Why does using the wrong starting scale for PDF evolution lead to incorrect predictions at the LHC?

Why the Starting Scale Matters for PDF Evolution

Parton distribution functions (PDFs) describe how quarks and gluons are distributed inside a proton—but this distribution changes depending on the energy scale at which you measure it. Think of it like examining a photograph at different zoom levels: the "grainy structure" you see depends on how closely you look.

The DGLAP evolution equation [AG-2025.09-678] governs how PDFs change with energy scale. If you start your calculation at the wrong scale, you're essentially beginning with an incorrect "photograph," and all subsequent predictions downstream will be skewed.

### The Practical Problem at the LHC

At the LHC, wrong starting scales create two concrete failures:

1. Distorted high-energy predictions. LHC searches for new physics often look at rare, high-energy events where quarks and gluons carry large momentum fractions. If your PDF evolution begins at an incorrect scale, the predicted rates of these rare events will be systematically wrong. This matters because discovering new physics depends on spotting excesses above the Standard Model prediction—if your baseline is off, you miss signals or see false alarms [AG-2024.04-1155].

2. Absorbing new physics into PDFs. More subtly, when global PDF fits use data from many experiments to constrain the quark and gluon distributions, they can inadvertently "fit away" hints of genuine new physics by adjusting the PDFs to accommodate anomalies [AG-2024.05-1304]. Starting evolution at the wrong scale amplifies this problem: it forces the fitting procedure to compensate for scale errors by warping the PDF shapes themselves, especially in the large-x (high-momentum) region where new-physics signals live [AG-2026.02-1300]. You then mask the very discovery you're hunting for.

### Why Precision Matters Now

As the LHC accumulates more data—especially heading into the High-Luminosity phase—theoretical predictions must match experiment at unprecedented precision [AG-2024.04-1155]. A wrong starting scale introduces a systematic bias that grows more obvious as statistical uncertainties shrink. Modern analyses now ask: should we use conservative PDFs that exclude potentially contaminated data, or fit PDFs and new-physics signals simultaneously to disentangle them? [AG-2026.02-1300]. Both strategies depend critically on getting the scale right from the start.