G25 source selection: how good models overfit and honest panels are built
Adding sources always lowers a G25 fit distance and routinely worsens the model — a worked demonstration where the fit improved from 0.0110 to 0.0068 while the story collapsed. The rules that keep a panel honest.
Related posts
What is a good G25 fit distance? The number everyone reads wrong
The fit distance measures how far a mixture still sits from your coordinate — not whether the model is right. Calibrated bands for reading one, why lower stops being better, and the checks that actually validate a model.
4 min read
nMonte explained: the R script behind every G25 admixture fit
Ger Huijbregts's nMonte turned Global25 rows into ancestry percentages and founded a whole tool tradition. How the Monte-Carlo search works, what nMonte3 changed, what the fit distance is, and the discipline the script never enforces.
3 min read
G25 scaled vs unscaled coordinates: which to use, how to tell them apart
Every Global25 row exists in two forms, and mixing them is the hobby's most common silent error. What scaling does mathematically, which form each tool expects, and the tell-tale signs of a mixed comparison.
3 min read