When you're getting a mortgage, someone will eventually ask if you'd like to "buy points". It sounds like an upgrade. It's actually a trade: you hand over cash at closing, and in exchange the lender knocks a bit off your interest rate for the life of the loan.
I think points are one of the most misunderstood parts of getting a mortgage, mostly because the decision gets framed as "do you want a lower rate?" Of course you want a lower rate. The real question is whether you want it at this price.
What a point actually costs
One point costs 1% of the loan amount, and it typically cuts the rate by about 0.25%. That's the whole mechanism. There's no magic in it.
So on a $320,000 loan, one point is $3,200, paid upfront at closing. In return, a rate that would have been, say, 6.75% becomes roughly 6.5%. You can buy more than one point, and sometimes fractions of a point, but the maths works the same way at any size.
Notice what you're doing here. You're prepaying interest. That's all a point is. You give the lender a lump of interest today so they charge you slightly less of it every month going forward.
The only number that matters: break-even
Because a point is prepaid interest, the question isn't "is a lower rate good?" It's "how long until the monthly savings pay back the upfront cost?" That's the break-even point, and it's simple division.
Take the $320,000 loan again. One point costs $3,200. Suppose the 0.25% rate reduction saves you roughly $50 a month on the payment. Then $3,200 divided by $50 is 64, so the break-even is 64 months. That's just over five years.
Before month 64, you're behind. You paid $3,200 and you haven't got it back yet. After month 64, every month you keep the loan is pure savings. The point earns its keep from then on, month after month, for as long as the loan survives.
Your own numbers will differ, so run them properly. Put your actual loan amount and both rates into the mortgage calculator and compare the two monthly payments. The difference between them is your real monthly saving, and the point cost divided by that saving is your real break-even.
Here's the catch nobody mentions
The break-even calculation assumes you keep the loan. Not the house. The loan.
Two things kill a loan early: selling the house, and refinancing. Both happen a lot more often than people expect when they're sitting at the closing table feeling settled. If you sell in year three, your 64-month break-even never arrives and the $3,200 was simply a gift to the lender. If rates drop and you refinance in year two, same result. The new loan starts from scratch, and any points you paid on the old one are gone.
This is why I'd think hard before buying points if there's any real chance you'll move within five or six years, or if rates are high enough that a future refinance looks likely. Paying points on a loan you half expect to replace is paying for a discount you won't be around to collect. If a refinance is already on your mind, run the numbers in the refinance calculator first and see how that scenario compares before you commit cash to points.
When points do make sense
Points aren't a scam. They're a bet on your own future, and sometimes it's a good bet. If you're buying a house you genuinely plan to stay in for a decade, at a rate you don't expect to beat by refinancing, then paying a point and clearing break-even at month 64 leaves you years of savings on the far side. Over a full 30-year loan, that adds up to real money.
They can also make sense if you have spare cash at closing that isn't needed for the deposit or an emergency fund, and you'd rather lock in a guaranteed monthly saving than invest it. A known saving on a loan you'll definitely keep is a decent, boring return.
My short version
Work out the break-even month. Then be brutally honest about whether you'll still have this exact loan when that month arrives. Keep the loan past break-even and points are worth it. Sell or refinance before it and they're not. Everything else is noise.