← Back to blog
A transparent glass balance scale on a dark floor, with a stack of glowing green coins on one pan and a stream of small green beads filling the other, as a thin green line crosses behind the level beam.
Fintech·November 22, 2026·10 min read

Mortgage Discount Points: The Break-Even Math Most Calculators Get Wrong

The standard "points cost divided by monthly savings" break-even is off in both directions. Here is the full math on a $400,000 loan, including faster principal paydown, opportunity cost and the refinance risk that decides most cases.

The usual advice on mortgage discount points is to divide the cost of the points by the monthly payment savings and see whether you will keep the loan that long. That formula is wrong in both directions at once: it ignores the faster principal paydown a lower rate buys you, which makes points look worse than they are, and it ignores what the cash could have done instead, which makes them look better. Once you correct for both, the answer to "are points worth it" turns almost entirely on one number the formula treats as an afterthought: how many months the loan actually survives before you sell or refinance.

What a discount point is

A discount point is prepaid interest. You pay the lender 1% of the loan amount at closing, and in exchange the lender lowers your interest rate for the life of the loan. How much the rate drops per point is not fixed by any rule. It comes from the lender's rate sheet on the day you lock, and it varies by lender, loan type and market conditions. A quarter of a percentage point per point is a common rule of thumb, but you will see offers that are noticeably better or worse than that.

On a Loan Estimate, points appear in the origination charges as a percentage of the loan amount. The mirror image also exists: lender credits, where you accept a higher rate and the lender pays part of your closing costs. The CFPB's explainer on discount points and lender credits covers the disclosure side. This post is about the math.

The formula everyone uses

The standard break-even calculation is:

break-even months = cost of points / monthly payment savings

It is the mortgage version of a payback period, and it shares the weaknesses covered in ROI vs. IRR vs. payback period: it ignores the time value of money and ignores anything that happens after the cutoff. Here it has one more blind spot, specific to amortizing loans.

A worked example

Take a $400,000, 30-year fixed loan:

The monthly savings are $66.12. Divide $4,000 by $66.12 and you get 60.5, so the simple answer is that you break even in month 61, a little over five years in.

That number is not terrible, but it is not right either.

What the formula misses: principal paydown

A monthly mortgage payment is split between interest and principal. At a lower rate, less of each payment goes to interest, so more goes to principal, even though the payment itself is smaller. As walked through in how amortization actually works, that split is where most of the real cost of a loan hides.

In the example, after 60 months:

No points (6.75%)One point (6.50%)
Monthly payment$2,594.39$2,528.27
Payments made, 60 months$155,663$151,696
Remaining balance at month 60$375,503$374,444

The one-point borrower paid $3,967 less in payments and owes $1,059 less. That $1,059 is not a paper gain. When you sell or refinance, the lower balance is what you pay off, so it comes back to you as cash at that moment.

Count both and the break-even moves from month 61 to month 48, a full year earlier. The simple formula undercounts the benefit of points because it only looks at the payment, not the balance.

What the formula also misses: the cash had other jobs

The $4,000 spent on points is $4,000 you did not keep in savings, invest, or put toward the down payment. Two ways to account for that:

Opportunity cost at a savings rate

If the $4,000 would otherwise sit in an account earning 4%, compound that forgone growth against the points (and credit the monthly savings with the same 4%, since you could invest those too). The break-even moves back from month 48 to about month 54. The mechanics are the same compounding covered in compound vs. simple interest: small rates on small sums, but over years it shifts the answer by months.

Prepaying the same cash instead

A sharper comparison: instead of buying a point, put the $4,000 toward principal at closing. You borrow $396,000 at 6.75%, the payment drops to $2,568.45 (a $25.94 saving), and you are ahead from day one, because that $4,000 is still yours as home equity. There is nothing to "break even" on.

Points win this comparison only once the lower rate has run long enough. In this example, buying the point overtakes prepaying the same cash at month 65. Before that, prepaying leaves you with more money; after it, the point does.

-$4k-$2k$0k$2k$4k$6k024487296120Months since closing1 point, full math (break-even month 48)1 point, payment-only (month 61)Same $4,000 prepaid instead

The chart shows the three views of the same $4,000. The dashed line is what the simple formula sees. The solid green line is the full picture including principal. The amber line is the do-nothing-clever alternative, and it is ahead of the point for the first five and a half years.

Pricing matters more than the formula

The rate reduction you get per point dominates every other input. Here is the same $400,000 loan under three offers:

OfferCostNew rateMonthly savingPayment-only break-evenFull break-evenWith 4% opportunity costInterest saved over 30 years
1 point for 0.25%$4,0006.50%$66.12month 61month 48month 54$23,803
2 points for 0.50%$8,0006.25%$131.52month 61month 48month 54$47,348
1 point for 0.125%$4,0006.625%$33.15month 121month 96month 122$11,933

Two things stand out. First, buying two points at the same price per quarter-point does not change the break-even at all; it doubles both the bet and the payoff. Second, a weak offer (half the rate cut for the same price) doesn't just double the break-even, it pushes it out to eight to ten years, which is longer than many loans actually last.

The 30-year interest-saved column is the number lenders and some calculators like to highlight. It assumes you keep the loan to the last payment, which is the least likely outcome.

The variable that actually decides it: how long the loan lives

Every break-even above is a statement of the form "points pay off if the loan survives at least N months." The loan ends when you sell, refinance or pay it off, and the points you paid end with it. There is no refund of unused prepaid interest.

This creates an uncomfortable asymmetry:

Points are therefore a bet that rates will not fall enough to make you refinance, layered on top of a bet that you will not move. When you are buying points to lower a rate that is high by recent standards, you are betting against exactly the scenario that would make you want to refinance. The break-even math doesn't capture this risk, because it treats the holding period as a number you know in advance.

A practical way to use the math: estimate the shortest realistic life of the loan (a job move, a growing family, a rate drop that triggers a refinance), not the most hopeful one, and compare it with the full break-even. If the full break-even is comfortably inside that shortest case, points are reasonable. If it is close, the expected value is thin and the downside is concentrated in the scenario where you lose the whole fee.

Lender credits: the same trade in reverse

Lender credits (sometimes described as negative points) flip the logic. You take a higher rate, and the lender pays some of your closing costs. You come out ahead if the loan ends before the break-even, which is exactly the case where points lose. Borrowers who expect to refinance or sell soon are often better served by taking credits than by paying points, and the same full-break-even method applies, just with the sign reversed: count the extra principal you will owe at the higher rate, not only the higher payment.

Taxes and APR, briefly

Two details complicate comparisons in practice:

Takeaway

Compute the break-even the full way: count the lower balance, not just the lower payment, and charge the cash a realistic opportunity cost. Then compare it against the shortest plausible life of the loan, remembering that a fall in rates is both the most likely reason to refinance early and the exact scenario where points are lost. In the example here, a quarter-point-per-point offer needs about four to five years to pay off and over five years to beat simply prepaying the same cash. If you want to run your own numbers, the mortgage points mode of the loan calculator amortizes both scenarios side by side, which is what the payment-only formula skips.

#mortgage-points#discount-points#break-even#mortgage-math#refinancing#fintech
ShareXLinkedInHacker News

Try it

Free, in your browser, no sign-up.

New posts by email

One email when a new post goes up. No other mail; unsubscribe in one click.

We store your email only to send new posts. See the privacy policy.

Related reading

Fintech
How Amortization Actually Works: Why Two Loans at the Same Rate Cost Different Amounts
Fintech
Debt Avalanche vs. Snowball: The Real Math, and When the Gap Actually Matters
Fintech
Compound vs. Simple Interest: Why the Same Rate Can Mean Different Money