It’s a Tuesday night, around 9:30, and you have a spreadsheet open. Five rows. Credit card with $940 at 22.4%. Old store card with $310 at 26.9%. Car loan with $11,200 at 6.1%. Student loan with $18,000 at 5.8%. A medical bill from last winter, $612, currently at zero interest but headed to collections if you ignore it for two more months.
You stare at this for a while. You move the rows around. You sort by interest rate. You sort by balance. You open a calculator, type in some numbers, close the calculator. You consider, briefly, just paying the minimum on everything for another month and dealing with this when life is less chaotic. You consider, also briefly, putting everything you have at the highest-interest one and starving for three months.
Neither of those is going to happen. The actual outcome of this Tuesday night, if nothing changes, is that you close the laptop and the spreadsheet stays exactly where it is for six more weeks.
This is the real problem with debt payoff advice. It’s not that there’s no good information. It’s that the available information assumes you’re going to follow the optimal plan, and the question of which plan you’ll actually finish never gets asked.
The two methods, defined cleanly
There are two well-known approaches to paying off multiple debts. They sound similar and produce very different experiences.
The avalanche method: list every debt by interest rate, highest to lowest. Pay the minimum on everything. Throw every extra dollar at the highest-rate debt until it’s gone. Then roll that payment into the next-highest-rate debt. Continue until done.
The snowball method: list every debt by balance, smallest to largest. Pay the minimum on everything. Throw every extra dollar at the smallest balance until it’s gone. Then roll that payment into the next-smallest. Continue until done.
The avalanche minimizes total interest paid. Mathematically, it is the correct answer. If you ran a simulation with five debts and a fixed extra payment amount and infinite discipline, the avalanche would always pay less interest and reach zero faster than the snowball. By a meaningful amount, often hundreds to a few thousand dollars depending on the situation.
So if you’re a calculator, you pick the avalanche, and we’re done.
Except the calculator is not the audience for either of these methods. The audience is a human being who has been paying minimums on five debts for two years and would like to stop. And for that audience, the math is not the binding constraint. Follow-through is.
Why the snowball wins in practice
In 2011, two researchers at Northwestern’s Kellogg School — David Gal and Blakeley McShane — published a paper called “Bad Habits, the Endowment Effect, and the Snowball Method” in the Journal of Marketing Research. They looked at actual debt payoff data, not simulations, and asked which approach was associated with people successfully eliminating their debts.
The result was counterintuitive: closing accounts in order of smallest balance — the snowball — was a stronger predictor of payoff success than any other variable they measured, including interest rates and total balance. People who started with the smallest balance, regardless of its interest rate, were more likely to stick with the program and reach zero.
The mechanism is motivational, not financial. When you eliminate a debt — any debt, even a small one — you get a clean, unambiguous win. An account that used to exist no longer does. That feedback is concrete. It happens on a real day. You can feel it.
When you put your extra dollars against your highest-rate debt under the avalanche, you also make progress, but the progress is invisible for months. The balance ticks down. There are no closed accounts. No lines disappear from your statement. The math is doing its job in the background, but your motivation system gets very little to work with. And your motivation system is doing the actual work of getting you to make that extra payment every month for the next three years.
This connects to broader behavioral research on small wins as motivators. Teresa Amabile and Steven Kramer at Harvard, in their work on the “progress principle”, found that small, visible progress is one of the strongest predictors of sustained motivation on long-running goals. The snowball isn’t a worse plan financially because it’s emotionally satisfying — it’s a better plan in real life because the emotional satisfaction is what keeps you running it.
The avalanche assumes you have a discipline budget that will last 36 months without external feedback. The snowball doesn’t make that assumption. It pays you, in dopamine, for showing up.
How to actually pick tonight
Here’s the heuristic that handles most cases.
If you’ve never paid off any debt before, or if you’ve started a payoff plan in the past and abandoned it within six months, use the snowball. List your debts smallest to largest. Pay minimums on everything except the smallest. Throw every extra dollar at the smallest until it’s gone. Repeat with the next-smallest. The math cost of choosing snowball over avalanche is real but bounded — usually a few hundred to a couple thousand dollars over the life of the plan. The math cost of starting the avalanche and quitting in month four is all of it, plus the interest you keep paying for years afterward on debts you stopped attacking.
If you’ve successfully paid off debts before — if you have evidence that you can run a multi-year financial plan without losing momentum — use the avalanche. List by interest rate, highest first, and put your extra dollars there. You’ve already proven you don’t need the small wins for fuel. Take the math.
A hybrid that works for some people: pay off any one small debt under $500 first, regardless of interest rate, just to get a closed account on the board. Then switch to avalanche for the rest. You get one early win to establish momentum, then optimize from there.
Whichever method you pick, two things matter more than the method itself. The first is automating the extra payment. Decide on a fixed extra amount — $50, $200, whatever you can sustain — and set it to autopay on the same day every month against the target debt. If the extra payment is something you have to remember to make, you’ll forget to make it about a third of the time, and a third of the time means the plan is two years longer than it needed to be. Pay yourself first by automating the transfer is the same logic applied to savings, and it works for the same reason: removing the decision from the loop.
The second is not adding new debt while you’re paying down the old. This sounds obvious and is brutally hard, especially with credit cards that stay open with available balances. The most common failure mode for either method is paying down the card by $400, then putting $300 on it the same month for an unrelated purchase. The balance moves a little, but the plan is functionally not running. If impulse purchases are part of how the debt got built up, the 24-hour rule is the closest thing to a structural fix, and it’s worth setting up at the same time you start your snowball or avalanche.
The real comparison nobody runs
The honest framing of debt snowball vs avalanche is not “which is mathematically optimal.” It’s “which one will I still be running in month eighteen.”
The avalanche, executed perfectly, beats the snowball by some amount of interest. The snowball, executed imperfectly, beats the avalanche abandoned in month four by all of the interest you would otherwise still be paying. These are not equivalent comparisons, and the second one is much closer to the actual decision you’re making.
Pick the one you’ll finish. Set it up so it runs without you remembering. Close the laptop. The spreadsheet is done with you for tonight.