Bringing in some thoughts from discussions elsewhere.
Because of Larry's high power potential and normally high RPM I didn't find his results all that surprising.
He has been training himself to a vary high cadence on short cranks using gears that work well for that. The Q-Ring is designed to minimize the resistance in the deadspot. So if you project that out. Someone on 175mm cranks; mashing at 60-65rpm in a big gear is going to have a different experience than someone spinning at 90-100rpm on 150mm cranks in a middling gear.
I simply find that he's trained himself in a manner that the delta between round and oval rings is going to be quite small (not a bad thing at all). Larry has reached the point that a lot of other variables will come into dominance. The number one variable is going to be the MBB providing an inconsistent pedaling angle which is going to have a big affect on the Vectors readings at the 3s range (30s average is probably less impacted). Only placing the bike on a stationary trainer is going to remove that problem. A kickr is probably the better place to do the test. Direct power mode at least will say; this cadence produced this power in this gear. At that point "perceived effort" data is probably very interesting, maybe even more so than HR data. It still not going to be complete because of the variability of the engine (o2 in and out; calories burned etc etc). but that doesn't make it junk data; it's just not high precision. So about the best we can do is Maximal effort for 1 minute, produces X power, Y Speed, Z Cadence for Gear Ratio B. If we had the data for multiple runs we could use comparative reduction (oh lord I'll have to get my diffy-q, and linear algebra books out) to make a graph with error ratios for the non-controled "perceive max effort" Yeah it won't be perfect but it would provide trend lines that could be compared. BUT and big BUT perceived effort makes these very "personal" results. Is it worth all the effort? Sure if it's "fun" and it was a rainy day. The only real way to test this would be to use a motor to turn the chain ring and measure in versus out. I'm sure Rotor has done that; and they don't publish the result.... I wonder why? no no I don't really wonder why.
As someone that only turns out 90rpm steady under perfect conditions riding on the nut, heres what I know subjectively. Going up hill at full effort (See larry I didn't say 110%) I have gears left with the Q-Ring and I don't with a Round ring. Mashing at 60rpm I can get more power out of my wimpy legs for just a bit longer. Those are both good; but the subjective part is the big one; my knees feel better after 80miles every time; and sometimes the subjective gains are all that matter. If you think you are faster wearing purple and it doesn't infact make you slower; then wear purple.
As for the pro's not using them; they complicate shifting smoothness and bike intechagability; and when you are training daily, spinning at high rpm; throwing out massive power; and using whatever form of epo or blood transfusions of the day the qring probably has so little bang for the buck that nobody cares. If everyone road "pain-agua" then maybe they'd use them.
So Larry if you are having fun keep testing; you are still learning things; remember when the HR data was suddenly confusing / suspect when you first learned about power.