The Roth conversion ladder has the strongest brand of any FIRE tactic. Subreddits love it. Bloggers love it. The math, when described in five sentences, is irresistible: convert traditional retirement money to Roth in low-income years, wait five years, withdraw the converted principal tax-free and penalty-free at any age. Compound for decades inside the Roth. Pay zero federal income tax in retirement.

The math, when actually run, is more conditional than the five-sentence version admits. I ran a 30-year retirement scenario through Thermal’s engine against all eight withdrawal strategies, with and without a Roth ladder layered on top. The ladder helped in some cases and hurt in others. Below is where the line is.

The mechanics, including the part most explanations skip

A Roth conversion takes pre-tax money (traditional 401(k), IRA, 457, 403(b)) and moves it to a Roth IRA. The converted amount is taxed as ordinary income in the year of conversion. From that point forward, the money grows tax-free. Withdrawals follow Roth rules.

The “ladder” is the strategy of running this play repeatedly across multiple early-retirement years, each year starting a new five-year clock on the converted principal. By year six of retirement, the conversion you did in year one is accessible without the 10% early withdrawal penalty.

The part most explanations gloss over: there are two five-year rules.

  1. Roth IRA five-year rule. Your first Roth IRA must have been open for five years before earnings can be withdrawn tax-free. Once any Roth IRA crosses the five-year mark, this clock is satisfied for all your Roth IRAs forever.
  2. Per-conversion five-year rule. Each conversion has its own five-year clock that controls only the 10% early withdrawal penalty for the converted principal. Convert in 2026, the principal is penalty-free in 2031.

These rules don’t affect tax. They affect penalty. After 59½, both rules become irrelevant. Before 59½, both apply.

When the ladder actually wins

You have a multi-year low-income gap.

The textbook case. You stop working at 50, have no pension, no Social Security yet, and are living off a taxable brokerage. Your ordinary income for the year is whatever interest and dividends the taxable account throws off. That puts you in the 10% or 12% bracket. Convert enough traditional to fill the 12% bracket every year, and you’re moving money from a future 22-32% withdrawal rate to a paid-up-front 12% rate. Real win.

You have a large traditional balance heading into RMD years.

Required Minimum Distributions start at 73 or 75 under SECURE 2.0, depending on your birth year. If your traditional balance is large enough that the RMD will push you into a higher bracket than you otherwise live in, conversions before RMD age are pulling future income forward into lower brackets. The bigger the traditional balance, the more this matters.

You’re moving from a high-tax to a no-tax state.

This one is asymmetric in your favor. Convert after the move (do conversions while you’re a Texas or Florida resident), and the conversion is taxed at 0% state. Withdraw the Roth later regardless of state, also at 0%. If you converted while still in California, you’d pay 9-13% state tax on the conversion that you’d permanently avoid by waiting one more year for the move.

You expect tax rates to rise.

This is the bet most ladder advocates make implicitly. Federal rates today are historically low. The TCJA brackets are scheduled to sunset, and the long-run fiscal picture suggests rates climb, not fall. If your tax rate at conversion ends up lower than your tax rate at withdrawal would have been, you win. If it goes the other way, you don’t. This is the part nobody can prove until they’re dead.

When the ladder loses (or just doesn’t help)

Your tax rate today is already as low as it’s going to get.

If you’re semi-retired, drawing from taxable, and already in the 0-12% bracket every year for the next 30 years, the conversion just adds tax for no benefit. Your traditional money was going to come out at 12% anyway.

You’re close to a Medicare IRMAA cliff.

Income-Related Monthly Adjustment Amount adds to your Medicare Part B and Part D premiums based on Modified Adjusted Gross Income from two years prior. Crossing a threshold by a single dollar can cost you thousands of dollars per year in higher premiums. A Roth conversion that nudges you across an IRMAA bracket can wipe out the entire tax-arbitrage benefit of the conversion. Conversions need to be sized to land safely below the next IRMAA cutoff.

Your Roth balance is already large.

If you’re a long-term Roth contributor and your tax-free bucket is already 40-50% of your portfolio, conversions are moving money from a slightly-better-tax-treated bucket to a marginally-better-tax-treated bucket. Diminishing returns. The cost (tax paid now) doesn’t justify the benefit (tax saved later).

You don’t have taxable cash to pay the tax bill.

The conversion creates an income tax liability the year you do it. If you have to use traditional money to pay that tax, the math gets significantly worse, because that withdrawal also triggers ordinary income tax (and potentially the 10% penalty if you’re under 59½). The strategy works best when you have outside cash to fund the tax bill cleanly.

What the engine showed

I ran a hypothetical: $2M portfolio split 45% traditional, 35% Roth, 20% taxable. Retiring at 50. $80k annual spending, inflation-adjusted. Eight withdrawal strategies (4% rule, guardrails, dynamic percentage, custom, VPW, floor-and-ceiling, endowment, BBD). Four withdrawal orderings (tax-efficient, traditional-first, Roth-first, pro-rata).

Layering a 10-year Roth conversion ladder on top of each strategy:

  • Tax-efficient ordering with Guyton-Klinger: ladder added roughly 2.5 years of additional portfolio life (across the median scenario), and lifted ending balance by ~12%.
  • Traditional-first ordering: ladder hurt. Pulling traditional out fast was already accomplishing what the ladder was trying to do.
  • Roth-first ordering: ladder helped marginally, mostly by ensuring there was enough Roth principal to draw from.
  • BBD strategy: ladder essentially irrelevant, since the goal is to never sell the assets at all.

Translation: the Roth ladder is a multiplier on a tax-aware strategy, not a strategy unto itself. If you’re not also thinking carefully about which bucket you draw from when, the conversion math is just shuffling money around.

The Roth ladder isn’t a strategy. It’s an amplifier on whatever withdrawal strategy you’re already using.

What “optimal” usually looks like

The optimal Roth conversion is rarely “convert everything you can.” It’s usually “convert exactly enough to fill a specific tax bracket without crossing the next one.”

For most early retirees that means: figure out what bracket you want to be in (often 12% or 22%), subtract your other ordinary income for the year, and convert the difference. Do it every year of the gap window. Stop when the math no longer favors it, usually when Social Security or pension income kicks in and your baseline income is already in the bracket you wanted to fill.

That requires running the actual math on your actual portfolio, not a generic calculator. The right conversion amount depends on your bucket balances, your other income sources, your state, and the specific brackets you’re trying to land in.

Where Thermal fits

Thermal models conversions year-by-year against your real bucket balances and runs the result against all eight withdrawal strategies in Monte Carlo. You can see the success-rate delta of converting versus not, on your specific plan, in your specific state, with your specific IRMAA cliffs. It’s the same engine running both scenarios, which means the comparison is honest.

For high-earner-specific scenarios where the tax math gets weirder (Section 199A interactions, NIIT thresholds, AMT aftershocks for ISOs in the same year as a conversion), High Earner Playbook has the longer-form work.