Optimize: Account-Type Withdrawal Allocation

The order you draw from your accounts — Traditional, Roth, taxable — quietly shapes your lifetime taxes and how long your money lasts. Drawing taxable first preserves tax-advantaged growth; drawing Traditional early can blunt later RMDs. The Optimizer can search these mixes and rank them.

First read the Optimizer overview. Background: Taxes and the Strategy tab.

Withdrawal allocation vs. asset mix

This optimizes your withdrawal allocation — which account types you draw from, and in what proportion or order. It does not optimize your asset allocation (your stock/bond/cash mix). In RetireRange, an account’s asset mix is set by how you’ve built your accounts and stays fixed through the simulation — there’s no automatic equity glide path. So this search is about sequencing your withdrawals, not shifting your investments.

Set it up

1. On the Optimizer, add the allocation percentage dimension(s) — for example, sweep how much of each withdrawal comes from Traditional vs. Roth vs. taxable. (Ordered drain priorities can also be explored.)
2. Choose your objective — after-tax legacy and success rate are the most useful here; lifetime taxes only with the usual success-rate anchor.

What to expect — this is one of the bigger searches

Allocation lives across several percentages at once, so the search space can be large — this is one of the searches most likely to run long, even into the hours for a fine grid. Manage it:

• Use Latin Hypercube Sampling rather than a full grid. A full grid tests every combination, so adding allocation dimensions multiplies the work explosively. Latin Hypercube instead scatters a fixed number of samples so that every dimension is evenly covered — giving you a representative read of the whole space for a fraction of the runs.
• Go coarse first, then refine around the best region.
• Set a budget cap, and consider a lower simulation count while exploring. The cap is a hard ceiling on how many candidates the search evaluates, so a run that could otherwise stretch into the hours stops at a size you choose — the sampler simply spends that budget on the best coverage it can, rather than running until the grid is exhausted.

Reading the results

• Look for a broad good region, not a single knife-edge winner — robust sequencing beats a fragile "optimal" point.
• Confirm finalists on the Compare tab, since allocation differences between candidates are often small and can be within the noise between optimizer batches.

RetireRange is for educational and planning purposes only and is not financial, tax, or investment advice. Optimizer rankings are modeled estimates based on your assumptions — not predictions or recommendations. Consult a qualified financial planner for advice specific to your situation. See Limitations.