19 February 2026 Thursday

W: 6 DW: 2 OT: 2 ex

• See above

• Useless

• See above

HI Stephane,

I agree this is definentely a point of confusion. Based on the presentations from the boosted teams I think the main issue in scaling is that Boosted analysis has significant statistical uncertianty on the expected 95% CL, coming from a non-trivial statistical uncertianty on the expected background prediction. (In the resolved analysis, the statistical on uncertianty on the expected background is much smaller.)

This effect is not currently captured in the Brazilian bands we use to report the expected limits. The yellow and green colored bands reflect the level uncertianty we expect on observed limits (eg: including both the statistical uncertianty on the background AND on expected SR data yeild). We dont separate out the statistical component on the expected limits, and simply quote the point estimate for the 95% CL.

We could potentially convey this with a modified version A (cartoon) of how we might show both pieces would something ilke:

• Triangles
• H gets gray belt!!

• Safety factor from 1.9 -> 2.11

• All well that ends well
• The English and Their History