The risk reward ratio in every decision we make
Most of us are trying to be right. After many years of building engineering teams, and trading accounts that have charged me real money for the lesson, I’m fairly sure that’s the wrong target. What decides how things end up is rarely how often we were right. It’s the risk reward ratio we accepted to get there.
The term comes from trading. Ignore that part. It works without knowing anything about markets.
Table of Contents
- What the risk reward ratio actually is
- Professionally: the risk reward ratio at work
- At corporate level: every roadmap is a set of bets
- Personally: where it stops being theory
What the risk reward ratio actually is
The risk reward ratio is one plain question: how much can I lose here, compared to how much I can win? Lose ten to win thirty and the prize is three times the price. The other way round, it isn’t.
Take a job application for a role that feels a size too big. It costs an afternoon and a polite no. If it lands, the next three years look different. That’s a very good risk reward ratio, because the win is far bigger than the loss, which means nine rejections out of ten still leave us ahead.
Now the other direction. A message answered at the wheel wins ten seconds. The best case is almost nothing. The worst is that there’s no next time. And it goes right hundreds of times in a row, which is exactly why it keeps feeling reasonable.
Two ordinary decisions, and what separates them isn’t how likely each one is to work. Getting home fine is far more likely than getting the job. It’s what each one pays if it works, against what it costs if it doesn’t.
That gap is the risk reward ratio, and most of the time nobody names it. The course we keep thinking about, the message we haven’t sent, the ten minutes we try to make up on the motorway. Each has an answer to the same two questions: what do I get in the best case, and what does this take from me in the worst. Most days are full of small versions, and the big ones don’t arrive labelled.
Being right is not the goal. Winning more when we’re right than we lose when we’re wrong is the goal.
Professionally: the risk reward ratio at work
The raise that never gets asked for. The cost is one uncomfortable conversation. The win is the money, and every salary after it that gets negotiated up from this one. Five asks and four nos still beat never asking. It gets put off for months anyway, because the discomfort is today and the win is abstract and somewhere later. The discomfort is real. It just isn’t the risk.
The position nobody warns us about is the comfortable job where nothing bad ever happens. The cost never shows up on a day anyone could point to, it’s skills ageing quietly while the field moves on. The win is capped at whatever this year’s pay rise turns out to be. That’s a lot of risk for very little, and every single day of it feels fine.
At corporate level: every roadmap is a set of bets
The bad ones share a shape. The eighteen-month replatform. The big-bang migration. The two-year programme with all the value parked at the end. Huge cost if it goes wrong, decent gain if it goes right, and it has to go right nearly every time. It gets approved in a room where everyone understands the technology and nobody has asked what happens if it doesn’t land.
Most good engineering practice is the same question in working clothes. A handful of users first. A pilot. Small releases, iterations, and things built so they can be undone. None of that makes the prize bigger. It makes being wrong cheap, which is worth just as much and is far more under our control.
Personally: where it stops being theory
Exercise, sleep, putting a bit aside each month. Small, boring, repeated costs against a payoff a long way off. Skipping one on any given day costs nothing anyone would notice, which is exactly why it’s so easy. Over ten years it’s the best deal available to almost anybody.
Then the one that matters more than every other line here: the bet there’s no coming back from. Not a bad outcome. A final one. The risk reward ratio only means anything if we’re still around for the next decision.
So I ask two questions before anything that matters. What do I get if this works? What does it take from me if it doesn’t? If the second answer is everything, I don’t take it, whatever the first one promises. It hasn’t made me right more often. It’s made being wrong cheap enough that it stopped deciding how things turn out.
