diff --git a/docs/FLOOD_FORECASTING.md b/docs/FLOOD_FORECASTING.md
index 3b91f5f..e798865 100644
--- a/docs/FLOOD_FORECASTING.md
+++ b/docs/FLOOD_FORECASTING.md
@@ -284,6 +284,19 @@ almost no positives (P.84 0.570, P.82 0.302 on 0.22% of test hours). P.92 and P.
 have no evaluable warning metric at all: neither crossed 3.0 m often enough in the
 test span (P.92 not once, P.20 in 0.09% of hours) to score.
 
+### The September 2025 flood, as the deployed configuration saw it
+
+![Observed vs predicted through the September 2025 flood — model trained only
+through 2024](img/backtest-2025-p1.png)
+
+This is the strongest single piece of evidence, because it uses the exact
+deployed configuration (train ≤ 2024-12-31) on an event it never saw: the
+model's **first alert came 26 September 2025 at 18:00, exactly 24 hours before
+the river crossed 3.70 m**, and it predicted a 4.00 m peak against an actual
+3.93 m — within 7 cm. Note the discrimination: the near-miss crest of 3.51 m on
+26 September never triggered an alert, the probability fires only for the real
+event, and it stands down as the water recedes.
+
 ### Headline validation: the October 2024 record flood
 
 The holdout above never sees a true extreme, because the 2024 flood is in the
diff --git a/docs/img/backtest-2025-p1.png b/docs/img/backtest-2025-p1.png
new file mode 100644
index 0000000..e607a0a
Binary files /dev/null and b/docs/img/backtest-2025-p1.png differ