The 10-year Treasury hit 5.11% on Sep 23, the 30-year 5.40%, and mortgage rates crossed 7%. Social media is four times louder about yields than it was in the first half. We put what the crowd blames next to what the curve is actually pricing, measured which stocks feel it, and checked how much the midterms show up in bond talk.
From the Feb 27 low, the 2-year rose 147bp, the 10-year 114bp and the 30-year 76bp. The front end led and the curve flattened. Almost all of the 10-year move is real yield: inflation expectations barely moved. The Fed cut three times in late 2025, then hiked 25bp at its September meeting.
An inflation or deficit scare usually lifts breakevens and steepens the curve. This did the opposite: breakevens flat, 2s30s flatter. That is the signature of a tighter expected Fed path.
Posts on the "treasury yields" topic averaged about 3,300 a month in the first half. September is running at about 14,200. The surge came in August and September as the 10-year pushed toward 5%. Day by day, the chatter tends to follow yield moves rather than lead them: on rates, the crowd explains the move more than it predicts it.
We tagged 16,899 rate-specific posts (topics "treasury yields" and "yields", plus bond-specific posts on "treasury", Jan 1 to Sep 23, 1.7B interactions) by the driver they cite. War and oil dominate, spiking in March, May and September. The Fed and inflation each draw less than half as much engagement. The curve says the Fed is the transmission: oil and war feed the inflation fear, the Fed responds, and real yields do the work.
| Driver | Crowd | Curve says |
|---|---|---|
| War / oil | #1 by far | Breakevens flat: not pricing lasting inflation |
| Inflation | ~8% | 5Y breakeven down 7bp |
| The Fed | ~9% | 2Y +147bp, curve flatter: the main channel |
| Deficits | ~5% | Would steepen the curve; it flattened |
| Midterms | 0.5% | Not priced yet |
Posts on the midterms have grown about 6x since January, to a pace near 300,000 a month. Yet only 0.5% of bond-market engagement mentions them. Where the two meet, the posts are about affordability, gas prices and the national debt, not the yield curve. Before Nov 3 the bond market is trading the Fed. After, the channel that matters is the fiscal path: which party controls Congress, and how much Treasury has to borrow.
For 923 of the 1,000 largest US stocks, we measured the average daily move for each +10bp in the 10-year, Mar 1 to Sep 22. Energy is the only sector that rises with yields, consistent with oil sitting underneath both. Homebuilders and the mortgage lender Rocket track the 10-year most tightly of any large caps. Airlines and cruise lines show up because they are hit by the same oil move, not only by rates.
| Stock | Corr. | Per +10bp day | Since Feb 27 |
|---|---|---|---|
| RKT | −0.70 | −5.8% | −32% |
| DHI | −0.63 | −3.1% | −10% |
| LEN | −0.63 | −3.3% | −32% |
| CCL | −0.62 | −4.3% | −29% |
| PHM | −0.61 | −2.9% | −11% |
| UAL | −0.61 | −4.2% | +8% |
| OXY | +0.55 | +2.8% | +8% |
| COP | +0.54 | +2.4% | +12% |
| XOM | +0.52 | +2.0% | +4% |
If the war-and-oil story is right, it rises. So far it is lower than in February. A move above 2.5% would change the read.
Flattening says the Fed. A steepener with the long end leading would say fiscal worry and term premium.
The midterm signal for rates is who controls Congress, not the affordability debate. One-party control raises the odds of a bigger deficit.
Treasury's borrowing plan lands right around the election; the Fed meets days before it. Watch whether social attention shifts from war to deficits.
Every figure on this page, free, no registration. The .md file is a companion brief for AI assistants: drop it into Claude or ChatGPT next to the data and your model starts from the right place.

The commercial feed runs this hourly with history: attention, sentiment and narrative for any ticker, topic or watchlist, point-in-time.
joe@lunarcrush.com