Revisiting the Turkey Problem

it took me 5 years to finish Nassim Taleb’s Incerto. it took me 5 years because i seldom read more than a page without jumping from my chair to pace the room in conflict. (hence ruminating)

here i want to pick on just 1 of Taleb’s many ideas to glean a better application of it in my personal and professional lives. specifically, the detection avoidance of [negative] Black Swans.

a Black Swan is an event that is unexpected, has extreme impact, and only explainable aftewards. Taleb’s lay example for readers like me is the well-fed turkey who enjoys consistent lunch buffets until one day, in late November, he is another family’s lunch buffet.

this unexpected event was rare (happened just 1/n lived days), extreme (irreversible), and later explainable (Americans express gratitude exactly 1 Thursday per year). it is the 3rd attribute, our butcher’s ability to predict, that i want to explore.

but to first stress why this element is so important, consider 2 more of Taleb’s examples (reminders): 9/11 was a black swan for office workers, not the terrorists; the 2008 crash was a black swan for “experts,” not for Nassim Taleb.

so my curiosity is shouting: how to detect the leading indicators of a black swan? thankfully i know better than to ask “Mr Nassim, when is the next crash?” or “Mr Nassim, how can i avoid a blowup?“, questions he already answered for critics by telling them, simply, to exit their industry.

in my reworked version of the question, without the ear of Dr. Taleb himself, i wonder if we can get closer to the truth via Post-Mortem: Turkey Edition.

hindsight vs foresight

our delicious friend may have avoided his fate with mere awareness of a number of cliche expressions. namely “there’s no such thing as a free lunch” or “if it feels too good to be true, it is.” but alas, turkeys don’t read or have English-speaking mothers. maybe here my hindsight bias is shining.


those who raise animals know they can be trained by causal techniques. for example, if the butcher wants a leaner turkey he may put food on the other* side of the turkey’s habitat, compelling exercise. if he wants the turkey to sport bigger legs he could develop a carrot-stick regimen for that as well. these Pavlovian methods will teach the turkey a thing or two about incentives.

thus through these trainings i reckon our turkey could learn the same lessons as children! which begs another question: what precautionary lessons are we being spoonfed about black swans, albeit through less obvious mediums?

the knower and the sucker

let’s assign labels to our current working black swan examples. the turkey is a ‘sucker’ and our butcher, the ‘knower.’ the terrorists are also ‘knowers’ and the office workers were sadly ‘suckers.’ (no offense)

if black swan events were startups we’d call these “2-side marketplaces,” akin to renters and hosts on Airbnb. but in reality the largest Black Swans appear to be “3-sided marketplaces,” such as in the 2008 market crash where bankers and governments alike (plus some consumers, plus some retail investors, etc) were all suckers. only a sliver of 3rd party thinkers like Nassim Taleb were “knowers.”

so it appears the generation of Black Swans follows a sort of rule: the presence of at least one bad actor. and to profit from them, what Taleb calls “increasing our exposure to positive Black Swans,” we need to identify this actor. but detecting who is who, is not so easy!

in our butcher’s case, while he obviously holds the knife, the consequence of his black swan is forgiven by the jetlagged family sitting around their table. meanwhile in other cases, the butchers (traders, CEOs, pharma “scientists”) don’t even realize they’re feeding a turkey to death. only later do they discover, somewhat ironically, that it was because they slit the turkeys throat that it transphormed into an undead demon.

Intel’s Andy Grove explains in High Output Management that systems are so complex, humans are essentially incapable of understanding them. our best alternative (coping mechanism) is what he calls “looking inside the Black Box,” which bears discussion here.

can we look inside the black box of brewing events to identify how many players are involved in it? if so, could we leverage this information to determine which events to avoid, and in which events to participate?

perhaps i’m just naive. but maybe a nuanced heuristic is to “avoid all 3+ sided black swan territories,” and make small bets only in butcher-turkey-like environments.

dangers of observing the wrong party

bonus compensation may be possible if, in the midst of a butcher-turkey event, one notices another keen observer who is convinced that the turkey is the real butcher. (after all, Turkeys can fly away or even peck the eyeballs out of the attending butcher)

in these cases it might make sense to bet against the observer (who is betting against the butcher), instead of betting on the butcher. our haunted 3-sided marketplace appears again.


does it make more [financial, emotional well-being] sense to avoid black swans, or to look them in the eye through the lens of player counts and leading black box indicators? additional shower thoughts TBD.