Source: Derived on-site from the Shiller CAPE series (expanding OLS, point-in-time)
Data through: September 2026 · updated Oct 1 · Updates: Monthly (current month refreshed nightly)
MThe three bars show the last 3 months for this indicator, oldest left to newest right. Bar height reflects each reading relative to the other two — the tallest is the highest of the three, the shortest the lowest.
Historical context: Currently in the 91st percentile historically — elevated vs historical norms.
What this means right now: CAPE is one to two standard deviations above its own trend — expensive against everything in its history to this point. This describes where the market sits; it is not a forecast.
CAPE: sigma vs trend · Monthly · 1881–2026
NBER recessions
Scroll to read · zoom and pan in Expand
Monthly values
CAPE: sigma vs trend — the last 12 months
Every month's reading, with the band the heatmap gives it. The chart above shows the same series in full.
Month
Value
Band
Sep 2026
+1.6σ
Concerning reading
Aug 2026
+1.6σ
Concerning reading
Jul 2026
+1.6σ
Concerning reading
Jun 2026
+1.6σ
Concerning reading
May 2026
+1.6σ
Concerning reading
Apr 2026
+1.4σ
Concerning reading
Month-end readings of the same series the chart shows, from cape_trend_deviation. Values are the current record, not point-in-time: a month that has since been revised shows its revised value.
What is the CAPE sigma vs trend?
How unusual today's CAPE is against its own history. A trend line is fitted through every CAPE reading up to each month. The reading is how far the latest value sits above or below that line, in standard deviations — Grantham's yardstick, with two standard deviations above trend his bubble line. Each month uses only the data available then, so past readings never change. It differs from the percentile on the Valuations page, which ranks a reading against all history: the trend allows for CAPE's slow upward drift over 150 years. The first 24 months after 1881 are unreliable while the trend settles. See Valuations for the record of expensive and cheap periods on the percentile basis.
Our heatmap colors this indicator against historically significant thresholds:
Below −2.0σ
Deeply below its own trend — as cheap as this measure gets
−2.0σ – −1.0σ
Below its own trend — the cheap side of its history; not a warning
−1.0σ – 1.0σ
Near its own trend — the normal range, holding 52% of its months
1.0σ – 2.0σ
Well above its own trend — expensive by its own history
Above 2.0σ
Grantham's bubble line — CAPE has been this far above its trend only in 1928–30 and 1997–2001, peaking in 1929 and 1999
Historical Extremes — What Happened Next?
When this indicator reaches extreme levels, history shows consistent patterns:
Sep 1929The 1929 peak
+3.01σ
The highest reading in the series, set the month before the October 1929 crash. CAPE was three standard deviations above its own trend — far outside anything its history to that point had seen.
Jul 1999Dot-com era peak
+2.95σ
The highest reading of the modern era, in July 1999. CAPE stayed above Grantham's bubble line (+2σ) for almost four years, from May 1997 to February 2001 — 46 months, the longest in the series — before the 2000–2002 decline.
Nov 1907Panic of 1907
−4.57σ
The lowest reading in the series, set during the 1907 banking panic. This is the cheap end of the scale: a market far below its own trend is cheap, not dangerous, on this measure.