I'm with ogden here. This is a number pulled out of thin air (excuse the pun).
Guys, I understand that there would be scrutiny and no offense taken. Koreans trained me

There as couple of published peer reviewed papers by me back in 2014 and 2016. IET Electronics letters and URSI, but they were not claiming a ZW wave back then.
Anyway coming to the number 56%.
This number has not been pulled out of thin air.
Way back in 2014-15, this number used to be as low as 12%-16%.
Back in 2014, for an input power of 20 watts on the Tx, I used to get barely 3.2 watts on the load end on the energy meter.
I also got stuck with one typical problem, i.e. the grounding!
If I used a lambda/8 length wire and a flat GI metal sheet lying on the ground arrangement connected to one end of the terminal of the Rx, the efficiency would jump to 33-34%.
Thats when I realized that the TM wave makes the metal sheet a quasi- equipotential surface.
The RLC lumped elements would not do anything to help the situation. There used to be simply not enough voltage oscillations across the terminals of the Rx. This is because, the damn thing needs an electrical length across its terminals.
HFSS simulation would not show up any resonance peak either, but the VNA would. Then came good old Balanis to rescue.
July 2018:
One fine morning while taking a massive dump in July 2018 (after a very frustrated night of beer drinking), I attached a Tesla Transformer "like" coil arrangement to the existing ground backed impedance resonator on HFSS. The simulation started showing up the peak at the expected freq zone.
In couple of days from that time, I optimized my design and built it.
When I measured the I/P Power (20 watts) and O/P (11.4 watts)using the energy meter, I got the Tx to 1 Rx efficiency of 56%. Then I used a second Rx and the overall efficiency jumped to 65%-67% range. I also built a makeshift partial shield box, it worked through the shield as well with a marginal drop of 3-4% in efficiency.
Until that point in time, people (folks at my Startup) called it various names-e.g. capacitive system, radiative system etc.
The HFSS clearly showed the ZW equiphases, Slow attenuation rate of E-field in the transverse direction and evanescent field decay away from the metal-air interface.
When I showed those results, the external expert was extremely impressed and said, "Sai, you dont have to prove anything to any of these skeptics. They clearly have no idea about Physics"
Long story short, I was kicked out of my startup, as the CEO acquired all my 7 patents from my PhD Uni. His management team didnt want to work with me as they thought, they gathered everything needed to do it themselves.
Lets wait for the peer review, if it gets through I will share the entire set of results. If it doesnt get through, I will just hang my boots and leave.