← Back to Blog

I Tested My Retirement Plan Against 5 Different Assumptions. Here's What I Found.

I've been planning for early retirement for years. Like most people, I plugged my numbers into calculators that assumed around 7-8% returns, saw a comfortable success rate, and felt good about my plan.

Then I started digging into what the professionals actually assume.

BlackRock expects 6.5% for US stocks. Vanguard says 4.5%. GMO—which correctly predicted the 2000s lost decade—says 0.5%.

Wait, what?

I decided to run my exact retirement plan through each of these assumption sets. Same savings, same spending, same timeline. Just different return assumptions.

The results were... illuminating.

My Setup

Here's what I was testing:

Pretty standard aggressive FIRE plan. Twelve more years of saving, then 40 years of retirement.

I ran this through a Monte Carlo simulator using five different assumption sets:

  1. Historical averages (1926-2024): 10.2% US stocks, 5.2% bonds
  2. JPMorgan 2024 LTCMA: 6.8% US stocks, 5.0% bonds
  3. BlackRock 2024 LTCMA: 6.5% US stocks, 4.8% bonds
  4. Vanguard 2024: 4.5% US stocks, 4.3% bonds
  5. GMO Q4 2024: 0.5% US stocks, 3.8% bonds

The Results

Assumption SetSuccess RateMedian End Balance
Historical91%$3.2 million
JPMorgan76%$1.4 million
BlackRock72%$1.1 million
Vanguard58%$420,000
GMO34%Ran out at 82

That's a 57 percentage point spread between the most optimistic and most pessimistic assumptions.

Under historical assumptions, I'm basically set. 91% success rate, likely dying with millions.

Under GMO's assumptions, I have a coin flip's chance of running out of money by my early 80s.

Same exact plan.

What This Actually Means

The Optimistic Case (Historical: 91%)

If markets return what they've returned historically (10%+ for stocks), my plan is rock solid. This is what most online calculators would show me.

But it assumes valuations don't matter (they do), the future will look like the past (uncertain), and the US will continue dominating global markets (possible but not guaranteed).

The Moderate Case (BlackRock/JPMorgan: 72-76%)

These are the assumptions that pension funds and endowments actually use. A 72-76% success rate isn't bad, but it's not comfortable either.

It means roughly 1-in-4 scenarios lead to trouble.

The Conservative Case (Vanguard: 58%)

Vanguard is known for conservative projections. At 58% success, I'm barely better than a coin flip.

The Pessimistic Case (GMO: 34%)

GMO has been bearish on US stocks for years. At 34% success, my plan is in serious trouble under their assumptions.

What I Learned

1. My Plan Is Assumption-Dependent

The 57-point spread tells me something important: my retirement success depends heavily on which future materializes.

If you showed me only the historical result (91%), I'd feel great. But that's cherry-picking the most optimistic assumption.

2. The "Right" Assumption Doesn't Exist

Nobody knows if BlackRock or GMO will be closer to reality. We won't know for 20 years.

What I can do is understand my sensitivity to assumptions and plan accordingly.

3. Flexibility Is Worth More Than Precision

Looking at these numbers, the best thing I can do isn't necessarily saving more. It's building in flexibility:

This flexibility doesn't show up in the numbers, but it effectively converts some "failure" scenarios into "adjustment" scenarios.

What I'm Changing

1. Extending My Timeline by 1-2 Years

Instead of hard-targeting retirement at 50, I'm thinking 50-52 depending on market conditions.

This flexibility alone moves my Vanguard success rate from 58% to 71%.

2. Building a Bigger Cash Buffer

I'm increasing my target emergency fund from 1 year to 2 years of expenses. This protects against sequence of returns risk.

3. Planning for Variable Spending

This guardrails approach dramatically improves success rates across all assumptions.

The Takeaway

If your retirement plan only shows you one number based on one assumption, you're missing the most important insight: how sensitive is your plan to being wrong?

A 91% success rate using historical assumptions might be 58% using Vanguard's assumptions and 34% using GMO's.

That spread matters more than any single number.

I'd rather have an 80% success rate that's stable across assumptions than a 95% rate that's fragile.

The professionals—pension funds, endowments, financial advisors—don't use historical averages. They use forward-looking estimates. And they test multiple scenarios.

You should too.


Test Your Plan Against Real Institutional Assumptions

Run your retirement plan through the same assumptions BlackRock, JPMorgan, Vanguard, and GMO use. See your success rate under each scenario. Understand your sensitivity.

Try QuantCalc Free