In particle physics, the term model building usually refers to a construction of new quantum field theories beyond the Standard Model that have certain features making them attractive theoretically or for possible observations in the near future. A model builder typically chooses new quantum fields and their new interactions, attempting to make their combination realistic, testable and physically interesting.

The phrase "model builders" is approximately equivalent to the phrase "phenomenologists." Model builders constitute a different group from the experimentalists and "pure theorists" (such as researchers of string theory); model builders are technically theorists, but with an emphasis on using current tools to fit data (see: instrumentalism), rather than on the more long-term pursuit of a 'final', uniquely 'correct' theory of nature (see: scientific realism).

Model building is mostly speculative because current particle accelerators can only probe up to a few hundred GeV, where physics is well described by the Standard Model. There is a result of renormalization group theory stating that at low energies, models flow toward universality classes and different models may flow to the same universality class. There reason why there are so many models beyond the Standard Model - hundreds or even thousands - that exist is because the Standard Model is a universality class. The only way to distinguish between these model based solely upon low energy experiments is to look for irrelevant couplings suppressed by the "cutoff scale". And there is no way most of these models can be right. Only one such model out of the hundreds can be right (possibly along with models which are dual to it or effective theory versions of it) and it may even be one which hasn't been considered at all yet!

This is analogous to a similar situation in condensed matter physics. If all known is the equations of hydrodynamics, the equation of state and the transport properties of a fluid, no one would be able to tell whether the fluid is made up of atoms or not. And you certainly wouldn't be able to tell if the particles are molecules made up of atoms or plain old unbound atoms or that an atom is made up of electrons orbiting a nucleus. Similarly, if all you know is the electrical properties of a dielectric/insulator/conductor/superconductor, you wouldn't be able to figure out the microscopic electron wavefunctions.

Unfortunately, all the attempts to look for irrelevant couplings, such as proton decay with flavor changing neutral currents, or anomalous magnetic moments with anomalous electric dipole moments) have come up with very strict tiny bounds. The only nonzero irrelevant coupling that we have measured is the neutrino oscillations coupling. This makes it possible to tweak most models so that the irrelevant couplings become tiny either by raising the scale at which new physics appears or introducing some ad hoc symmetries forbiding these couplings.

Of course, it may be argued that there are patterns among the marginal and relevant couplings so that among all the possible models which give rise to the Standard Model universality class, some of them are much simpler than the others and the simpler models have fewer free parameters than there are Standard Model parameters and thereby allowing us to make postdictions, but it is certainly an article of faith that the "true" UV completion of the Standard Model is simple.

See also

• epicycle
• Rube Goldberg machine

Particle physics