tonio
made their 3rd forecast (view all):
Probability
Answer
0% (0%)
Less than or equal to 1349
35% (+5%)
Between 1350 and 1499, inclusive
65% (0%)
Between 1500 and 1649, inclusive
0% (-5%)
Between 1650 and 1799, inclusive
0% (0%)
More than or equal to 1800
Why do you think you're right?

As of September 11, 2025, the CDC reports 1,454 confirmed measles cases, but systematic analysis of state health department websites (@ctsatsreveals the true current total is likely around 1,463 cases when accounting for reporting lags. This means only 37 additional cases are needed to cross the 1,500 threshold over the final three weeks of the forecast period (requiring just 12.3 cases per week). According to the CDC data, ~70% of weeks since June 2025 have exceeded 12/week, with weeks in late August showing 20, 23, and 29 new cases.

With multiple active transmission sources remaining (Wisconsin, Colorado City) most U.S. schools have reopened – and while no major school-associated outbreaks have been confirmed, measles' 10-14 day incubation period means any school transmission effects would only become visible in late September.

Altogether, despite lower reports the last two weeks, crossing the 1,500 threshold is more likely than not.

Files
Why might you be wrong?

The most recent CDC data taken at face value sugguests a deceleration in new cases, which could sugguest public health containment efforts may have achieved a step-change reduction in transmission. The major outbreaks that drove case counts earlier in the year (Texas, New Mexico) are over or slowed – removing two significant sources of new infections. There also doesn't appear to be any confirmed classroom transmission yet, despite weeks having passed since most schools reopened. If the deceleration in the data represents a genuine epidemiological shift rather than a temporary reporting artifact, the final count could remain just below 1,500.

Files
ctsats
made a comment:

The most recent CDC data taken at face value sugguests a deceleration in new cases

I don't think this is actually the case: the increase in the aggregate cases reported in the last 3 weeks has been +23, +23, and +33, for an average of ~26/week; in the 5 weeks prior there were +10, +14, +23, +9.5*, +9.5* cases, for a weekly average of ~13/week. So, the current weekly average seems to be double that of most of August.

[*imputed; what was actually reported was 19 cases for 2 weeks - one report was not released as scheduled]

Even confining the discussion to "new" cases, i.e. the ones reported in the last week only (to be revised upward in subsequent reports), numbers have gone up relative to August: notice that, around Aug 22, it was very natural (and not unreasonable...) for people to assume that the last week reported number would be routinely zero (0); but this number was 2 in the previous report of Sep 3 and 3 in the latest one.

Just sayin'...

Files
Files
Tip: Mention someone by typing @username