So I'm using that analysis on geoarcher's data because that is all the data I have. That said, the analysis is still actually perfectly valid. Let me explain why.
Adam Karpowicz actually has his own analogous metric to this one. While we divide kinetic energy output by poundage at full draw, he divides kinetic energy stored by poundage at full draw. While slightly different, really, they're trying to look at similar things. Dividing energy stored by poundage at full draw tells you how efficient the bow is at storing energy with a ratio that allows fair comparison across a wide range of different poundage/power level bows. This was probably chosen by him because, as a bowyer, he was more interested in comparing how efficiently his bows generate stored energy. Now contrast this to "my" metric, energy output over poundage at full draw. (I should note I was unaware of Adam's metric at the time I came up with mine) Less interested in energy storage efficiency, I was more interested in practical bow efficiency.
So lets take a quick example here. You're looking to shoot some particular arrow with a certain power. If you were infinitely strong, it wouldn't matter, as you could simply buy ever more powerful bows until you achieved that terminal energy. Being a human of finite strength though (I assume) you can't achieve that, so you look for a bow that, relative to the maximum draw weight you can hold, produces the highest kinetic energy output. So this is the best metric for you, and it allows level comparisons of bows of vastly different draw weight.
But wait, there is more. So if you integrate a force draw curve and divide the energy output by the energy stored, you do get the bow's actual efficiency. (more or less, you're excluding various sources of loss that aren't the bow's fault) But then what? This has limitations though, because in some respects it is the special olympics of comparisons as it allows everyone to compete on their own level. That means bows inefficient at storing energy get to be compared to bows efficient at storing energy on level footing, as they only are asked to ever spit out as much energy as they stored in the first place. While all the bows I tested in my analyses and most of the bows seen here are highly reflexed so store a comparatively large amount of energy, this is not universally true. Take an English longbow which, when unbraced, is only roughly straight. Or take wood selfbows from a variety of arid climates, like Nubia/Egypt, which are highly deflexed and hold little or no string tension at brace height. (someone correct me, last I looked, years ago, it was still a debate whether they were braced under tension at all) These bows would store comparatively little energy as early draw weight is very low. So say they were, hypothetically, capable of 75% efficiency. (I genuinely have no idea how efficient they are) Would it be fair to compare one to a composite bow which also is capable of 75% efficiency? What if the ratio of stored energy to poundage at full draw of the composite bow was 1.05 whereas the arid climate selfbow was a mere 0.68? (again, just a shot in the dark) Probably not. And then you look at the outer limb mass of these selfbows next to a short slick efficient composite. Is stored energy over poundage really fair? Unless you plan to shoot extremely heavy arrows with both, again probably not. So now we compare energy output over poundage at full draw. And that is it, that is where the rubber meets the road: how much energy can this bow spit out for an archer of a given strength.
I hope that all made sense.
*edit*
I stumbled on something and didn't want to double post:
http://www.cinnabarbow.org/marinerbows/scorpius.html
So I like short, fast, and efficient bows. (obviously) This jumped out at me though, 210fps with a 416 grain arrow on a 30" draw 40#s.
So the most efficient bow in Adam Karpowicz's set of inhumanly efficient Turkish bows managed 94% efficiency shooting a 23GPP arrow. If you divide energy output over poundage at maximum draw for that particular setup, you get .932.
A 416 grain arrow traveling at 210fps is 40.75 foot pounds..... from a 40# bow drawn 30". That gives you an efficiency ratio of 1.02. Turkish style bows, what with their higher brace heights (which this has) and shorter draws (which this also has) are disadvantaged in this metric, as they store less energy. My Spitfire, which is quite efficient at storing energy with its strong reflex, low brace height, and long draw, pulls 43#s at 31" and has stored 46 foot pounds of energy. So lets give this bow the benefit of the doubt, and say that it is storing as much energy as a lower brace height longer draw bow. So shooting a 10.4GPP arrow that'd be 88.6% efficient.
All of these numbers strike me as a little difficult to swallow, particularly for a glass bows since, these days, the most efficient bows use carbon and for good reason. I don't suppose anyone has gotten their hands one one of these bows and can provide some data?
*edit2*
Just found on their facebook page:
"50# @ 28" = 57" @ 30"" (ref. https://www.facebook.com/bambooarchery/ … 496827919)
While not totally dramatic stacking, it hardly is indicative of an unimaginably fat force-draw curve that'd store unusual amounts of energy. Amusingly, it also seems to match the Grozer almost perfectly.