You will be able to describe a decision as a bet on an uncertain future, the way Annie Duke does in Thinking in Bets.
Pick one thing you told someone this week would "probably" or "should" happen: a project finishing on time, a client renewing, a candidate accepting. Write down the percentage you had in mind when you said it. If you find it hard to settle on a number, that difficulty is the subject of these notes.
Annie Duke is a professional poker player who became a writer on decisions. Her book, Thinking in Bets, argues that almost every decision is a bet. When you choose one option over another, you are betting it will turn out better than the alternatives. You make that bet with incomplete information, because you cannot know the future, and luck plays some part in the outcome.
Poker made this obvious to her. A player never sees the other players' cards. They decide from what they can see, an estimate of the odds and a sense of the other players at the table. A good decision can lose a hand, and a bad decision can win one. Duke's point is that life works the same way.
Taking a job is a bet that it suits you better than staying. Hiring a candidate is a bet on how they will perform. None of these come with certainty. Seeing each choice as a decision as a bet does two things for you. It reminds you that you are working with probabilities, and it separates the quality of a decision from the luck of its outcome. Module 8 builds on that second idea.
Siti is an HR business partner at a regional logistics company. Her director asks whether the new hybrid policy will cut resignations. Siti's bet is that the policy keeps more people than the old one did. She answers, "I think it should help." The director says "Good" and moves on.
Six months later, resignations are about the same. Whether Siti was wrong cannot be judged. "Should help" could mean she was almost certain, or that she thought it barely more likely than not, and neither she nor her director knows which she meant.
Words like "probably", "likely" and "should" feel clear but mean very different things to different people. Sherman Kent, an American intelligence analyst, noticed this decades ago. Colleagues who had agreed on the wording of a report turned out to mean quite different odds by the same phrase.
A percentage fixes this. Suppose Siti had said there was about a 60 percent chance that resignations fall noticeably over the next six months. Her director would know she thinks it is better than even but far from certain, and could ask for more confidence before spending money on the policy. A stated number lets the listener say whether they need more confidence before acting, and it opens a conversation about what would raise it.
A number also clarifies your own thinking. Once Siti says 60, she has to ask herself why not 80 and why not 40. The answers are her reasons, and some of them may prove weaker than she thought.
The number does not need to be precise. On a question like this, nobody can tell the difference between 62 and 65 percent. Rough bands of around 10, 30, 50, 70 or 90 percent are fine. You are after clarity about the size of your confidence, and you should not pretend to an accuracy you do not have.
To check whether the number is honest, use Duke's question, "Wanna bet?" She describes it as a question that changes conversations. When someone confidently states something and is asked whether they would bet on it, they often hesitate and then soften the claim. Being asked to put something at stake makes people check how sure they really are.
You do not need to make real bets with colleagues. Ask yourself silently: would I bet a month's salary on this, a week's salary, or lunch? The amount you would risk is a quick read of your real confidence, and it is often lower than the confidence in your voice. With a team, a gentler version works well: "If you had to put a number on it, how likely is that?" It invites a probability without suggesting the person is wrong.
Very few things at work or in life have a probability of exactly zero or exactly a hundred percent. Projects that are "definitely on track" slip. Clients who "would never leave" leave. Candidates who are "a sure thing" turn down the offer.
Saying 90 percent instead of "definitely" is accurate, and it does not mean you lack conviction. It leaves room for the one time in ten when things go the other way. When that happens, you have not been proved wrong, because you said it could. Lesson 1.3, "Qualifiers and rebuttals make a claim stronger", covered qualifiers on a claim, and a probability as qualifier is the most precise qualifier there is.
The opposite trap is saying 50 percent about everything. It is a way of never being wrong, and it tells nobody anything. Commit to the number you actually believe, even when that feels uncomfortable.
At her next one-to-one, Siti restates her view. She says there is about a 60 percent chance that resignations fall by a noticeable amount within six months, mostly because exit interviews mention commuting a lot, and that she would go higher if the pilot team's numbers hold up. Her director now knows exactly what she thinks and why.
When you give a view on anything uncertain, work through these:
Treat the choice as a bet that this option beats the alternatives, made with incomplete information and some luck. Replace "should", "probably" or "likely" with a percentage. Use a rough band, around 10, 30, 50, 70 or 90, rather than false precision. Ask why not higher and why not lower, and note the answers as your reasons. Check your real confidence with "would I bet on it?" at rising stakes. Avoid 0 and 100, avoid defaulting to 50, and commit to what you actually believe. State the number with your main reason and what would change it. Record the prediction so you can check it later.
Making numeric predictions takes practice, and you only improve if you check them later. The activity below starts your first set, and Lesson 7.4 brings you back to check how they turned out.
Write five predictions about your work over the next three months, give each one a probability, and set a reminder to check them.
Junxiong-WFG Organisation is an authorised representative of AIA Financial Advisers Private Limited (Reg. No. 201715016G).