Here is my take right now. Exercise science, just like all science works in a reductionist realm. With tweezer-like precision they pick out all the smallest parts to find out how everything works. They picked apart the metabolic pathways, the anatomists picked apart every muscle they could and as a result lost sight of the whole. Training routines were thus made to maximize individual metabolic pathways and muscular gains. I don't think that CF contends that extensive time spent in long endurance events won't develop one's VO2 max... or that lateral raises won't develop big/strong deltoids. I think the intension is that working at a high intensity with functional movements, a high power output, is what will improve perfomance in all arenas of human ability. This, to a large degree, is not what exercise science focuses on. In trying to figure out how everything works, we lost sight of everything working together.
This is just my opinion of where exercise science and Crossfit clash. I'm willing to hear of any other takes on this as well...
"you better have a contingency plan if the unthinkable happens, because it most likely will"
But if it's unthinkable, how can the unthinkable happen?
I ask this not to be clever (ok, maybe a little), but to point out that the unthinkable happens, and thus, cannot be planned for, precisely because it's unthinkable.
If the Port Authority had "planned for the unthinkable" (i.e. being hit with planes), then that would mean the unthinkable was no longer unthinkable, because there was a plan to deal with it.
So...how can you plan for an undefined "unthinkable" event?
#17- To expand on what G said, you won't find many guys with physiology degrees coaching professional or collegiate athletics. What you will find are a lot of guys who have been living and breathing that sport for most of their lives. Training camps today look no different than training camps 30 years ago. The quality of the equipment may have improved, but the way it's used is the same. It's what works, it's what has stood the test of time and competition.
To put this in more general terms, the function of the human body has not undergone any profound changes in the last 100000 years, let alone the last 6000 years of recorded history that we as a race have been training our bodies for combat and sport and whatever else. The idea that sports science is suddenly going to come up with some breakthrough move or technique in training that has somehow been overlooked during the past however many millenia is vain to say the least.
A great example of this philosophy is Andrue Swasey, the strength coach for the Miami Hurricanes football team. During Andrue's tenure Miami has earned a reputation for manufacturing some of the best talent in the NFL. Here's a great article about his program:
The best way I can some it up is with the old saying, "You train the way you play." That pretty much says it all. It's just common sense. If you have a skill or performance factor that you want to improve, you do exercises that stress that factor or simulate that skill. That's all there is to it.
The reality of sports science is that all it does is come up with ever more complicated reasons to do what the people who know have always been doing.
I'm not familiar with Taleb's work and can't dash off anything clever or insightful. I'm sufficiently intrigued to buy his book. What amazes me is the never ending cornucopia of interesting subjects Coach and Lauren post on rest days. Don't you guys ever sleep?
Fantastic video! I too didn't want it to end!
#30/gbass - makes me think of that line from the Untouchables; "don't hope it happens, do expect it to happen, just be ready when it happens."
#17/bklyn - Coach's point in that line is that exercise science is not providing meaningful information to practitioners. Kilgore's articles in the CF Journal illuminate why that is the case. Bottom line: no one is taking what 'exercise scientists' are working with and using that information to produce top performing champions. The work of exercise scientists is relevant to fitness practitioners like an accordion is relevant to deer hunting. It's not a rejection of science - it's a rejection of the flaccid exercise science that is happening now.
One understanding of science could roughly be described as form a theory, test it, evaluate the results and refine the theory - then submit your effort to the wider population to see if it can be validated by others who have different biases than you. If you participate in this process, you could described as a scientist even if you don't claim the title. If you buy that, Coach is a scientist and as you observed, uses many aspects of that discipline in attempting to determine what behaviors produce athletic adaptation in humans.
#29 - yes.
Reading the BS stuff, I'm thinking I don't get it yet, otherwise, I'd know why Coach thought it was important enough to post for rest day.
Someone had thought of using planes in an attack prior to 9/11/01, and it was part of a report. That information was simply ignored. And I am not saying this to place blame or anything like that, just saying using planes as a weapon was not unthinkable.
you cannot prepare for all possibilities - obviously
but you can prepare for events that would be outliers on a bell curve
for instance, had the statistical wizards running the hedge fund Long Term Capital Management just put a simple stop loss in place in case something crazy happened, they would have saved 7 Billion Dollars.
cut your losses short people
Sorry this is kind of long, and takes some twists and turns throughout... And it may double post because it got stuck for using stars.
Apolloswabbie, maybe this is why it is important for a rest day.
I am currently reading The Black Swan, and it is fascinating to me. Just so we're on the same page Black swans are rare events that have a high impact that you cannot see coming. If you want to prepare for these events, you would have to imagine them. But think about how your imagination comes up with possible events... so often our predictions are based on history or experience. So with the stock market, you hear people say: "What's the next Google?" But they don't realized that they missed the first "Google", they don't think, oh wait, how did I miss that? But then, when that happens, a narrative fallacy can be created. Where we try to come up with a story about our past experiences, and this 'narrative' we come up with will not be exactly what happened (it will exclude tons of information, that is just too overwhelming to include)... but our "hindsight is 20/20" will make us think that this narrative is accurate... Leading to a certain degree of arrogance for predicting the next Black Swan.
The problem with the Black Swan is its not very important to a lot of people, meaning their lives are not dicated by the black swan daily. But their lives are affected by black swans when they occur. (I hate to use this example, but I think its important)
9/11, the people on the airplanes that were on their way to crashing into the Twin towers were probably thinking: "Ok, this is going to be like other hijackings. They want ransom or some friends released from jail. They will take us somewhere, that is not our original destination, and negoitate with authorities.." so on and so on. Yet the black swan here was not predicted. There was no ransom and no negoitation. But the passengers on Flight 93 (?) heard about the other hijackings and decided to defend themselves and do something.
And this is partly why I like CrossFit (outside of the intelligent driven community)... CrossFit promotes a level of fitness that is not specialized, it is not training for soccer or the pole vault. It is training for the highest level of preparation. Preparation for what? for anything.
And the scary thing about today is that Terrorism is a people's war. Terrorists choose not to attack uniforms because they do not have the means to win against an army. They attack unarmed, unprepared civilians, who are weak. Who knows, maybe one day I will be walking down into the subway in NYC and see something bizarre going on. Maybe I'll ask, "Hey, what are you doing?" as someone with a bag looks up and suspiciously starts running. Maybe since I've done CrossFit for 7 months I will be a little faster and I can catch up with him and see why he is running. Maybe something like that will never happen, but some other thing I can't forsee will happen. And I want to be prepared to the best of my ability.
#35/Appolloswabbie, #25 Angry G, #32/pocketmonster. I think what you are saying is becomming much more clear to me. I should get the CF Journal and read those articles but from what you are all telling me, and what i DEFINITELY agree with, is that the "exercise science" that is in "Men's Fitness" and other garbage magazines as well as Arnold's "Bodybuilder's Bible" is total trash and nonsense.
By the same token, it seems to me (i can be wrong) that you guys forget that the things that are exploited in CF training are heavily based on exercise scientist's findings. Tabata sets are based on a study formulated by I. Tabata et al. in the 1990's...about 14 years ago. We implement those into our sets too because of their exploitation of HIIT's benefits.
The body in itself is still very unknown. The Journal of Neuroscience has some recent articles explaining how our brains can cause variability and error in our movements rather than muscular variability. Little things like that effect our training and can't go ignored. From what I can tell, CF and Coach does NOT ignore this stuff as you can see from the movies of seminars and certs. This is also exercise science in that respect.
I guess the term "exercise science" was just a bit too broad for me....but I understand what the statement was meant to say now, i think..
"Shattering the Bell Curve" Comments;
Mr. Taleb's comments that we should not follow the bell curve but a power curve is as an arguement as ridiculous as they come. The examples he cites are in themselves black swans. We have little choice but to follow a normal probability curve as most occurences in nature automatically follow one. If you remove the normal curve you would have chaos and not in a good way. How would the FDA function? All drugs are ok'd for human consumption based on performance on a bell-curve whether it be for their efficacy or their side-affects. I'm sure we don't want a free-for-all when it comes to drug production. Hey lets just throw some prescription-drugs out into the public and see what happens. I would hypothesize that 99% of adults don't know what a power curve is or that the rules change depending on what type of power curve you are using. Just looking at drug production alone we can see how chaotic and expensive drugs would become if we changed the scale we were using. Theory isn't always the best medicine so to speak and maybe Mr. Taleb can afford increasing drug prices but I sure can't. He is right in saying income doesn't follow normal probability law, but anyone that has had a basic algebra course already realizes that. I'm trying not to flame Mr. Taleb but it seems his his arguement is designed for publicity not reality.
Chris I agree with you completely. Much of what CrossFit is about is backed by good undeniable science and scientific study. I remember awhile ago one of the videos showed Brandon being hooked up to a machine to try and determine the effect on the heart from a CrossFit workout. I would love to see the results. Coach states that in the video today, there is science behind CrossFit. The Tour De France riders would die shoveling snow in Canada for any period of time due to their specialization. Also we know that the aerobic and anaerobic pathways differ greatly, but we also know through CrossFit when we have short-duration high-intensity anaerobic workouts (many of the girl named WODs) it has an effect on the aerobic pathway (my 10k time goes steadily down, even though I run only when prescribed by WOD).
I think in no way does Coach ignore good science. Because of this, CrossFit creates good science to pull data from.
#28 Chris H. “If you flipped a coin 99 times and got 99 heads, you should be suspect that it's either not a "standard coin" or a "standard toss."
And others with the same hypotheses…I must agree with your conclusions…So if we were betting men in say Vegas…the odds spinners would be betting (Probably, I am not a statistician or bookie) against the coin falling for the 100th time on the heads position…Giving the best odds on the coin not hitting the heads again.
The best wager would be money on the coin landing yet again on the heads…Since we seem to agree the coin is either not a real coin or the toss is somehow rigged…but if this were the case, why couldn’t the rigger make the coin do what they wanted thus making them the most profit based upon what the gamblers have decided? Wouldn’t this seem to fall within the auspices of Mr. Taleb’s theories?
Seems that many systems out there do exactly that, to contrive the most profit for “the system” with the average Joe’s out here trying to “Outsmart” the systems through logic and rational via “Odds” and utilization of practical theories when most of it is simple luck and what many try to believe as some divine intervention?
In the example of the coin toss, I am quite sure I would have lost my shirt, house, and everything else way before and prior to getting close to witnessing the 99th toss and having anything else to wager on the 100th toss being either…Since the primal law of averages has told us the coin should land even at 50/50 over time. I would be one of the sorry saps who after seeing the coin land on heads for the 50th time would continue to place hard-earned possessions and anything else I could borrow on the coin’s tail being due…
Simply put, this is why I do not enjoy gambling…Vegas did not get that big by people winning…
While a lot of natural phenomena can be modeled with a Gaussian (bell curve) distribution, it is hardly the only useful one. For example, an exponential distribution models the rate at which electronic components break down and also radioactive decay, the Pareto (power law) distribution has been shown to successfully model financial markets, and the Cauchy distribution is a solution to some interesting differential equations in physics and also models absorption spectra.
Some of those distributions, like the Cauchy distribution and certain varieties of the Pareto (power law) distribution, have infinite or undefined mean and variance, making statistics based on samples from populations so distributed largely useless for prediction.
Taleb argues for identifying the correct distribution for various events rather than defaulting to an "easy" one like the Gaussian simply because we're fond of our ability to make predictions with it.
That Taleb's examples are "black swans" is precisely the point. The probability of such events, if a Gaussian distribution were an accurate model, is so low that they should never happen. Not only in our lifetime, but in many many lifetimes of the universe. See my previous post (#28) for an example of a similar sort of event.
This isn't doing something for the hell of it, rather it's for getting the model to better fit reality. That is: doing good science.
#40 Angry G
"I think in no way does Coach ignore good science. Because of this, CrossFit creates good science to pull data from."
Thats the best point i have heard in a long while
Not quite. The numbers guys would say that barring a scam (you bet heads, the coin tosser forces the coin to land tails. you bet tails, the coin tosser forces the coin to land heads), the coin should land heads with such certainty that they'd be willing to give you the world and everything in it, if you bet $1 on tails and won.
But you are right that it does come under Taleb's ideas because it does describe such an extremely unusual event that one must question the model that describes the event (in this case a fair coin that lands heads 50% of the time, tails 50% of the time).
The Axeman article does eventually get around to what I have long considered to be the dark side of the 'Black Swan´ mentality, which is the defeatism it sometimes engenders. I wouldn´t go so far as to say that a deep respect for randomness fosters disdain for "imaginative and creative faculties" or "excellence and insight, elegance and grace", but it does promote unwarrented agnostisim.
But I´d take a few lazy "you never can tells" if with them comes the ability to cut through the layers of logical falacy that pollute modern existance. Heh. "modern existance2 is such a posuer phrase. Specifically I care only about NFL sportscasters, who are so mired in fallacy that the gmae is unwatchable.... Argghh!!! And yes, I tried MUTE, but they blast their meaningless stats all over the screen! I've got your 20 carries right here yah pinko!!!!
Ok, sorry. It´s getting close to the fall. I get a litttle tense is all.
Perhaps you all miss my point. Bruce Lee said, "It's like a finger pointing at the moon. Don't concentrate on the finger or you will miss all that Heavenly glory". It is not the coin that you need to concentrate on, but the grenade!
The manual tells us that a grenade fuse will burn between one and five seconds. That is a (+/-)two second margin for error. In the case of a three second fuse it is not comforting to think that the manufacturer's MILSPEC allows such loose tolerances. But the Black Swan might say that we wish to avoid thinking about the mathematic probabilities and rely instead on a comforting narrative. So here it is; The important thing to remember is that once you pull the pin and pop the spoon, Mr. Grenade is no longer your friend. In fact, he is equally angry at everyone in the room.
I'm interested enough in the math theory and philosophy to buy the book. But I'm disinterested in anyone as a human being who thinks so highly of themselves that the rest of us are merely "turkeys" because we don't see the universe the same way.
"Don't bridle, though, at Taleb's elegant scorn." Sorry, I'm bridling. But maybe Taleb is being mischaracterized by the author.
Einstein was so far (intellectually) beyond most of humanity and yet he was a humanist at heart. No contempt from him - yet Taleb is going to condescend to the rest of us? How about this, bud - how condescending are you to "simpleton" police officers when you're getting mugged? And how useful is their "bell curve" view of where crime happens when they happen upon you in a hotspot? I got your "power law" right here.
Okay, so that's not terribly helpful. This turkey will probably buy the book and discard my previous "poultry" view of the universe...and kick Taleb's a$$ if I see him in the street.
Just to prove his theory correct, of course.
#2 and #24
I'm in Ft lauderdale relagated to LA Fitness, unfortunately. Been training about 7 wks haven't missed a day. I'll be in south beach for ESM sprint triathlon in Sept. I think.
# 46 Dale
I think the author of that review is partially grinding his own axe, as he's from the Discovery Institute. The notion of random improbability doesn't settle well with divine intervention.
I am basically a Taleb fan - having four of his books (two editions of Fooled By Randomness).
In some of my past work I had to fight against the entrenched use of the Gaussian distribution.
On the other hand, like Hume and Popper, I think Taleb goes a little too far and gets a tad arrogant in defending his position.
If we forsake induction then you can forget Crossfit's "black box". Just because Coach has been successful in the past doesn't give us a clue about what will work tomorrow (according to Popper).
Fact is, most of what we call progress has been based on induction. Newton's laws allowed us to build bridges, automobiles, airplanes, etc. - all based on induction, and proven by Einstein to not hold in all cases. I'll still take the progress, and Coach's methods.
I must quote E. T. Jaynes (PhD in physics from Princeton, professor at Stanford and University of Washington, St. Louis):
"Philosophers have argued over the nature of induction for centuries. some, form David Hume (1711-76) ... to Karl Popper ... have tried to deny the possibility of induction, ALTHOUGH ALL SCIENTIFIC KNOWLEDGE HAS BEEN OBTAINED BY INDUCTION." (emphasis added) from "Probability Theory, The Logic of Science"
Dale (#46) I read this part of one of the reviews (by Eo) at amazon.com:
"The second thing that bothered me is that Taleb seems to be somewhat amoral to me; in this type of book where plenty of his own emotions come through, plenty of his personality, he has plenty of criticism of others for their wrong models and wrong view of the world, and how this has hurt the world, but there remains a lack of moral responsibility to his advice."
This is readily seen in his notebook page, and I think that is why I have found his writing style rather distasteful. I like his theories, but the man himself seems like he'd be rather lonely monitoring the hedge fund he is in charge of, and not someone I'd like to spend day-to-day time with. I prefer people who teach with the notion that the teaching and learning can be fun and create motivation, not just telling everyone they are full of crap and to figure it out.
So am I saying everything needs to be flowers and candy? No, of course not. But I would venture a book called "Going Postal" by Mark Ames where he looks at causes of office killings. The harsh approach is like sandpaper. Nobody likes to be rubbed by sandpaper.
Sick yesterday so I did it today.
100# I didn't even make it halfway
85# very close but not a full PU
26m 150# 5'9"
Did the squat cleans today.
215 PR Nice form too.
225 Fail x2
215 Fail x2
185 Fail for form break.
Good stuff. Thanks CF.
This was catch up day for me today. I am still nursing a bruised rib from friday but it felt better so I went ahead and practiced OHS again. Getting better with balancing the weight overhead, but the shoulders get tired and it made me bail twice on the last rep on 2 sets at 85lbs. I simply wasnt committing the shoulders enough. Still managed to do 6 Sets of 10 reps.
80-85-85-90-90-90(PR). I am steadily moving towards my end of summer goal of 100lbs 10 reps. I hope to get it in a month maybe.
Did Box Jumps Scales (3 sets of 10 reps on 12"-18"-24"-30"-12" boxes).
Did Firday's BP 1-1-1-1-1-1-1 (on the smith machine because shoulders were pretty tired from OHS). 185lbs-195-200-205-210-205-205.
Next rest day I'll probably do the weighted Pull Ups. CrossFit Rocks!...
To: Comment #36 by bret kleefuss
Well, it was ignored. Which is not the same as "unthinkable" in my mind.
The fact that someone thought of it, makes it a "thinkable" act. Thus, since it's not unthinkable, it can be planned for.
Regardless of whether it actually WAS planned for.
I think the closest thing I've read about the "plan for an unthinkable act" was the comment by Gbass (#37) with reference to LTCM putting in a stop-loss on the fund.
It's my opinion that in some instances, "planning for the unthinkable" can work well, in others it cannot.
As we see in this instance, a hedge fund could put in a stop loss to prevent a meltdown of the fund's value if something whacky happens. In this case, they don't even need to define what "whacky" might mean...they are planning against a wide range of things (yet unknown) that could affect the value.
In other circumstances, the capability to plan for the unthinkable may not be possible. Someone already thought of the possibility of someone flying planes into the WTC.
I have thought about the possibility of a plane flying into my house. However, the probability of it happening, Taleb notwithstanding, is small enough that I won't be spending money to harden my rooftop against the odd chance a Boeing 737 decides to test Gravity's Will over my home.
I second what's been said above; I would definitely buy Coach's lectures if they were ever sold.
no rest for me. i did livelong CF's "black"
w/u 800M on elliptical (6')
2x10 fat man pullups
15' more on elliptical
80 walking lunges
pretty fun workout. well... you know for crossfit. ;-)
20 min AMRAP
5 OHS BW (180)
5 Pull-up w/40# DB (biggest we have)
Rest 5 min
1 round for time:
10 Broad jump
21 swing 1.5 pd
21 wall ball
5:52 - pretty smoked from prior
Rounds in 20 min of
Jump rope x 100
Back extensions x 15
Ring dips x 7
9 rds +80 skips. Easy pace.
hurt my foot yesterday, so no running today and had to take it easy at the gym.
30 man makers (15#dumbells)
shoulder press: 65-65-65-65-65
bench press: 75-75-75-75-75
weighted pull-ups: 15-15-15-15-15
my right shoulder has been acting up again, bench and pull-ups were painful. I guess I should take a few days off soon, but I just don't want to!
no rest for me.. go hard or go home lol
ran 4km to the gym in the rain with my 10lb backpack haha i wanna add like 5lbs just cause of the wet clothes haha
anyways got there, stretched for a bit
30 double unders
10x 95# x 2sets -clean and jerks
10 x95# x 1set -clean squat jerk
12x bw+25# x 3 sets -bar dips
6x bw+25# x 3sets -pull up hands facing away
6x bw+25# x 3sets -pull up hands facing towards
50 double unders new PR- 28 in a row
ran back home in the rain 4km with the soaked backpack
good times i tell yeah
o and 21/m/160# 5"9 for the above
Did a track interval (220 repeats) workout today, since it had been a while and I got my new running shoes a few days ago.
8 220s, range 34-39 sec each, plus a couple 100s afterward at about 15 sec. Especially fun in the hot sun.
#34 Tim W and others:
You would be surprised how little is unthinkable, because there are intelligent people being paid to throw ideas around and think up these "unthinkable" scenarios and how to counter them. If you can imagine any scenario for some sort of black swan attack, rest assured someone has brainstormed it to death already.
Taleb had(has?) a hedge fund called Empirica. The premise of the fund was, basically, to bet on black swans since, in Taleb's estimation, the market is so poor at pricing this risk properly. Last I heard the fund was returning money to investors and being wound down.
I might agree that the market misprices extreme events, but I don't think they are systematically over or under priced, Taleb's results might confirm this.
Ben - thank you for pointing that out. I got to the bottom and while I didn't have the sense that he was "grinding an axe", I did sense that Taleb's work gnawed at some core belief. Notwithstanding, and though I haven't read the book, I did peruse some of the author's page and I think 2 other things bothered me.
1 - Mr Taleb: Dude, you've got a theory. Relax. A lot smarter people than you have postulated bigger stuff and been dead wrong, so get over yourself.
2 - Little to none of the people I know live their lives based on "Gaussian distributions". How many people have even taken stats. Seriously?
In fact, I believe the opposite is actually true. If people lived by the bell curve, why is insurance so prevalent, in almost every walk of llife? The statistical probability of a flood in many areas of the country is about three or four standard deviations away from the mean of that curve, but people still buy flood insurance just in case, because while the "probability of loss" is minimal under the standard bell curve, the risk of loss is simply unbearable. People worry about the black swan, not the odds of the unlikely occurrence.
The same is true for almost all kinds of insurance.
The same logic applies to lotteries. The "odds" say 1 in 176 million, yet millions of people play, hoping for the black swan to land on them.
Contrariwise, people build condos in areas very likely to be hit by hurricanes and then tell themselves it won't happen to them - the inverse black swan. (But likely buy insurance at a higher premium, just in case).
So, I don't know. I'll have to read the book, I guess, right after Harry Potter 7.
I subbed KB Swings in Angie because I have a date with Fran tomorrow.
100 kb swings 16kg 1x20 + 8x10
100 push ups (hopelessly broken)
100 sit ups 50/20/30
100 squats 5x20
Next time I'll break 30 with pull ups!
Based on the reading list tangential to Dr. Taleb’s actual writings, he is out-to-lunch on the subject of his discourse. He is a philosopher, but a philosopher-entertainer, not a philosopher-mathematician or -scientist. He has an MBA from Wharton, and a Paris PhD in financial mathematics, of all things.
The Black Swan theory is not new. It is a repackaging of the Purple Cow nursery rhyme. It is also a hypothetical for the truism, absence of evidence is not evidence of absence.
If Taleb did in fact generalize “about the wealthy--whom he regards as ‘inelegant, dull, pompous, greedy, unintellectual, selfish, and boring’”, this does more to calibrate Taleb’s success than anything else. This is congruent to Dennis Prager’s perpetual harangue about PhDs, another philosopher sans credentials. Not that there’s anything wrong with that; Eric Hoffer is an example. It’s just a warning that one should choose his targets with care.
I noted that Gilder left Katrina a Black Swan by implication, but Chernobyl an ex post facto Black Swan. The Katrina disaster was in fact predicted with great accuracy (the levies were only good up to a slow moving category 3, and that was indeed Katrina). It was a Black Swan only in the media and on the Left.
Responding to Gilder, some humans are indeed turkeys. Consider people who worked for Stalin, or now Putin. Or Kamikazes, and those modern versions, the terrorist suicide bombers.
If Gilder is right, that “Taleb believes that these statistical patterns show that the world is chaotic and unpredictable”, then science does not exist. Gilder would be right again that Taleb is the fool of randomness.
Taleb harbors misunderstandings of his favorite topic, the Gaussian (aka normal) probability distribution. He’s right that its misuse is commonplace, but from that observation, he trivializes it.
Suppose one were to lob rounds or guide missiles to a distant surface target. With a few generalizing assumptions, the distribution of the weapons around the target is well represented by a two-dimensional Gaussian distribution. That is, pick any two right angle coordinates, Cartesian x-y or North-East, the miss is normally distributed in each axis, centered at the target or not.
Now ask the question, what is probability distribution of the miss distance? This is not Gaussian at all. It resembles the distribution of wealth more than it does a normal distribution. In fact, it generally has a Rayleigh distribution. It’s zero probability density at zero distance, and grows linearly to a maximum, then declines rather like one side of a Gaussian distribution.
What this example demonstrates is that the probability distribution depends on the coordinate system. The problem is set up as a two-dimension normal distribution, but then the question asked is the radius in a polar coordinate system. There are scores of well-known probability distributions, but they coalesce into just a few by a change of coordinate systems. It’s all in how you ask the question.
For Chris H #46, if a particle’s energy is exponentially distributed (a condition for thermodynamic equilibrium under Planck’s quantum hypothesis), then its displacement along one coordinate is Gaussian.
The Taleb observation is Gaussian misuse, but that can sometimes be traced to a problem with the coordinate system.
Probability theory lies at the philosophical core of epistemology and science. An abstract discourse on probability theory then is the philosophy of philosophy of philosophy -- a playground, or a nest for fuzzyheads.
"2 - Little to none of the people I know live their lives based on "Gaussian distributions". "
I think you are missing the point there. Most people, in all they do, make some sort of informal assessment about the possibilities for benefits and losses in their choices. Agreed, this is definitely not using a bell curve in a formal sense, and mostly not in some loose sense either.
When formal statistics is applied, however, there is indeed a bias for using the bell curve. Formal statistics are applied in so many areas and on so many levels of society that everyone is living their lives in a world to some part constructed through choices others have made with the bell curve as a basis.
The strength of the bell curve is just that, it's a quantified basis for decision that gives you credibility more so than the everyday informal reasoning and thereby giving you an advantage in convincing others about your ideas and making a decision in your favor. That the model does not perfectly match reality is usually of lesser meaning to you, than that power to convince others.
Being skeptic towards ingrained "bell curve credibility" is very sound IMO.
Dale may be missing Taleb's point, but he makes a good point about people's faith in their particular pet black swan. It seems that Taleb may have fallen victim to his own hubris by putting such faith in his pet black swan theory that he sank his hedge fund by neglecting to give any importance to narrative context in which the market selects among black swan securities (i.e., what gives a stock/company real value).
What is unpredicted is not necessarily unpredictable or unknowable. It simply is far enough out of the ordinary that the majority of people do not give it credence. Clearly the venture capitalists who invested in Google saw the potential, if not the inevitability of profits. The narrative context retains value inasmuch as it is founded on accurate analysis. Coach doesn't pull the WOD from the hopper every day. If he did, we wouldn't spend so much time on the central functional movements of pullups and squats or have a logical sequence to our workouts.
The power law is an important distribution to consider in evaluating information (perhaps especially for people like the FDA where overreliance on Gaussian distributions could (and sometimes does) miss side-effects on marginal populations (e.g., children).
Up front, I may be looking at this the wrong way. I guess I am a simpleton. This talk of bell curves, power law, and black swans is moot.
I don't gamble unless I can bet on Me.
The gem is found through action. Not prediction.
If something is to happen consitantly, you have to remove luck. If something isn't going my way, I need to learn to make lemon-aid out of lemons.
It does not matter to me if it has not been successful before or if it always has been sucessful, I am the deciding factor. To do otherwise is to give power to imaginary hobgoblins.
Never give me the odds.
I was amused by Tariq's mention of Taleb on 7/16 in his comment and then Coach posting article's about him the following day.
#71 Jeff Glassman-
I, like Dale, was intrigued by Taleb's thoughts enough to consider buying the book to read, but turned off from Taleb's arrogance. Perhaps now I will just see if the library has the book. I'm curious if coincidences are covered and what the thoughts are on those.
At any rate, it was a fascinating set of articles.
Thanks Coach, for the past few rest day articles, they've been great.
Comment #43/ArrowGift: Not the point of your post, but if you study the FDA, my guess is you'd find they are failing to provide a benefit to US health, driving drug costs much higher than needed, and in general are an obstacle to medicine vice a boon.
The FDA has killed Americans by restricting their choices in therapies and driving costs higher, and I've yet to see any convincing effort that those deaths are balanced by any protective effect the FDA provides.
#41/James, Aha! Thanks - yes, that's the connection. Crossfit is preparation for life's BS events. Paul
The force is strong in you. I love the Han Solo quotes.
I 2nd Kate's thanks for Jeff's post. Generally, Jeff, I enjoy chewing the meat of the bones of your posts for a couple of days. Downside of that is that I often don't have anything to respond with until days later. That's OK - two ears, one mouth was for a reason. Paul
Comment #67 - Posted by Chris M
#34 Tim W and others:
You would be surprised how little is unthinkable, because there are intelligent people being paid to throw ideas around and think up these "unthinkable" scenarios and how to counter them."
Exactly my point.
They say 72 percent of statistics are made up right there on the spot
And 85 percent of people believe them whether they're accurate statistics or not
Now I don't know what you believe but I know that there's no doubt
I need another double shot of something 90 proof, this is too much to think about
- todd snyder
did kind of a hybrid workout today:
120 pull ups/120 dips: 19:37 (not really an elite time but im working my way down!)
1 mile run
10 minutes max sets of: 10 steps 35lb lunges
10 ankles to the bar
1 mile run
todays WOD looks good..gonna save it for next week sometime
wow, Greg is a powerhouse.
I know it's bad when Greg can't smile (at the end of the WOD he tries to smile - but it comes off weakly due to pain, I am sure).
Nice to see you in action, Amundson.
Nassim Taleb is one of the great thinkers of our time. 'Dynamic Hedging' was very influential in derivatives trading and risk management, 'Fooled By Randomness' is a great book, especially its treatment of Monte Carlo simulations, and I am thrilled to read 'Black Swan' since it tackles another issue, outlier events that sit in the tails but happen more often that that would suggest. His work with Mandlebrot is unique in Financial Mathematics and I appreciate his philosophical, almost Hume/Popper approach. I am thrilled to see CrossFit is "in" on Taleb and his approach to thinkning!!
I've read Dynamic Hedging and it is a decent book, but most of the experienced derivatives traders and risk managers I know, are of the opinion that the book shed no new light on the subject and just packaged what traders already knew.
Functionally, what use is the observation that jaw dropping and uexpected events occur from time to time? Duh. Anyone who fails to understand this should be washing dishes somewhere.
The more interesting question is how we engineer resilience into a system. Unexpected event X happens. Social System A collapses, Social System B makes it through, Social System C gets better and stronger. Those are approximately the options available.
I was listening to a set of CD's on the History of Freedom, and in it the lecturer--Rufus Fears--describes the battle of Cannae, in which the entire main Roman Army was destroyed by Hannibal. This should have led to the Rome's defeat and subjugation.
However, two things happened: 1) Hannibal determined to seek a negotiated peace, for whatever reason. That decision was pivotal. 2) According the Greek historian Polybius--who seems to have been one of the principle influences on James Madison, who was the "Father of the American Constitution"--it was the Roman Constitution, based on Republican ideals, which ultimately, systemically, enabled the Roman social system, culture, and empire to endure, and ultimately destroy Carthage.
It is quite obvious, once one learns how the Romans ran things, and what the meaning and use of the Fasces actually was within the system, how indebted our own system is to theirs.
Pointing out where people are dumb can be useful, but only if one either offers better specific information, enabling greater "smartness" in some specific sphere, or offers better tools for thinking to engineer systemically better thought processes which are generally applicable.
His basic thought that unexpected small things can make large differences is, to me, neither new nor particularly interesting, and certainly not enough to sustain his apparent Gallic hauteur.
It's what you do--what you can be predicted to do--when you get knocked down, that matters. And that we can in fact prepare for.
I will add more generally that this appears to me to be an example of what I have elsewhere termed "failurism". This is the intellectual equivalent of moral equivalence. Ultimately, he seems to be arguing, if we can't predict the future, we shouldn't try.
But in point of fact, many people do predict the future accurately most of the time. And the fact that sometimes things go SNAFU does nothing to change the intrinsic value of planning and thought.
I had one more idea "pop" in this morning that was very interesting. I had to check the dates, but they are unambiguous: the Black Swan quote has to be based on William James' assertion that "in order to disprove the notion that all crows are black it is sufficient to find one white one."
What is interesting about this, is that James is arguably the greatest thinker America has produced, and yet he is almost unknown. His influence on this epistemological bon mot was apparently not remarked upon by anyone.
Reason? His interest in the paranormal and specifically spiritualism. That particular quote was addressed to sceptics. His understandable and correct contention was that in order to disprove the idea that there is no such thing as a ghost, it sufficient to find one.
However, due to the ideological impurity of such notion, it has appparently been found necessary to create a version of: "very nice idea, Prof. James, would anyone here that is respectable care to repeat it for inclusion in the public record?"
All of this is very, very interesting.
bike circuit 34:17
Thanks for the links on today's post. At first I wanted an explanation with each and was wondering "what are these about?" then I realized that by just posting the links Lauren was suggesting I take a look and think for myself.
Thanks again for pushing all of us to think and act in new ways and not giving it to us the easy way like we've been taught to expect.
plz give some diagrams of different probability curves.