Avalanche or snowball? Run both.

What the simulation actually does

Every month, in order: interest is added to each balance at that debt's APR divided by twelve. Then the plan budget — the sum of all your minimums plus your extra — is spent. Each debt receives its minimum. Whatever remains goes entirely to one target debt. When a debt reaches zero, its minimum doesn't leave the plan; it joins the pot attacking the next target. That rolling is the engine, and it's why the last debt disappears so much faster than the first.

The only difference between the two columns is which debt is the target. Avalanche sorts by APR, highest first. Snowball sorts by balance, smallest first. Everything else is identical, including the total you spend each month, which is what makes the comparison honest.

Worked example, using the numbers loaded on the page

Four debts: a $5,200 card at 22.9%, a $900 medical plan at 0%, a $2,400 store card at 28.9%, and an $11,500 car loan at 6.9%. Minimums total $615. Add $200 extra and the plan budget is $815 a month.

Avalanche looks at the rates and picks the store card at 28.9%, because every $100 sitting on that card costs $2.41 a month while $100 on the medical plan costs nothing. Snowball looks at the balances and picks the $900 medical plan, which it clears in about four months — a real account closed, one fewer statement, one fewer due date.

Snowball spent four months attacking a debt charging 0% interest, which in pure arithmetic is the worst possible target. That's the cost of the momentum, and the page shows you exactly what it comes to in dollars.

Which one to pick

Avalanche wins on math. It always has and always will — mathematically it is impossible for snowball to cost less, because avalanche by definition removes the most expensive interest first. If you're comparing the two on total cost, the argument is already over.

Snowball often wins on follow-through, and follow-through is the variable that actually determines the outcome. A plan you abandon in month seven costs you every dollar of the debt you didn't pay. If the difference here is a couple of hundred dollars and snowball gets you a closed account in the first quarter, take snowball and don't feel clever about it.

A middle path most calculators won't tell you about: run snowball until you clear one small debt, then switch to avalanche. You get the early win and most of the savings. Nothing requires you to pick a religion in month one.

What this doesn't model

Minimum payments here are fixed. Real credit card minimums are usually a percentage of the balance with a floor, so they shrink as the balance falls. Holding them constant makes the simulation slightly optimistic on debts you aren't targeting — a difference of a few months on a long plan.

It assumes you add nothing new to any of these balances. If you keep using the card while paying it down, none of these numbers apply. Freeze the cards, or at minimum take them out of your phone's saved payment methods.

It ignores promotional rates that expire, deferred-interest offers that back-charge you retroactively, late fees, over-limit fees, and penalty APRs. Any of those can add hundreds. It also ignores the credit score effect of the order you pay things in — paying a maxed card down below 30% of its limit can move your score faster than closing a small loan, which matters if you're about to apply for a mortgage.

And it treats the car loan like any other debt. Secured debt is not like any other debt: miss enough payments on that one and the car is repossessed, regardless of how well the rest of the plan is going. Never divert a secured minimum into an accelerated payoff on a card.

Before you start either plan

Call every card issuer and ask for a lower APR. It's a five-minute call, the script is "I've been a customer for X years and I'm looking at a balance transfer offer — can you review my rate?", and it works often enough to be worth the twenty minutes. A rate cut is free progress that no payoff strategy can match.

Then set the plan budget as an automatic payment split across your accounts on the day after payday. The single biggest predictor of finishing is whether the money moves before you see it.