What is the regret minimization framework?

It's a decision heuristic Jeff Bezos used to leave his Wall Street job to start Amazon — imagine yourself at 80 looking back, and ask which choice you'd regret more. The framework is asymmetric: low-probability + high-regret actions (creative bets, life-shaping asks, irreversible windows like fertility, relationships, or specific job openings) often clear the regret bar even when financial EV is negative. The calculator expresses the framework as math — multiplying honest probability, upside, downside (after reversibility), and a dollar-equivalent regret cost over a lookback horizon.

How do I price 'annual regret value' honestly?

Two anchors. First: ask yourself 'how much would I pay annually to NOT have this regret in 30 years?' If you'd pay $5K/year for 30 years to avoid the lingering 'what if,' that's $5K. Second: scale by the regret category — passing on a small windfall ≈ $500-2K/yr; passing on a meaningful career bet ≈ $5-15K/yr; passing on a family / relationship / creative life-defining call ≈ $20K+/yr. Most people under-price by 2-3× because regret feels distant; the future-self looking back rarely thinks the dollar amount was the issue.

Why multiply regret by probability?

Because you only regret skipping a path that WOULD HAVE worked. If the action was 30% likely to succeed and you skipped, you don't regret in the 70% of timelines where it would have failed anyway — you regret in the 30% where you missed the chance. The honest regret cost is probability × annual-regret × years. Without this weighting, low-probability paths get over-credited for regret cost.

What does reversibility actually mean?

% of the financial downside you can recover. Quitting to start a company with a strong professional network (you can return to a similar job within 6-12 months) is ~80% reversible. Moving abroad and burning the apartment lease + selling everything is ~30% reversible. Investing $50K in equity that vests over 4 years and could go to zero is ~10% reversible (you can't 'unspend' the capital, but you keep the experience). Be honest — most attempts are more reversible than people think.

What does 'attempt for regret' mean as a verdict?

It means pure financial EV is negative (math alone says don't), but the regret cost of skipping over your lookback horizon flips the math positive. This is the genuine Bezos use case — actions where the dollar math is bad but the soul math is decisive. Most career-defining bets land here. The verdict is honest: you're trading expected money for expected non-regret. The calculator surfaces both numbers so you see the trade.

What's the breakeven probability number?

The probability at which pure EV = 0 (regret excluded). Below that, financial math says don't; above, financial math says go. Example: $100K upside, $20K downside, 70% reversible → effective downside $6K → breakeven = $6K / ($100K + $6K) = 5.7%. So you need to believe success is more likely than 5.7% for pure financial math to favour the attempt. Regret-weighting then adjusts from there. The breakeven number is a sanity-check — if it's high (40%+), the action genuinely needs to be likely; if low (<10%), even modest odds clear the financial bar.

What about non-financial upside / downside?

Translate to dollar equivalents — what would you pay to have / avoid them? Reputation hit from a public failure: $5-30K depending on visibility. Network expansion from a year of trying: $10-50K. Health impact from extended stress: $10-100K (use the equivalent of a year of premium healthcare). The framework requires dollar-honest inputs, but the dollars don't have to be cash — they're 'utility-equivalent' dollars. Being precise here is more important than being exact.

How is this different from a generic expected-value calculator?

EV calcs stop at probability × upside − probability × downside. That misses the asymmetric weight of regret over time. A 30% chance at $100K vs a $20K downside has EV +$22K — go. But if you'd regret skipping at $5K/year for 30 years × 30% probability = $45K of regret cost, the regret-weighted decision value is +$67K — strongly go. Conversely, a high-EV action with zero regret (e.g., a passive investment) doesn't get inflated. The framework respects that some decisions are mostly EV, and some are mostly regret.

Does this work for irreversible decisions (kids, marriage, surgery)?

Set reversibility to 0 and use a long horizon. Irreversible decisions ARE the strongest case for the regret framework — they're where 'skip' regret compounds longest and there's no second attempt. Lower the success-probability bar that you'd accept; raise the annual regret value (life-defining decisions often warrant $20K+/yr regret pricing). The calculator will reflect this with a strong 'attempt for regret' verdict in the asymmetric high-regret + irreversible case.

What if I don't know the probability?

Use the 'honest 50/50 odds' lever in the detail rows — when you genuinely don't know, 50/50 is the calibrated default. Then iterate: as you learn more (research, conversations, base-rate data), update the probability and re-run. Most decisions in the 'should I try' bucket are between 20% and 50%; outside that range you usually know which side it falls on without the calculator.

Should I include opportunity cost in the downside?

Yes — this is the most common error. If attempting means giving up a $80K/yr job for a year of trying, the downside includes ~$80K of foregone wages (plus benefits, equity vesting). Most calc users put 'lost capital' but forget 'lost wages.' Include both. The calculator's effective-downside × reversibility helps — if you can recover most of the wage stream by returning to the job, the reversibility-adjusted downside is small, even if raw downside looks big.

Is regret pricing different at different ages?

Yes. Younger = longer horizon (more years of regret accumulating) but lower per-year regret weight (paths can be re-tried). Older = shorter horizon but higher per-year weight (specific paths close, no second attempt). The calculator handles this via the lookback-horizon × annual-regret product — a 25-year-old with 50 years × $3K/yr regret = $150K, while a 55-year-old with 25 years × $8K/yr = $200K. The age-adjusted regret cost ends up similar; what changes is which probability and reversibility profile clears the bar.

Decision & LifeFree · No signup · 55K+/month

Regret Minimization Calculator — Bezos Framework as Math

Imagine yourself N years from now looking back — what would you regret more, having tried and failed, or not having tried at all? Plug in honest probability of success, upside if it works, downside if it fails, reversibility of failure, lookback horizon, and the dollar-equivalent annual regret of skipping. Calculator returns pure expected value (financial), regret cost of skipping, regret-weighted decision value, and a verdict tier that aligns dollar math with regret math. Low-probability + high-regret actions (career pivots, creative bets, asks-you'll-never-make-again) often clear the regret bar even when EV is mildly negative.

Instant result
Private — nothing saved
Works on any device
AI insight included

Reviewed by CalcBold EditorialLast verified May 4, 2026Methodology

Probability of success

Honest odds the action works out. Calibrate against base rates: startup success at 2-yr mark ≈ 50%, 5-yr ≈ 20%; landing a specific job after applying ≈ 15-25%; book getting traditionally published from cold pitch ≈ 5-10%; career pivot to new field landing target role ≈ 30-50%. If you'd put your money on it being lower, use the lower number.

Upside if it works ($)

Net financial value if the path succeeds — incremental income × years, equity outcome, business sale, etc. For non-financial upside (reputation, autonomy, network), assign a dollar equivalent — what would you pay to have it? Most knowledge-worker pivots: $50-300K total upside over 5 years.

Downside if it fails ($)

Net financial cost if it doesn't work — lost income during attempt, capital deployed, opportunity cost vs current path. The honest version subtracts what you would have earned otherwise. Most reasonable career attempts: $10-50K of downside before reversibility.

Reversibility of failure

% of downside you can recover. 100 = fully reversible (return to old job, redeploy capital). 70 = mostly reversible (some time / network / status loss). 50 = partial. 0 = total bridge-burn (expensive irreversible move). Most career attempts: 60-85% reversible.

Lookback horizon (years)

Years from now until you imagine looking back. Bezos used '80 minus my age' (so ~50 yr at age 30, ~25 yr at age 55). Longer horizons amplify the regret cost — small annual regret × 50 years > large annual regret × 5 years.

Annual regret value ($/yr)

Dollar-equivalent of the annual regret if you skip and it would have worked. Mild regret (passing on a bonus): $1-3K/yr. Career-defining miss (didn't pursue creative work): $5-15K/yr. Life-shaping omission (didn't take the relationship leap): $20K+/yr. Be honest — most people under-price regret by 2-3×.

Embed builderDrop the Regret Min on your site →Free widget · 3 sizes · custom theme · auto-resizes · no signupGet embed code

What This Calculator Does

The Regret Minimization Calculator turns Jeff Bezos’s famous heuristic into actual math. The original framework is qualitative: “imagine yourself at 80 looking back — which choice would you regret more?” Powerful, but fuzzy. This calculator forces honest numbers into the frame: probability of success, upside if it works, downside if it fails, how reversible failure is, the lookback horizon, and a dollar-equivalent annual regret of skipping a path that would have worked. It returns the pure financial expected value, the regret cost of skipping, and a regret-weighted decision value that aligns dollar math with soul math.

It's not a generic EV calculator. Vanilla EV stops at P × upside − (1−P) × downsideand misses the asymmetric weight that regret accumulates over decades. The framework respects the truth most decision tools dodge: some choices are mostly EV (a passive index investment), and some are mostly regret (creative bets, irreversible windows, asks-you’ll-never-make-again). The calculator tells you which kind you’re looking at.

The Math

Two non-obvious moves in this math. First, regret is multiplied by probability — you only regret skipping a path that would have worked. In the timelines where it would have failed anyway, there is no regret to pay. Second, downside is multiplied by reversibility. Most career attempts are 60-85% reversible (you can return to a similar job within 6-12 months), and the math should reflect that. Only true bridge-burns (selling the apartment, leaving the country with no return ticket) approach 0% reversibility.

A Worked Example — Career Pivot

You're considering leaving a stable role to pursue a career switch. Honest numbers:

Probability of success: 30% (calibrated against base rates for landing your target role within 18 months).
Upside if it works: $100,000 — incremental comp + equity + autonomy translated to dollar-equivalents over 5 years.
Downside if it fails:$20,000 — foregone wages + courses + opportunity cost minus what you’d earn pivoting back.
Reversibility: 70% — your network and skill stack let you return to a comparable role within a year.
Lookback horizon:30 years (Bezos’s rule of thumb at age 50: 80 minus your age).
Annual regret value:$5,000/yr — what you’d pay yearly to have NOT carried this “what if” in 30 years.

Plug it in:

effectiveDownside = $20,000 × (1 − 0.70) = $6,000
pureEV = 0.30 × $100,000 − 0.70 × $6,000 = +$25,800
regretIfSkip = 0.30 × $5,000 × 30 = $45,000
netDecisionValue = $25,800 + $45,000 = +$70,800
breakevenP = $6,000 / ($100,000 + $6,000) = 5.7%

The verdict tier comes back ATTEMPT — math AND regret both say go. The breakeven probability is striking: you only need to believe the path has a 5.7% chance for the pure financial math to favour the attempt. Your honest 30% clears that by 24 percentage points. And the regret cost ($45K) is nearly twice the pure EV — telling you regret is the dominant signal in this decision, not finances.

When This Is Useful

The framework earns its keep on decisions where pure EV and regret cost diverge. Generic EV calculators handle the easy cases — an obviously profitable bet or an obviously bad one. The hard cases are where finances mildly lean one way and regret leans the other. Those are also the most consequential decisions of a life: career pivots, creative bets, asks of people who won’t be available later, fertility-window decisions, geography moves, founder leaps, and skipped education. Run the calculator any time the choice has a narrative dimension that won’t age — anything where the future-you might say “I always wondered.”

Reading the Levers

The detail rows surface two synthetic scenarios you should always cross-check. The full reversibility levershows the math if reversibility went to 100% — the cheapest version of the action, where every unit of downside is recoverable. The honest 50/50 levershows the math if you assumed coin-flip odds. Use it when you don’t actually know the probability — 50/50 is the calibrated default for genuine uncertainty, and most decisions in the “should I try” bucket sit between 20% and 50% anyway.

Common Mistakes

Forgetting opportunity cost in the downside. The single most common error. If attempting means giving up an $80K/yr job for a year, the downside includes ~$80K of foregone wages plus benefits plus equity vesting — not just the capital you deploy. Skip this and the calculator overstates the case for attempting.
Under-pricing the annual regret.Most users put $1-3K/yr by default. The honest number for a career-defining miss is $5-15K/yr; for a life-shaping omission (the relationship leap, the creative work never made, the move never tried), it’s $20K+/yr. Future-you rarely thinks the dollar amount was the issue — the regret weights bigger than people predict.
Over-stating success probability. Calibrate against base rates: startup success at 5-yr mark ≈ 20%, landing a specific job after applying ≈ 15-25%, book getting traditionally published from cold pitch ≈ 5-10%. If you’d actually bet money on a lower number, use the lower number.
Confusing reversibility with severity. A $50K equity bet feels severe but may be 80% reversible (you keep the experience, network, and option to redeploy capital). A $5K move with public visibility may be 30% reversible (the reputation hit lingers). Reversibility is about whether you can return to the prior state, not whether the loss feels big.
Treating verdicts as recommendations. The calculator scores the math; it doesn't tell you who you are. An ATTEMPT verdict on a path you’d secretly resent doing is a worse outcome than a DON'T ATTEMPT verdict on a path you’d feel proud of trying. Use the verdict to clarify the trade, then decide.
Running the calc once and stopping. The honest workflow is iterate: run with your gut numbers, see what dominates, then revisit each input with research and conversations. Probability tightens with base rates; downside tightens with budgets; regret tightens by talking to people who skipped similar paths a decade ago and asking how it aged.

Related Calculators

Once regret-min returns ATTEMPT, the Quit-Job Runway Calculator tells you exactly how many months you can fund the attempt before income returns — the two compose into a complete go-decision. The Salary Negotiation Counter Calculator helps price the “ask” class of regret bets — the counter-offer you’d regret not making is exactly the one this framework flags. For purely financial alternatives, Retirement Savings and Can-I-Afford anchor the boring-but-honest alternative against which any regret-weighted attempt is measured.

Frequently Asked Questions

The most common questions we get about this calculator — each answer is kept under 60 words so you can scan.

What is the regret minimization framework?
It's a decision heuristic Jeff Bezos used to leave his Wall Street job to start Amazon — imagine yourself at 80 looking back, and ask which choice you'd regret more. The framework is asymmetric: low-probability + high-regret actions (creative bets, life-shaping asks, irreversible windows like fertility, relationships, or specific job openings) often clear the regret bar even when financial EV is negative. The calculator expresses the framework as math — multiplying honest probability, upside, downside (after reversibility), and a dollar-equivalent regret cost over a lookback horizon.
How do I price 'annual regret value' honestly?
Two anchors. First: ask yourself 'how much would I pay annually to NOT have this regret in 30 years?' If you'd pay $5K/year for 30 years to avoid the lingering 'what if,' that's $5K. Second: scale by the regret category — passing on a small windfall ≈ $500-2K/yr; passing on a meaningful career bet ≈ $5-15K/yr; passing on a family / relationship / creative life-defining call ≈ $20K+/yr. Most people under-price by 2-3× because regret feels distant; the future-self looking back rarely thinks the dollar amount was the issue.
Why multiply regret by probability?
Because you only regret skipping a path that WOULD HAVE worked. If the action was 30% likely to succeed and you skipped, you don't regret in the 70% of timelines where it would have failed anyway — you regret in the 30% where you missed the chance. The honest regret cost is probability × annual-regret × years. Without this weighting, low-probability paths get over-credited for regret cost.
What does reversibility actually mean?
% of the financial downside you can recover. Quitting to start a company with a strong professional network (you can return to a similar job within 6-12 months) is ~80% reversible. Moving abroad and burning the apartment lease + selling everything is ~30% reversible. Investing $50K in equity that vests over 4 years and could go to zero is ~10% reversible (you can't 'unspend' the capital, but you keep the experience). Be honest — most attempts are more reversible than people think.
What does 'attempt for regret' mean as a verdict?
It means pure financial EV is negative (math alone says don't), but the regret cost of skipping over your lookback horizon flips the math positive. This is the genuine Bezos use case — actions where the dollar math is bad but the soul math is decisive. Most career-defining bets land here. The verdict is honest: you're trading expected money for expected non-regret. The calculator surfaces both numbers so you see the trade.
What's the breakeven probability number?
The probability at which pure EV = 0 (regret excluded). Below that, financial math says don't; above, financial math says go. Example: $100K upside, $20K downside, 70% reversible → effective downside $6K → breakeven = $6K / ($100K + $6K) = 5.7%. So you need to believe success is more likely than 5.7% for pure financial math to favour the attempt. Regret-weighting then adjusts from there. The breakeven number is a sanity-check — if it's high (40%+), the action genuinely needs to be likely; if low (<10%), even modest odds clear the financial bar.
What about non-financial upside / downside?
Translate to dollar equivalents — what would you pay to have / avoid them? Reputation hit from a public failure: $5-30K depending on visibility. Network expansion from a year of trying: $10-50K. Health impact from extended stress: $10-100K (use the equivalent of a year of premium healthcare). The framework requires dollar-honest inputs, but the dollars don't have to be cash — they're 'utility-equivalent' dollars. Being precise here is more important than being exact.
How is this different from a generic expected-value calculator?
EV calcs stop at probability × upside − probability × downside. That misses the asymmetric weight of regret over time. A 30% chance at $100K vs a $20K downside has EV +$22K — go. But if you'd regret skipping at $5K/year for 30 years × 30% probability = $45K of regret cost, the regret-weighted decision value is +$67K — strongly go. Conversely, a high-EV action with zero regret (e.g., a passive investment) doesn't get inflated. The framework respects that some decisions are mostly EV, and some are mostly regret.
Does this work for irreversible decisions (kids, marriage, surgery)?
Set reversibility to 0 and use a long horizon. Irreversible decisions ARE the strongest case for the regret framework — they're where 'skip' regret compounds longest and there's no second attempt. Lower the success-probability bar that you'd accept; raise the annual regret value (life-defining decisions often warrant $20K+/yr regret pricing). The calculator will reflect this with a strong 'attempt for regret' verdict in the asymmetric high-regret + irreversible case.
What if I don't know the probability?
Use the 'honest 50/50 odds' lever in the detail rows — when you genuinely don't know, 50/50 is the calibrated default. Then iterate: as you learn more (research, conversations, base-rate data), update the probability and re-run. Most decisions in the 'should I try' bucket are between 20% and 50%; outside that range you usually know which side it falls on without the calculator.
Should I include opportunity cost in the downside?
Yes — this is the most common error. If attempting means giving up a $80K/yr job for a year of trying, the downside includes ~$80K of foregone wages (plus benefits, equity vesting). Most calc users put 'lost capital' but forget 'lost wages.' Include both. The calculator's effective-downside × reversibility helps — if you can recover most of the wage stream by returning to the job, the reversibility-adjusted downside is small, even if raw downside looks big.
Is regret pricing different at different ages?
Yes. Younger = longer horizon (more years of regret accumulating) but lower per-year regret weight (paths can be re-tried). Older = shorter horizon but higher per-year weight (specific paths close, no second attempt). The calculator handles this via the lookback-horizon × annual-regret product — a 25-year-old with 50 years × $3K/yr regret = $150K, while a 55-year-old with 25 years × $8K/yr = $200K. The age-adjusted regret cost ends up similar; what changes is which probability and reversibility profile clears the bar.