I do not trust Dr Bussards scaling. He did not have enough data to make those scaling claims. The University of Sydney (Gummersall, 2013) scaled some factors in their simulations:
A. Current in the rings (Amps or AmpTurns)
B. Size of the rings (Meters)
C. Energy of electrons (KeV)
But, they were looking at how many electrons were trapped - not fusion rate. They found some bounds for these machines:
I have not had time to fully digest this plot. It compares the current in the rings against the energy of the material. The higher the current the stronger "the container" trying to hold in plasma. The higher the energy the more plasma is trying to leave. These bounds appear to show where containment can happen. On the face of it, it makes sense. Too much energy or too weak of a container, and the plasma leaks.
I do not think anyone has scaling understood yet. If they do, they have not published (aka the Navy).
A new textbook "Inertial electrostatic confinement (IEC) fusion fundamentals and applications" by Dr. George Miley, offers some basic scaling for fusors (from page 209):
Fusion rate ~ Ions heading towards the inner cage^2
By contrast, polywell scaling is very different. It is based on the same input, the rate of ions flying inward. On page 376, Miley looks at fusion scaling in polywells:
Fusion rate rises, as more ions fly into the center --- by a factor of 5? That is a serious claim. I am still working through this to try and understand it. It is based on simulations, so I am very skeptical. Also, why should it be so different than scaling in fusors?
For source material Miley references two papers, the first which I need to get: Tzonev IV, DeMora JM, Miley GH (1996) "Effect of large ion angular momentum spread and high current on inertial electrostatic structures" and Chacon L, Barnes DC, Knoll DA (1998) "An implicit boundary-averaged Fokker-Plank BAFP code to model spherical interial electrostatic confinement fusion systems."