ValidAnytime
  • Try it free
  • Observatory
  • Pricing
  • Docs
Sign inStart free
ValidAnytime

A guaranteed false-alarm budget across your whole fleet, valid no matter how often you look.

Made by Compiled Intelligence — a frontier AI lab working on quantitative finance from first principles; ValidAnytime is the monitoring we built for our own model fleets, productized.

Product

  • How it works
  • Pricing
  • Observatory
  • Try it free
  • Dashboard

Solutions

  • Who it's for
  • For LLM engineers
  • For MLOps engineers
  • For SRE & platform
  • LLM eval regression
  • Model drift detection
  • Alert fatigue

Developers

  • Docs
  • Quickstart
  • Python SDK
  • API reference

Learn

  • Benchmark
  • Detector guides
  • Glossary
  • Compare
  • Why dashboards lie
  • Blog
  • Anytime-valid 101
  • Winter Storm Uri case study
  • Incident ledger
  • Observatory methodology

Company

  • Contact
  • Privacy
  • Terms

Observatory alarms & product updates, by email

Rare emails only. We store your address and the tag “footer” — nothing else. Unsubscribe anytime.

© 2026 ValidAnytime. All rights reserved.

Every alarm ships with its guarantee_tag and theorem_ref.

Observatory
SO

Has Southern Company (Southeast)'s demand forecast drifted?

SOCO · updated Jul 6, 22:08 UTC

No regime change detected

Southern Company (Southeast) — no forecast-error regime change detected

The day-ahead forecast's error has stayed within its calibrated band. Not a certificate that nothing happened.

quiet streak 1 daybudget ≈1 false page per BA-yearwatching since Jan 2024
1.83%
mean abs. forecast error

of demand (MAPE)

27,271 MW
mean demand
291 MW
typical error

median |error|

3,155 MW
worst-hour error

99th pctile |error|

The evidence, last two weeks

Because the baseline is the operator’s own forecast, seasonality is subtracted for free — what’s left is the miss. The detector pages only when that miss enters a new regime and clears the page line.

0 = the operator’s forecast is exactly rightforecast error — actual demand minus day-ahead forecast (MW)page line (≈1 false page per grid-year)accumulated evidence (anytime-valid log-statistic)Jun 20, 2026Jul 5, 2026
Real EIA-930 data for Southern Company (Southeast) — the last 336 hourly points. The forecast is the grid operator’s own day-ahead demand forecast (not ours); the error is actual − forecast, peaking at +11,642 MW in this window. The evidence statistic stayed under the page line: no regime change detected. That is not proof nothing happened. Back-in-band is a display heuristic, not a certificate of absence.

Forecast-error regime changes10 in the monitored history

Each is a dated page inside the stated budget. Labels attach only where a window overlaps a documented grid-stress event — never invented. We detect that the forecast’s error changed regime, not that the grid failed or why.

  • EndedJul 3, 2026 → Jul 5, 2026Late-June 2026 heat wave
    peak evidence
    9.43
    forecast miss at onset
    −6,896 MW
    load at onset
    93th pctile of this grid
    back in-band
    Jul 5, 2026
  • EndedMar 4, 2026 → Mar 6, 2026
    peak evidence
    9.67
    forecast miss at onset
    +495 MW
    load at onset
    50th pctile of this grid
    back in-band
    Mar 6, 2026
  • EndedFeb 25, 2026 → Feb 26, 2026
    peak evidence
    9.40
    forecast miss at onset
    +354 MW
    load at onset
    62th pctile of this grid
    back in-band
    Feb 26, 2026
  • EndedFeb 1, 2026 → Feb 3, 2026January–February 2026 cold wave (Winter Storm Fern)
    peak evidence
    10.32
    forecast miss at onset
    −4,248 MW
    load at onset
    79th pctile of this grid
    back in-band
    Feb 3, 2026
  • EndedJan 30, 2026 → Jan 31, 2026January–February 2026 cold wave (Winter Storm Fern)
    peak evidence
    9.19
    forecast miss at onset
    −3,396 MW
    load at onset
    64th pctile of this grid
    back in-band
    Jan 31, 2026
  • EndedJan 18, 2026 → Jan 19, 2026
    peak evidence
    9.16
    forecast miss at onset
    −2,946 MW
    load at onset
    53th pctile of this grid
    back in-band
    Jan 19, 2026
  • EndedDec 11, 2025 → Dec 12, 2025
    peak evidence
    9.73
    forecast miss at onset
    −4,849 MW
    load at onset
    43th pctile of this grid
    back in-band
    Dec 12, 2025
  • EndedSep 13, 2025 → Sep 15, 2025
    peak evidence
    9.59
    forecast miss at onset
    −2,736 MW
    load at onset
    67th pctile of this grid
    back in-band
    Sep 15, 2025
  • EndedFeb 20, 2025 → Feb 23, 2025
    peak evidence
    11.59
    forecast miss at onset
    −4,266 MW
    load at onset
    68th pctile of this grid
    back in-band
    Feb 23, 2025
  • EndedJan 31, 2025 → Feb 1, 2025
    peak evidence
    9.22
    forecast miss at onset
    −2,113 MW
    load at onset
    22th pctile of this grid
    back in-band
    Feb 1, 2025

How this verdict is formed

The hourly error runs through a conformal band that learns Southern Company (Southeast)’s own normal forecast-error rate, into a Shiryaev–Roberts e-detector on a bounded rolling window. An alarm fires only when accumulated evidence crosses the page line — and because the statistic is anytime-valid, checking as often as you like never inflates the false-alarm rate. Full recipe, and exactly what we do not claim, on the methodology page.

Embed this badge

A live status pill for Southern Company (Southeast), refreshed every publish. Drop it in a README, status page, or dashboard — it always shows the current verdict and links back here. It reads no regime change while the forecast holds, and flips to forecast drift detected when the e-detector pages.

Southern Company (Southeast) demand-forecast monitor · ValidAnytimelive preview
Markdown
[![Southern Company (Southeast) demand-forecast monitor · ValidAnytime](https://validanytime.com/observatory/badge/SOCO)](https://validanytime.com/observatory/SOCO)
HTML
<a href="https://validanytime.com/observatory/SOCO"><img src="https://validanytime.com/observatory/badge/SOCO" alt="Southern Company (Southeast) demand-forecast monitor · ValidAnytime"></a>
Monitor your metricsBacktest your data