What σ means, and how unusual today's rates are
σ measures how far a reading sits from normal. Today's borrowing rate sits further from the old normal than four decades of data would ever produce by chance.
The symbol
σ is the Greek letter sigma. In statistics it stands for the standard deviation: the typical size of the wobble around an average or a trend. If a trend line says rates "should" be 3% and they usually sit within about a point of it, σ is about one point.
A reading of +2σ means "two typical wobbles above the trend". The sign tells you the direction; the number tells you how many wobbles.
The 68-95-99.7 rule
When the wobble follows the familiar bell curve, about 68% of readings land within 1σ of the trend, 95% within 2σ and 99.7% within 3σ. Going one way only, a reading at least:
- +1σ above happens about 1 time in 6,
- +2σ about 1 time in 44,
- +3σ about 1 time in 740,
- +4σ about 1 time in 31,600.
How the clock uses it
The clock fits a straight line through the blended Treasury rate from its 1981 peak to its 2020 low: forty years of steadily cheaper borrowing. The wobble around that line gives the old regime's σ. Because a line that kept falling past 2020 would go below zero, the clock holds the old trend flat at the level where it ended and asks how far today's rate sits above that point. That is the figure on the front-page chart, updated with each day's data. In November 2026 it was about +3.7σ.
How unusual is that?
If today were just one more month of the old regime with bad luck, a reading this high would turn up roughly once in ten thousand months. That is longer than the United States has existed. The plain reading is that the old regime is over: rates are being set by different forces now, which is why every projection on the clock starts from the 2020 turn and fits a new channel from there.
The honest caveats
Treat the "1 in 10,000" as an order of magnitude, not a precise probability. Three reasons:
- Months aren't independent. Rates drift: a high month tends to be followed by another. That makes long runs away from a trend more likely than the bell curve says.
- Markets have fat tails. Big moves happen more often than a bell curve predicts.
- The trend is a choice. A different start date or a curved line would give a different σ.
None of these comes close to closing the gap. Even if the true odds were a hundred times better than the bell curve says, a reading like today's would still be about a 1-in-100 event under the old regime, against a simple alternative that explains it easily: the regime changed.