Risk-reward ratio is the comparison of what you can gain to what you can lose on a bet. If you risk $1 to make $3, the ratio is 1:3 (or 3:1 reward to risk). It doesn't tell you whether to take the bet — that depends on probability — but it frames the trade. A bet with a 1:3 reward-to-risk ratio only needs to win one time in four to break even; a 1:1 bet needs to win half the time. The ratio sets the bar: the better the ratio, the lower the win rate you need for the bet to be positive expectancy.
In practice, you define risk and reward explicitly. Risk is usually the distance from entry to your stop (or the max you're willing to lose). Reward is the distance to your target (or the gain you're aiming for). A trader might say "I'll risk 2% to make 6%" — a 1:3 ratio. The discipline is to not take bets where the potential reward is small relative to the risk unless the probability of winning is high enough to compensate. Many losing strategies are ones where the reward doesn't justify the risk: small upside, large downside, and win rate that doesn't fix the math.
The mental model extends beyond trading. Any decision where you put something at risk for a potential gain has a risk-reward structure. Hiring, launching a product, entering a market — you're risking time, capital, or reputation for a payoff. Making the ratio explicit forces you to state what you're risking and what you're aiming for, and to check whether the odds make the bet rational.
Section 2
How to See It
Risk-reward appears whenever you can define a downside (what you lose if wrong) and an upside (what you gain if right). Look for: explicit targets and stops, and decisions where the upside is being compared to the downside.
Investing
You're seeing Risk-Reward Ratio when an investor says: "I'm buying at $50 with a stop at $45 (10% risk) and a target at $65 (30% reward). That's 1:3. I need to be right one in four times to break even, so I only take this if I think the probability of reaching $65 is meaningfully above 25%."
Business
You're seeing Risk-Reward Ratio when a founder evaluates a product bet: "We're risking 6 months and $500K. If it works, we add $5M ARR. If it doesn't, we write off the cost. The reward is 10x the risk — but we need to weight that by the probability of success. If we think success is 20% likely, the expected value is 0.25M - 0.80.5M = $600K positive."
Strategy
You're seeing Risk-Reward Ratio when you choose between two options: Option A has a small upside and a large downside (bad ratio); Option B has a larger upside and a capped downside (better ratio). All else equal, you prefer the bet where the payoff structure is favourable — and you avoid bets where you're risking a lot for a little unless the odds are very high.
Personal
You're seeing Risk-Reward Ratio when you consider changing jobs. You're risking stability, seniority, and relationships for a higher salary or better role. The "risk" is what you give up; the "reward" is what you gain. The ratio frames the decision: is the potential gain large enough relative to the downside to justify the move?
Section 3
How to Use It
Decision filter
"Before committing to a bet, state the risk (what you lose if wrong) and the reward (what you gain if right). What's the ratio? What win probability do you need for the bet to have positive expected value? If the ratio is poor, you need a high win rate — make sure you have it, or pass."
As a founder
Frame big bets in risk-reward terms. What are we risking (time, capital, opportunity cost)? What's the payoff if we're right? What's the ratio? If you're risking a year and $2M to make $500K incremental revenue, the ratio is bad unless the probability of success is very high. Prefer bets where the upside is a multiple of the downside — and be honest about probability. Many failed initiatives are bad risk-reward that looked good because the team overestimated the chance of success.
As an investor
Define risk and reward before you invest. Risk: capital at risk (often 100% in equity) or the drawdown you're willing to tolerate. Reward: the upside case (e.g. 3x in 5 years). The ratio might be 1:3 — you're risking 1 to make 3. Then ask: what probability of the upside case do I need for this to be positive expectancy? If you need 40% and you think the probability is 25%, the bet is negative expectancy even with a good ratio. Combine ratio with probability.
As a decision-maker
When evaluating a proposal, ask for the risk-reward structure. What do we lose if we're wrong? What do we gain if we're right? What's the ratio? What's the assumed probability? Often the ratio is undefined or the probability is optimistic. Making it explicit surfaces whether the bet is rational. Reject or reshape bets where the reward doesn't justify the risk given realistic odds.
Common misapplication: Chasing high risk-reward ratios without considering probability. A 1:10 ratio is useless if the chance of the reward is 1%. The expected value is probability * reward - (1 - probability) * risk. Ratio and probability together determine whether the bet is good.
Second misapplication: Defining risk too narrowly. "I'm only risking my stop at $45" ignores that the position might gap through your stop, or that the opportunity cost of the capital is a cost. Define risk as the full downside you're willing to accept — and be conservative.
Buffett has said he looks for bets where he can't lose much and can win a lot — asymmetric risk-reward. He avoids situations where the downside is large and the upside is limited. His "margin of safety" is one way to improve the ratio: buy so cheap that the downside is capped and the upside is large. The discipline is to only take bets where the reward justifies the risk.
Ed ThorpMathematician, author; early quant investor
Thorp framed every bet in expected value terms: probability of win × payoff - probability of loss × loss. Risk-reward ratio is the payoff structure; probability is the other half. He used the ratio to screen: avoid bets where you're risking a lot for a little unless the win rate is very high. In blackjack and in investing, he sought favourable ratios and then sized position to capture the edge without excessive variance.
Section 6
Visual Explanation
Risk-Reward Ratio — Upside vs downside. Break-even win rate = risk / (risk + reward). Better ratio → lower win rate needed for positive expectancy.
Section 7
Connected Models
Risk-reward ratio connects to expected value, position sizing, and how we handle loss. The models below either complete the decision (expected utility, probability), scale the bet (position sizing), or describe the payoff shape (asymmetric upside, margin of safety).
Reinforces
Expected Utility Theory
Expected utility weights outcomes by probability and utility. Risk-reward ratio is the payoff structure (gain vs loss); probability is the weight. The two together give expected value. The ratio doesn't tell you the answer without probability; expected utility is the full framework.
Reinforces
Margin of Safety
Margin of safety is the buffer between your price and the value you need. It improves risk-reward: you're risking less (you have a cushion) for the same upside. A large margin of safety means a better ratio — limited downside, same or larger upside. The two are complementary: margin of safety improves the ratio; the ratio frames whether the bet is worth taking.
Tension
Loss Aversion
Loss aversion is the tendency to weight losses more than gains. Risk-reward ratio is symmetric in the math — risk and reward in the same units. But psychologically, people need a better ratio to feel comfortable because losses hurt more. The tension: the "rational" ratio might be 1:2; the loss-averse person might need 1:3 or 1:4 to act. Know your bias.
Tension
Probability Theory
Probability theory gives you the chance of each outcome. Risk-reward gives you the payoff structure. You need both: ratio sets the break-even probability; your estimate of probability determines whether the bet is positive expectancy. The tension: people often focus on ratio and ignore probability, or fix probability and ignore ratio. Use both.
Section 8
One Key Quote
"We look for one-foot hurdles we can step over, not seven-foot bars we need to jump. We want to risk little and have the chance to gain a lot."
— Warren Buffett, Berkshire Hathaway
Buffett is describing favourable risk-reward: low risk, high reward. The "one-foot hurdle" is a bet where the downside is small and the upside is meaningful. The quote is a policy: only take bets where the ratio is in your favour. Don't risk a lot for a little.
Section 9
Analyst's Take
Faster Than Normal — Editorial View
Risk-reward ratio is the shape of the bet. It doesn't tell you to take the bet — probability does that — but it sets the bar. If you need to win 80% of the time to break even, the ratio is bad unless you're very confident. Prefer bets where the reward is a multiple of the risk so that you don't need a high win rate to have positive expectancy.
State risk and reward explicitly. Many decisions are made without defining either. "We're launching this product" — what are we risking? What's the payoff if it works? Making risk and reward explicit forces clarity. If the reward is vague or the risk is understated, the ratio is wrong and the decision can be wrong.
Probability is the other half. A 1:5 ratio is great if the chance of the reward is 30% — expected value is positive. The same ratio is bad if the chance is 5%. Don't fall in love with the ratio; combine it with a realistic probability estimate. Expected value = p*reward - (1-p)*risk.
Asymmetric upside is the goal. You want to risk a little to make a lot. That means seeking situations where the upside is large relative to the downside — and avoiding situations where you're risking a lot for a small gain. Margin of safety, optionality, and good entry points all improve the ratio.
Define risk and reward in the same units. If risk is "we might lose 6 months" and reward is "we might make $2M," you're mixing time and money. Convert to one unit — e.g. opportunity cost of 6 months in dollars, or NPV of the $2M — so the ratio is comparable. Vague definitions produce vague ratios and bad decisions.
Section 10
Test Yourself
Is this mental model at work here?
Scenario 1
A trader buys a stock at $100, sets a stop at $95 (5% risk), and a target at $120 (20% reward). So reward-to-risk is 4:1. They say they only need to be right 20% of the time to break even.
Scenario 2
A company is considering a project that could generate $200K profit. The cost of the project is $2M. If it fails, the full $2M is lost. The CEO says the risk-reward is good because $200K is a solid return.
Scenario 3
A trader has a 1:2 reward-to-risk trade (risk $1 to make $2). They estimate a 50% chance of winning. Expected value = 0.5*2 - 0.5*1 = 0.5 per unit risk — positive.
Scenario 4
Two projects: A has 10% upside and 30% downside. B has 30% upside and 10% downside. Same probability of success. You prefer B.
Section 11
Summary & Further Reading
Summary: Risk-reward ratio is the comparison of potential gain to potential loss on a bet. It frames the trade: better ratios mean you need a lower win rate for positive expected value. Define risk (what you lose if wrong) and reward (what you gain if right) explicitly; combine the ratio with probability to decide. Seek asymmetric upside — risk a little to make a lot — and avoid bets where you risk a lot for a little unless the win probability is very high. Applies to investing, product bets, hiring, and any decision with defined upside and downside.
Buffett on asymmetric bets, margin of safety, and "one-foot hurdles" — risk-reward in practice.
Leads-to
Position Sizing
Position sizing is how much you bet. Risk-reward is the shape of the bet. A 1:3 bet with 5% of capital risks 5% to make 15%; the same bet with 20% of capital risks 20% to make 60%. The ratio is the same; the impact is not. Size the position so that the actual risk (position size × risk per unit) is acceptable.
Leads-to
Asymmetric Upside
Asymmetric upside is when the potential gain is large relative to the potential loss. Risk-reward ratio is the way you measure that asymmetry. A 1:3 or 1:5 reward-to-risk ratio is asymmetric in your favour. The model makes the asymmetry explicit so you can seek it and avoid the reverse (risk a lot to make a little).
