Finkelstein’s long game

Daniel Finkelstein Why decision-makers play the long game The Times Saturday Review Jan 12 2019, p14

This is a review of Steven Johnson Farsighted: How We Make the Decisions that Matter the Most Riverhead 2019.

If the decision-making had been poor, even brilliant execution cannot save things. Yet with good decision-making, the situation can be saved even if things go slightly awry … .

The most important piece of advice [Johnson] offers is to spend as much time as possible on what he terms “full spectrumanalysis. That is, to lay out all the possibilities before you fix your view of the situation in front of you.

You generate such as analysis by involving diverse groups who bring different perspectives. And you ask some of them to challenge any ideas you come up with. Or to conduct premortems – imagine we did this and it went wrong, what would we identify afterwards as the cause of our error? What would we decide that we overlooked?

Mathematics plays a role – properly weighting the different outcomes in terms of their likelihood and importance to you – but the author ends up putting more emphasis on imagination and intuition.
Johnson adds a novel idea to the decision-making process. Literally. He recommends reading novels.

The author argues strongly for the role of science fiction, because he believes that one of the hardest parts of long-term decision-making is envisaging futures that at this moment seem almost impossible.
We may, he says, be ignoring what science fiction hints could be a huge problem – that artificial intelligence may end up posing an existential threat to human beings.

[After] reading ‘Farsighted’ you feel as if you have learned something, and that you might be a somewhat better decision-maker, but you certainly don’t feel that you have mastered the topic. It isn’t systematic enough.

[If] you were establishing a course on decision-making, you wouldn’t make this book … the recommended text.


The term was introduced by Gary Klein.

[Prospective hindsight]—imagining that an event has already occurred—increases the ability to correctly identify reasons for future outcomes … the premortem operates on the assumption that the “patient” has died, and so asks what did go wrong. The team members’ task is to generate plausible reasons for the project’s failure.

A typical premortem begins after the team has been briefed on the plan. The leader starts the exercise by informing everyone that the project has failed spectacularly. Over the next few minutes those in the room independently write down every reason they can think of for the failure—especially the kinds of things they ordinarily wouldn’t mention as potential problems, for fear of being impolitic.

Next the leader asks each team member, starting with the project manager, to read one reason from his or her list; everyone states a different reason until all have been recorded. After the session is over, the project manager reviews the list, looking for ways to strengthen the plan … in describing weaknesses that no one else has mentioned, team members feel valued for their intelligence and experience, and others learn from them. The exercise also sensitizes the team to pick up early signs of trouble once the project gets under way. In the end, a premortem may be the best way to circumvent any need for a painful postmortem.

The Guardian had previously commented  commented that:

Daniel Kahneman describes it as his favourite method for making better decisions.

“It’s a sneaky way to get people to do contrarian, devil’s-advocate thinking without encountering resistance.”

But prospective hindsight can be used to imagine things going right, too.

My Comments


  • The notion of the premortem captures some important best practice.
  • The reference to ‘full spectrum analysis’ and diverse groups adds something. It seems to me that many decision failures are because the decision makers were – at least with hindsight – too narrow and closed minded. To be genuinely ‘full spectrum’ one ideally needs a full spectrum of inputs. But many institutions find this difficult.
  • The reference to story telling is insightful, and has become common since Klein’s original work.
  • In my experience getting someone to quote popular if wacky fictional sources can give attendees ‘permission’ to air their ideas without feeling they are risking appearing stupid. It can also be an easy way to open up a conceptual space that might otherwise be overlooked.

But mainly, I agree with Finkelstein that a more systematic approach would be helpful. (Hence my blog.)

Artificial Intelligence

Farsighted refers to AI. Coincidentally I recently discussed this issue with a former colleague who now works with the relevant UK all-party parliamentary group. Something troubled me about their recent report. Reading Danny’s article it seems to me that it falls far short of reflecting an adequate premortem: there is surely far more downside than ‘ethics’. For example, the ‘drive’ towards autonomous vehicles seems fraught with unexplored potential down-sides. On the other hand, I also think that if the objective is improve transport then a premortem might identify a role for semi-autonomous vehicles that would seem to me even more desirable than solely autonomous ones. As Johnson argues, a review of science fiction might help, if only to allay the public’s fears.

Dave Marsay

%d bloggers like this: