Random Baseball Champs
Several years ago, my suburban Chicago town's high school won the state baseball championship. The team played terrific ball, running the eight-game, single-elimination playoff table after rising to the top in the early newspaper polls and remaining for the duration of the season. They were quite talented. Two seniors were drafted by major league baseball, and more than half a dozen were recruited to play in college.
Following the final game, a triumphant convocation was held in the school gymnasium to honor the conquering heroes. Coaches, school administrators and local politicos gloated in unison as the championship banner was hoisted. About 45 minutes into the ceremony, the emeritus ex-coach was invited to deliver a homily. He started out dutifully following the chorus of the others - acknowledging the team's skills, work ethic, resolve, resiliency, etc. - but then, unpredictably, veered off track.
Citing the need to remain humble in the face of success, the ex-coach reminded the team that they had caught several fortunate breaks in route to the championship. Had the opposing pitcher in the semifinals not been afflicted with uncharacteristic late-inning wildness, he noted, the champs might never have gotten the opportunity to salvage a game in extra innings they could well have lost in regulation. Or had several bang-bang plays not gone their way in two other nail-biting sectional games, the team might have been eliminated earlier in the competition. His message: enjoy the magnitude of your accomplishment, but remain humbled in the knowledge that there was at least some luck or good fortune in your success. His corollary warning? Be prepared to be on the other side of such fate at some time in the future.
The reception to the ex-coach's words was not very positive, to say the least. Team parents complained the speech cast a pall on the championship. How dare he minimize the team's accomplishments! The ex-coach attempted to clarify his remarks on several occasions, noting he meant no disparagement to the team, seeking only to ground the high schoolers for their volatile life journeys. He was quoted in the local newspaper as saying he witnessed such capriciousness firsthand many times in his 20 years of coaching. Twice when he absolutely "knew" he had the best team, he fell short in the tournament, losing to "inferior" competition. The year his team won state, the ex-coach had felt they'd be lucky to make the round of 16. In his view, talent, preparation - and luck - were needed to be a champion.
The Random Black Swan
Perhaps it is personal bias - or luck - but I seem lately to be uncovering more business-related books that interest me. Not the formulaic, feel-good, how-to-be-successful tomes, mind you, but meaty books that focus on intelligence, analytics, failure, delusions, psychology and randomness. Three years ago, I stumbled on Fooled by Randomness, by Nassim Nicholas Taleb, while browsing in Border's, and enjoyed the read thoroughly. This spring, I purchased Taleb's The Black Swan, a follow-up that examines the role of rare events with an extreme impact. I vacillate in my assessment of Taleb's prose: at one sitting, viewing his style as irreverent; at another, denouncing it as pompous. Even when the latter, however, I forgive him, for the depth and substance of his thinking certainly more than compensates for his meandering style.
Taleb's combined series, what is playfully called Fooled by a Random Black Swan, is very substantive stuff, challenging readers in philosophy, history/history of science, finance, economics, psychology and mathematics/probability. Taleb is an academic tour de force, having earned both MBA and Ph.D. degrees from prestigious schools. A one-time serious quants trader, he now appears to be more engaged in mundane scholarly pursuits.
The point of departure for Fooled by a Random Black Swan is that the human brain sees the world as less random, and conversely, more well-behaved than it actually is. We often mistake pure luck for skill, and indeed, often elevate lucky fools to guru status. We are wired for certainty, determinism and causality, even when they don't exist. We think linearly, continuously and symmetrically, elevating the bell curve to religious status.
Black swans, those events that are outliers, carry extreme impact, and are not predicted - but nevertheless explained post facto - are of much higher importance than we would like to attribute, according to Taleb. He contrasts two types of randomness, the utopian Mediocristan, which is close to equality, and behaves according to the bell curve with continuous progression; and the winner-take-all Extremistan, with extensive skewness in population values and progression in jumps. Mediocristan, typified by a population height distribution or IQ, is impervious to black swans. Extremistan, typified by a population wealth distribution or sizes of companies, is vulnerable to black swans. Alas, our lives are more influenced by Extremistan than we'd like to think.
Financial markets and trader performance are the grist for much of Taleb's randomness and black swan theorizing. He holds "acute successful randomness fools" from the investment world in highest disdain, citing arrogance and utter insensitivity to the role of pure chance in their success. Such fools almost routinely manifest the psychological flaws of overestimating the accuracy of their beliefs, of being married to their positions, of constantly rewriting history, and of denying their failures. More often than not, they regress from seven figure compensations to trader cemeteries overnight.
Random Wisdom for BI
There is much wisdom for both business intelligence and research to be gleaned from the combined Fooled by a Random Black Swan. The assumption of randomness that dominates Taleb's thinking is an excellent point of departure for BI. Randomness, or the absence of a systematic explanation for performance outcomes, represents the null hypothesis that must be disproven by intelligence/research. Only when a well-designed "study" finds strongly related performance factors should the explanation of simply random results be considered for rejection.
Equally important for BI is design consideration for randomness' insidious cousins, delusions that derive from the psychological and mathematical untidiness of business. Indeed it is here where the observations of Taleb look much like those of Phil Rosenzweig in The Halo Effect. Hindsight bias - I knew it all along - is a delusion that allows history to be perfectly understood after the fact. Experts at back testing stock market performance fall victim - and often lose money - to this bias.
Survivorship bias is a second randomness-related foible noted by Taleb - and a host of others, including Rosenzweig. Survivorship bias can make the results of intelligence investigations invalid because the weak die and are thus not included in the findings - skewing results toward the successes still alive. And randomness may well be the explanatory culprit with such intelligence. Given a large enough starting sample and at least a small likelihood of success, many survivors might be expected by chance alone.
Successful businesses and entrepreneurs seldom accept randomness as an explanation for their good fortune; businesses often use randomness to write off their failures. With a short enough time sampling, a certain amount of business success might be truly random - or at least attributable to factors entirely outside the control of the business. Over the long term, however, Taleb notes the property of ergodicity will ensure that randomness is not the primary explanation for success or failure. Taleb's ergodicity is Rosenzweig's delusion of lasting success, and is in turn methodology's regression to the mean (http://dmreview.com/article_sub.cfm?articleId=1079220).
The Black Swan notes several delusions that are also important to understand and accommodate for business intelligence. Taleb's narrative fallacy, which has to do with how we fool ourselves with stories and anecdotes, is akin to Rosenzweig's, how little we know, the almost comical depiction of the media's discussions of the up and down fortunes of Cisco and ABB. There is also significant overlap between Taleb's errors of confirmation, the bias to look at what affirms existing knowledge, and Rosenzweig's halo effect, the tendency to evaluate someone or something on all dimensions based on knowledge of a single dimension.
Taleb struggles with issues of induction, with the desire to make strong inferences from the particular to the general, to formulate laws based on the occurrence of limited observations of recurring patterns. He refuses to acknowledge the validity of the statement: "No swan is black", after viewing 4000 swans and finding none. However, he will acknowledge the alternative: "Not all swans are white", on observing a single black swan. Taleb finds consolation in philosopher Karl Popper who notes only two types of theories, those known to be wrong (falsified) and those not yet proven wrong (not falsified). Calling this perspective Skeptical Empiricism, Taleb espouses a bottom-up approach to business - ideas from skepticism with minimal theory, that seek to be broadly correct rather than precisely wrong. This course of action sounds suspiciously like evidence-based management (http://www.dmreview.com/authors.cfm?authorid=131), which uses the method of doubt - acting with the knowledge at hand while constantly challenging, testing and confirming the effectiveness of decisions.
Indeed, for business intelligence and research, the risks from Taleb's biases are almost identical to those from Rosenzweig's delusions: overstated findings and conclusions derived from inferior intelligence designs and methods. One potential solution to counter the biases and delusions? Elevating the rigor of methods and designs used to gather information. The evidence hierarchy noted in http://dmreview.com/article_sub.cfm?articleId=1084174 offers guidance.
An article by Paul Saffo in HBR, Six Rules for Effective Forecasting, provides final affirmation for the non-linear, untidy evolution of business success. Saffo's Rule 2 is "Look for the S Curve". "Change starts slowly and incrementally, putters along quietly, and then suddenly explodes, eventually tapering off and even dropping back down." The challenge for entrepreneurs is to make big bets as close to the takeoff inflection point as possible, avoiding the often long startup trajectories, but taking no risk at missing the explosive launch. Taleb addresses such nonlinearity as well, quoting a ditty by Malcolm Gladwell, author of The Tipping Point: "Tomato ketchup in a bottle - None will come and then the lot'll." Business is messy, no doubt!
- Nassim Nicholas Taleb. Fooled by Randomness - The Hidden Role of Chance in Life and in the Markets. Thomson, Texere. 2004.
- Nassim Nicholas Taleb. The Black Swan - The Impact of the Highly Improbable. Random House. 2007.
- Phil Rosenzweig. The Halo Effect - and the Eight Other Business Delusions That Deceive Managers. Free Press. 2007.
- Paul Saffo. Six Rules for Effective Forecasting. Harvard Business Review. July-August 2007.