From guts to data driven decision making

Source: Wikipedia, License: CC0
There is a wonderful cartoon by Loriot, a German humorist (1923 - 2011), about a couple sitting at a breakfast table, arguing about how to boil a four-and-a-half minute egg. The answer appears simple, but husband and wife argue about how to measure the time using experience, feelings and expert judgment (wife) or a clock (husband).

The whole sketch is hilarious and is often regarded as a fine observation of miss-communication.

Yet, I think it really points out two different approaches in decision making: You can trust your guts or use data/measurements to support your decision.

A lot of industries are going through this transitions at the moment, and this transformation is not without frictions and power shifts, as I observe it in the insurance industry.

Without a stop watch or an egg-timer you need someone with good intuitive time keeping skills to cook a four-and-half minute egg. Experience and a good track record build credibility. Those people are scarce and expensive. With an egg-timer everyone can cook an egg. The magic is gone and with that the high salary for the old egg master. All you need is an instrument that can measure the observations on which your decision is based. Or, so it seems.

Hence, as more data is becoming available I believe we will see a power shift between the 'old guys', who rely largely on their experience, relationships (that means gossip) and guts to convince their management boards and the 'new guys', who use data to validate and support their story, who can compensate some of their lack of experience with analytics and come to a conclusion in a more objective way.

Those 'new guys' will make mistakes, but the smarter ones will work with the 'old guy', learn that there is a lot more context to know to truly understand the data. And they will work with the 'old guy' and the data to test lots of hypothesis/gossip and learn quickly.

Thus, if you are young and sharp, trained in the scientific method, know how to analyse and present data and like to engage with people, then you are well positioned to accelerate your career. Just make sure the clock is actually working! Otherwise it is garbage in, garbage out again. Or, as Seth Godin pointed out, having a clock which is randomly wrong is the worst of all clocks and you would be better off without one.


  1. It already happened in trading, where programmers/stats/math guys took over traders.

  2. Sometimes, I think solid practical experience doing mathematical (and statistical) analyses is more important than training in the scientific method. I interact with scientists on a frequent basis who are performing analyses that are either irrelevant to the research question or are improperly applied to the data. I have to explain things like why a log transform isn't appropriate, or how a significant p-value won't actually be meaningful to the question being asked.

    It's mathematical experience gaps like this that make the "new guys" (many of whom are older than me) less convincing to management over time. Mathematics requires experience concerning the question being asked in order to be applied and interpreted correctly. The intuitive gut-check is the red flag that makes you run back through your analysis and check your work to either find your mistake or confirm the abnormality, and this gut-check is what makes your work valid.

    P.S. Anyone reading this who wants to be more accurate with their analyses, I highly recommend ditching point-and-click software (SPSS, JMP, etc.) and learning a stats software that forces you to be more hands-on with your data. My personal recommendation is R, but folks dealing with larger data sets or who need a multi-user server solution may want to look at SAS, instead. I will personally guarantee that learning either software will drastically improve your ability to do stats.

  3. I agree, mathematical experience and intuition combined with sound business knowledge are essential to know what to do in the first place, which questions to ask and to understand if the answers are plausible, or scientifically correct but nonsense.

  4. Indeed, yet I believe it is important not to loose touch with the traders. The technical skills are absolutely necessary, but not sufficient for success.