“Current mortgage rates” is one of the few finance terms holding ~20,000 searches per 24 hours this week — and the anxiety is justified: Freddie Mac’s 30-year fixed average printed at 6.76% in its latest survey, Bankrate’s weekly lender panel shows 6.78% and rising, and the Federal Reserve concludes its September 15-16 meeting today, with markets priced for a hold — or even a surprise hike — rather than the cuts borrowers have been waiting two years for.
That reversal is the story of this refi cycle: the “lock in before rates drop” plan has become “wait, they’re going the wrong way.” This guide gives you today’s real numbers, why the Fed’s decision may barely move your mortgage rate, and the honest decision framework for whether to refinance now, later, or never.
Today’s Mortgage Rate Snapshot (September 16, 2026)
- 30-year fixed (conforming): 6.76% — Freddie Mac PMMS; 6.71-6.78% across Bankrate and Credit Karma panels
- 15-year fixed: 6.09% (Freddie Mac) — the cheapest mainstream product in years relative to the 30
- FHA 30-year: ~6.8% per daily indexes; VA 30-year: ~6.8%
- Jumbo 30-year: ~7.1-7.4% depending on lender and lock period
- Anchor: the 10-year Treasury, which touched 5.00% this week per Treasury.gov — its highest sustained zone since 2007
If you are sitting on a sub-5% mortgage from 2020-2021, you are in the famous “rate lock-in” cohort — and if you refinanced into an adjustable during the scramble, your reset math is different again (our refinance guide covers the 5/1 and 7/1 reset waves).
Why Today’s Fed Decision May Barely Move Your Mortgage Quote
The counterintuitive truth of 2026: the Fed does not set mortgage rates — the bond market does. The 30-year tracks the 10-year Treasury plus a spread, and that spread is currently driven by three things the FOMC cannot control:
- Inflation expectations. With CPI running hot enough that prediction markets price a real chance of a hike today, long bonds demand higher yields. A hawkish hold raises mortgage rates even though “the Fed did nothing.”
- Treasury supply. Record deficit financing floods the market with bonds, pushing yields up and competing capital away from mortgages.
- Mortgage-bond demand. Agency MBS flows fluctuate with bank balance sheets and global capital; wide spreads mean your 6.76% despite a 3.6% fed funds rate.
The practical translation: a 25bp Fed move in either direction typically shifts mortgage quotes by less than that, sometimes in the “wrong” direction. Rate volatility — quotes moving 0.25% in days — is your actual enemy, which is why locks and float-downs matter more than perfect timing this month.
The Refi Decision Framework: Run Your Own Numbers
Step 1: The break-even math
Total closing costs ÷ monthly savings = months to break even. Example: $4,000 costs, $310/month saved on a $300,000 loan going from 7.5% to 6.7% = 13 months. Stay five years? Excellent deal. Moving next spring? Skip it.
Step 2: The 15-year shortcut many borrowers ignore
At 6.09%, a 15-year fixed is the best relative value it has been in a decade. If you can carry the payment, refinancing a 30 into a 15 near your remaining term can cut total interest by tens of thousands without waiting for a Fed cut that may not come.
Step 3: The no-cost trap
“No-cost refis” swap closing costs for a rate ~0.25-0.50% higher. In a falling-rate world you can later cheapen the loan out; in a flat-or-rising world (today’s base case) you are paying the premium indefinitely. Compare a true no-cost quote against the paid version over your planned stay — often the paid version wins now.
Step 4: Rate-and-term vs cash-out — different animals
Cash-out refis price 0.25-0.75% above rate-and-term, and most 2026 guidelines cap LTV at 80%. Pulling equity at ~7%+ to fund consumption is the single most expensive household loan decision; pulling to consolidate 24% card debt usually still pencils out. The difference is what the money does after it lands.
What Would Actually Trigger a Refi Wave in 2026-2027
History is clear: mass refinancing happens when rates fall ~0.75-1.00% below the prevailing stack of existing loans. At today’s 6.7%, the sub-4% cohort needs roughly a 3% world to rationally refi — which is not forecast anywhere. The realistic 2027 scenario is 5.8-6.3%, which serves borrowers who bought or refinanced in 2024-2025 at 7%+, not 2020 lock-in holders. Plan around the cycle you can actually get: our refi-freeze exceptions guide lists the moves available at today’s rates — and our renovation financing comparison shows when a second lien beats moving house.
Frequently Asked Questions
Will mortgage rates drop after today’s Fed decision?
Probably not materially. A hold is expected and a hike is genuinely on the table; either outcome leaves the 10-year anchoring your quote. Watch the CPI releases and Treasury supply, not the fed funds headline.
Should I lock now or float through the announcement?
If you are within 30-45 days of closing, lock with a float-down if the spread is small (often 0.125%). A hawkish surprise today can move quotes a quarter point overnight; the float-down is cheap insurance against exactly that scenario.
Is an ARM smarter right now?
The typical 5/1 ARM runs 50-75bp below the 30-year fixed. If you are certain you will move or refinance inside five years, that discount is real money. Short horizon plus rate uncertainty cuts both ways — model your worst-case reset before choosing.
My credit improved since I bought — does that change the math?
Yes, hugely. A 26% improvement in score band is worth 0.25-0.50% in rate, which on a refi is thousands over the loan. Pull your report first; fix what’s fixable (see our credit-building guide), then apply.
Bottom Line
Today’s Fed meeting is the most-watched rate event of the month, and still the least important input to your mortgage quote — the 10-year Treasury at 5% is writing your rate sheet, not the 3.6% fed funds target. The winners of this cycle are borrowers who do boring things: quote three lenders in one week, lock with a float-down, run the break-even before the emotional math, and consider the 15-year instead of waiting for a 2021 déjà vu that the data says isn’t coming. Your move now beats the perfect rate later.
