Rare Kaon decay may shed light on new physics....maybe

[member 656954]

TL;DR Summary

Florida State University physicists believe they have an answer to unusual incidents of rare decay of a subatomic particle called a Kaon that were reported last year by scientists in the KOTO experiment at the Japan Proton Accelerator Research Complex.

The phys.org summary had me intrigued and the resulting new paper makes for heavy reading at my level of knowledge, but it was the underlying reasoning of the methodology that I really question:

One of the best ways to search for new physics (NP) beyond the SM is to look for events that are predicted to be extremely rare in the SM by a theoretically clean calculation. An observation of just a few such events could then constitute a robust evidence of NP.

One of the authors notes "It's so rare, that they should not have seen any."

From that, the authors propose two mechanisms to explain it involving new physics, but it seems a literal 'jump the gun' to me.

According to their calculations, there could be two possibilities for new particles. In one scenario, they suggest that the Kaon might decay into a pion—a subatomic particle with a mass about 270 times that of an electron—and some sort of invisible particle. Or, the researchers in the KOTO experiment could have witnessed the production and decay of something completely unknown to physicists.

Interestingly, to me at least, FSU Assistant Professor of Physics Kohsaku Tobioka said, "It might be noise, but it might not be. In this case, expectation of noise is very low, so even one event or observation is very striking. And in this case there were four."

Four events (with one of them actually suspected to be noise) seems like a too small number to propose anything apart from the special data runs that are now planned to investigate this further, but I was hoping that more knowledgeable minds might have some thoughts on the method and the proposed mechanisms.

 

[Vanadium 50]

First, I consider having theorists publishing based on preliminary experimental data bad practice and borderline unethical. The experiments say "we're not sure and we're checking this", and the theorists go full speed ahead, often putting words in the experimenters' mouths.

Next, the experiment has said

At the KAON2019 Conference:

• We did not claim the observed events as signals
• We did not give any numbers on the branching ratio or physics results (emphasis mine)
• We simply described what we had found and what we knew

Why?

• In blind analysis, the standard way is to give a result regardless of the contents inside the signal box.
• However, the number of events in the signal box was beyond our reasonable expectation.
• To be scientifically correct, we decided to do further checks on the events, detector status, and background estimations before announcing a result.
• To be honest to the scientific community, we should not hide the factthat we had opened the signal box.
• To be scientifically correct and to be honest, at the KAON2019 Conference, we showed the events and explained exactly what had happened.

Given this, it is particularly obnoxious that theorists (and not just this team) is going around publishing "KOTO" branching fractions, in effect "scooping" the experiments themselves. It's also disingenuous for theorists to behave this way on the one hand, and on the other criticize the experiments for not showing their results until they are almost ready to publish.

Now, let's look at what happened. The experiment ran in 2015 and saw no events. Then the accelerator and detector made some configuration changes (hoping that they would be improvements) and got 1.5x as much data. They see 4 events, one of which they say is certainly background. They also say that the known backgrounds sum to 0,05. So just looking at background, they see 20x as much as they expect. Clearly the backgrounds are not yet under control. Declaring an observation of signal when the backgrounds are not under control would be scientifically irresponsible, which is why KOTO did not do that.

Next, there is a fifth event near but not in the signal region. That increases the likelihood of background "sneaking in".

Next, the theorists blew it when calculating the branching fractions. They looked at only the later data, ignoring 2015. But unless they want to propose that kaon decays changed in 2016, they need to include the 40% of data from 2015 with zero events. So every plot is wrong.

Color me annoyed.