The Caddy you mention has GM's 3.6 liter DI engine, correct? That engine is doing horrible things to all oils. I don't believe even RLI's 5w40 has been able to withstand shearing in it, and that was the one oil that was holding up to the RS4's engine (albeit for moderate run lengths)
All DFI engines oil suffers from problems related to excessive fuel intrusion, including Porsche; but some oils (those with ACEA A3/B4 and A5/B5 ratings, for example) do considerably better than other products in these applications. Unfortunately, far too many small "blenders" (including some you have mentioned) make marketing claims designed to indicate their products meet those specification, when in actual fact they do not; often holding no ACEA ratings at all as they do not participate in the ACEA testing protocols. That nonsense makes it difficult for the average consumer to understand what is real from what is marketing hype..... And, just for a reference point, we solved the Caddy's issues solely by switching it to a different brand and weight of oil. It now goes 6-7K miles between changes with no problems.
As far as TBN and 50% of baseline,that's bunk as far as I know. Almost all oils will fall down by 50% within 5k miles. Amsoil SSO, which has a baseline TBN of 12, doesn't stay at TBN 6 after 5-6k miles....despite being an oil rated for 15k-25k+ miles (by amsoil). Take my comments for whatever they are worth to you, but as far as I know TBN is fine until it starts to:
1. Reach 2.0-2.5 or below
2. Be very low in relation to TAN in a traditional oil (that doesn't apply to oils like RLI, which start with a high TAN from their basestock chemistry)
Cannot agree with you there; far too much data (and not just ours) says otherwise. One miss conception is that the use of synthetics automatically allows huge increases in mileage between changes, which is not necessarily the case. A more interesting comparison is the delta in TBN (or TAN) in relation to the change in viscosity, a strong indication of how the additives package is holding up to heat and shear. TBN values do a pretty good job of that. An oil that starts as a 0W-40 looks more like a 0W-20 (or worse) about the same time the TBN retention value drops below 50%. For a lot of highly touted "full synthetic" oils, that is no where near 6,000 miles. That alone is why a lot of Porsche engine builders are telling their clients to change the oil much more frequently; often in the 3,500 to 5,000 mile ranges. The "delta" values are much more important than you give them credit for.
I understand proprietary datasets - makes alot of sense when developing something privately or for marketing purposes. Doesn't make alot of sense to me when using the data to advise the public. I'm not guessing which you are using yours for, I'm simply making a general observation.
In the repair shop industry, customers pay you for what you know, as well as what you can do for them. Quite often the knowledge portion is hard to come by, as well as
very expensive to obtain. Just about any decent tech in a shop knows "how to beat the flat rate book" without any sacrifice in the quality of their work; that is how they earn their living. But you would be hard pressed to find a book or website that tells you how they actually do it; that data is their "bread and butter". Years ago, it became very obvious that a lot of what the oil manufacturers (or blenders) were saying was not panning out in the real world. So, to quote an old Russian maxim, specialty shops took the position of "trust, but verify", and started testing programs on their own, and not just on engine oils. Quite often, the results could only be described as "eye opening". And that information obviously has considerable commercial value; if someone asks you for your recommendation on oil change intervals on weight "A" vs. weight "B" of brand "X" oil, you can give them an answer based upon data rather than intuition, knowing that your answer will be a safe one and not come back to haunt you. But that does not mean that I (or anyone else in the trade) intend to post that data on some website; while I do not have any issues answering a question (with some supporting data) on a website, at the end of the day, I do like to eat well.
Regarding shear and harm to engine: Toyota under-spec'd an oil and it did harm to the engine. That doesn't have a direct correlation to how a good oil will perform post-shear. BMW M's "famous" castrol 10w60 routinely shears down to a 30 grade in normal use, but it still protects those engines just fine. Other oils also shear routinely and protect fine. I'm not saying it's reassuring; I am observing that shear does not = enhanced engine wear.
You do realize that BMW went to the 10W-60 only after a significant number of their M3 engines prematurely croaked on lighter weight oils? The 10W-60 was "band-aid" to keep them from having to warranty even more engines than they already had done for crankshaft related failures due in some measure to their choice of a lighter oil to try and get CAFE rates up, which is exactly what Toyota did, with similar results. The only apparent differences were that more M3 owners ********************ed about it on websites, and threatened lawsuits, than Toyota pickup owners did; and that the 10W-60 oil took a lot longer to shear down to the "danger zone" weight than their original weight recommendation oil did. But even the 10W-60 can get there given enough time.
If someone had a massive data set of UOAs on a particular engine, they could determine if the incidence of shear corresponded to a linear/exponential increase in particular wear metals. But it'd need to be substantial enough to be worthwhile. My .02
They could, but to be completely meaningful, it would need to include more than just the UoA’s detected metals levels, you would need to know how those values relate to actual internal engine component conditions; which would require the tear down, examination, and careful weighing of critical components within the engines, and assuming that the OEM does not make significant component changes within given model years. And, like most OEM's, Porsche uses multiple vendors for most components, which would also need to be factored in, along with developing another dataset of where the critical wear materials "inflection points" are (the level of wear requiring further repair or replacement actions). This requires considerable resources, and is probably why Porsche requires their dealer network to return failed engines to Germany for evaluation. Unfortunately, such a program is well outside the resources of all but the most sophisticated specialty engine building shops; and definitely outside the resources of my business.