On practical naturalness and its implications for weak scale supersymmetry

TL;DR

This paper compares various measures of naturalness in weak scale supersymmetry, highlighting their differences, overlaps, and implications for model selection, especially considering the string landscape's influence on naturalness and finetuning.

Contribution

It provides a comprehensive comparison of naturalness measures in supersymmetry and discusses how the string landscape impacts their interpretation and the likelihood of natural models.

Findings

EW naturalness is the most conservative and model-independent measure.

Different naturalness measures can overestimate finetuning by factors up to 1000.

The string landscape favors EW natural SUSY models over finetuned ones.

Abstract

We revisit the various measures of naturalness for models of weak scale supersymmetry including 1. electroweak (EW) naturalness, 2. naturalness via sensitivity to high scale parameters (EENZ/BG), 3. sensitivity of Higgs soft term due to high scale (HS) radiative corrections and 4. stringy naturalness (SN) from the landscape. The EW measure is most conservative and seems unavoidable; it is also model independent in that its value is fixed only by the weak scale spectra which ensues, no matter which model is used to generate it. The EENZ/BG measure is ambiguous depending on which ``parameters of ignorance'' one includes in the low energy effective field theory (LE-EFT). For models with calculable soft breaking terms, then the EENZ/BG measure reduces to the tree-level EW measure. The HS measure began life as a figurative expression and probably shouldn't be taken more seriously than that. SN is closely related to EW naturalness via the atomic principle, although it is also sensitive to the distribution of soft terms on the landscape. If the landscape favors large soft terms, as in a power law distribution, then it favors m(h) ~ 125 GeV along with sparticles beyond present LHC reach. In this context, SN appears as a probability measure where more natural models are expected to be more prevalent on the landscape than finetuned models. We evaluate by how much the different measures vary against one another with an eye to determining by how much they may overestimate finetuning; we find overestimates can range up to a factor of over 1000. In contrast to much of the literature, we expect the string landscape to favor EW natural SUSY models over finetuned models so that the landscape is not an alternative to naturalness.