Your 6.4% Refi Window Closed This Week. Here Is What Actually Moved.

Update, August 7, 2026: Rates never came back. Freddie Mac’s 30-year average hit 6.69% for the week ending August 6, a fifth straight weekly rise and the highest reading since late July 2025. It is also the first time in 10 months the 30-year has sat above its own year-ago level of 6.63%. The 15-year eased to 6.01%. The July 2 low of 6.43% now looks like the best number of the summer, and the Fed held on July 29 with three members voting for a hike instead. Everything below still stands. If you are still waiting for a 6% handle, you are betting against five straight weeks of tape.

If you were watching for a 6% mortgage rate to lock a refi, that window opened last Thursday and closed by Wednesday. Freddie Mac’s weekly survey hit 6.43% on July 2, a seven-week low. Six days later, Bankrate’s daily 30-year rate was back at 6.52%. That is not the Fed. That is oil.

Here is what happened. The Iran ceasefire unraveled Wednesday, July 8. Oil jumped from about $69 a barrel to $73.61 in a single session, per CNBC. The 10-year Treasury yield climbed to 4.57%. Mortgage rates followed within hours. “Oil prices have surged, bringing bond yields and mortgage rates higher once again,” Bankrate reported.

Translation: bond investors treat rising oil as rising inflation, and rising inflation means they demand a higher yield on the 10-year Treasury to hold it. Mortgage rates ride on top of the 10-year with a spread. When the 10-year moves, your rate quote moves the same afternoon. The federal funds rate has nothing to do with it in the short run.

On a $400,000 loan, moving from 6.43% to 6.52% costs about $24 more a month, roughly $8,600 over the life of the note.

That’s a small monthly hit and a real total. If you were 30 days from closing at 6.43%, you either had a lock, or you just paid for a headline. If you were waiting to lock, the July 2 window is gone.

Here’s the move. If you have an active rate lock: check the expiration date on your commitment. If you close before it expires, sit tight, your rate is protected. If your closing is close, call your loan officer and ask two things. What does a 15-day lock extension cost, usually $150 to $500. And do you have a float-down option, which lets you capture a lower rate for a small up-front fee if the market improves before closing. If you paid for a float-down and the market drops, use it. If you did not, you are stuck at your locked rate, which is not the worst outcome if the trend keeps going up.

If you were about to lock but hadn’t yet, don’t chase yesterday’s number. Pull quotes from two lenders today, on the same day, with the same credit pull, and compare the all-in cost, not just the headline rate. Ask each for both a lock and a lock-and-shop option. If the math on your break-even works at today’s rate, lock it. Compare current options on our best mortgage lenders page and run your own numbers through the mortgage calculator before you sign.

If you were waiting for a 6% flat, wait longer. The Fed held on July 29 and left the target range at 3.50% to 3.75%, and three of the twelve voters wanted a quarter-point hike instead. A Fed cut is not the trade the market is arguing about right now. The Fed does not move mortgage rates directly anyway. What moves them is the 10-year Treasury, which is what moves when Iran, oil, or inflation surprises the market. None of those are on a schedule.

For the technical readers: mortgage rates typically run 1.5 to 1.9 percentage points above the 10-year Treasury yield. That spread widens when the market thinks prepayment risk is high (rates falling) and narrows when it thinks refis are dying (rates rising). It has been running above the long-run average since 2022. With the 10-year around 4.67%, a normalized spread would put the 30-year in the low 6s instead of pushing 6.7%. That day may come. It has not come yet.

How Candid Yak makes money. Some of the products we write about pay us if you apply or sign up through our links. That never changes our verdict, our rankings, or the numbers in this article. We call a bad deal a bad deal whether it pays us or not. Some brands shown in our comparison tools are placeholder examples while we finalize partner agreements, and we label them as such.