Vavloline seems to be plucking favorable single data points from the wide variety of tests needed to meet the overall criteria and then touting them as an performance advantage. This is probably not true. If they were really "better" then they would show fleet test results, something all oil marketers have access to. (While you do have to pass the engine dyno tests, you don't have to conduct fleet tests on each and every oil formulation. But you do have to have field experience showing your formulation would be good in the field - this puts the burden on the marketers so they don't come up with a formulation that beats all the tests, yet stinks in the real world.)
The %Phosphorus retention data shows Valvoline had more of this good material in the oil after a 20 hour test. While it is true that phosphorus is part of the additive that helps protect the engine against wear and that having more in the oil would be a good thing, this test was put in place to measure something else. The OEMs are concerned that phosphorus harms catalysts and that the phosphorus additive is volatile (boils off during high temperatures). The test was put into place to measure the amount of phosphorus additive that boils off and wear measurements are secondary. This is really a catalyst protection evaluation. Valvoline is doing the "if A=B and B=C, then A=C" thing here.
The other data shown is that the Valvoline engine wear after 100 hours is 4X less than Mobil. (Note that they are switching tests here, so this is the B=C part.) The graph I found does not have a scale. One might assume it means that engines with Valvoline will last 4X longer, but I see this as a Valvoline oil had 1 Part Per Million of iron wear metal in the used oil, Mobil had 4 PPM, and to pass the test, the result had to be less than 100 PPM. While I have no doubt the numbers are statistically significant, I cannot believe it is significant to the consumer. Why not? Because all these engine tests are designed in cooperation with the OEMs, the oil marketers, and the additive companies. The fundamental objective is to simulate an engine's lifetime in a short period of, in this case, 100 hours. The variability of normal engine testing to predict such use has always meant that the results are pass/fail only. If Mobil passes and if Valvoline passes, then the entire lubricant industry has agreed (by their participation in the statistically designed testing process) that there is no difference in the field.
You still might ask, so what? If there is even one little difference, isn't that enough to slant your purchase toward Valvoline? Not really. Additives are all surface-active and they compete for metal surface sites. Friction modifiers, anti-wear additives, and detergents all push each other around to find metals. Each one has a preferred temperature at which it can stick to the surface the best. The 100 hour test (it is not named) might be a low temperature wear evaluation, which does not necessarily predict high temperature wear. You might ask Valvoline what the wear results from all the other tests wear, and ask how Mobil did in those, too.
As far as not being full synthetic - recall that a highly refined mineral base oil, called Group III, is able to be called a synthetic. Castrol started this and most marketers use Group III instead of "original" synthetics because they are cheaper and the performance is equivalent. Don't pick on Mobil for this and believe Castrol is being a saint.
The HTHS viscosity is a good indicator of oil film thickness. However, it is now part of the SAE specfications to place an oil into a viscosity grade. To say one oil is better is a stretch, since it is like saying one SAE 30 is a little thicker than another SAE 30, so it must be better. Well, then, just use an SAE 40 for the same justification.
Here's the takeaway - the technical folks found a statistical difference in the myriad of test results, told the marketing folks, convinced the lawyers it's safe to say, and they made a commercial out of it. They did not say engines last longer, did not say that cars in the real world have lower wear, and did not show any fleet tests that evaluate a lifetime of use (over multiple oil changes); they only showed results of highly shortened, compressed, abbreviated tests (one oil change) that are meant to predict service. Is it really better?