Fit equation, where N's are numbers of types of events, G's are skewed Gaussians for pion and eta, and B is a 3rd order Chebyshev polynomial

Each skewed Gausian has 3 fit parameters
B(m) has 3 fit parameters
The N's are also fit parameters

Each summand function is independently normalized so that the overall normalization is the 'fitted' total number of events:

Letting D(m) denote the data histogram, the pion signal purity is

And the impurity (background + eta "purity" in the pion signal window)

The attached document is a fit to run 12 pp510 mass distributions, for:
- pT above threshold and below 10 GeV
- E between 30-40 GeV
- Large Cells
--- each page is a "run-set", usually 5-7 runs each, in chronological order
--- the first 3 pages show some summaries of fit parameters and purity; left column is distribution, right column is vs. run-set