Let's think about this a little (long)..... (archive)

The complete automotive resource for buyers, sellers, and owners like you.

Barring the kind of catastrophic "infant failure" that results in throwing a rod after 2 weeks of ownership, you won't know how good your oil is until the car has enough mileage on it to evaluate wear and tear. And you can't evaluate wear and tear without taking the engine apart and checking the susceptible parts (rings, cylinder walls, valve stems / tips / guides, main and rod bearings, crank journals etc etc etc).

And even if you did this, a series of one is hardly a basis for decision-making. Suppose you did evaluate a series of engines, half of which ran on Syntec and half on Mobil 1 (under IDENTICAL conditions and over identical kinds of roads / temperatures etc, of course). Say you used piston-to-cylinder-wall clearance as your test parameter, and you got an average increase of 0.00237" for the Syntec group and 0.00194" for the Mobil 1 group after 100,000 miles. If you wanted to use a simple Student's T-test to decide if the difference was real, you'd need at least 20 engines in each group to show that a 50% difference in your test parameter was significant (depending on the standard deviation of each series). And the smaller the difference you observe between groups, the larger your series have to be to show that the apparent difference is really significant.

Suppose you did this for 10 engines in 10 M coupes, and you only looked at whether an engine "failed" - whatever you define as failure, be that average crank journal wear over 0.010", threw a rod, or whatever. If you used a nonparametric analysis like a chi-square or a Fisher's Exact test (since the total group is too small for a more sophisticated or specific analysis), and your test statistic was whether or not the engine failed (by whatever criterion you judge as failure), you wouldn't find a significant (p=.05) association between oil use and engine failure unless 1 of 5 failed in one group and 4 of 5 failed in the other - and even that scenario just barely achieves significance. If one or two engines failed in one group and 3 or 4 in the other, the odds would be greater than even that there was no association at all between oil brand and failure.

You just don't have any way of knowing whether one oil is "better" than another, assuming no weird effect in your particular car. And there's usually something else going on to cause the weirdness, if you look. I guess if your oil temp is consistently 10 degrees cooler at the track or under other high-stress conditions with one oil than another, I'd go with the cooler one (all other things being equal). If you find that one oil smokes out your exhaust but another doesn't, go with the one that doesn't smoke (AOTBE). If you find a difference in comsumption, go with the one that hangs around your sump the longest(AOTBE). But these things are not likely to happen to most of our cars with any top-line oil.

I ran my formula vee on Pep Boys house-brand synthetic for 10 years. I started tearing the engine down after every other race. That proved a waste of time and money (gaskets, sealants etc). I went to every 5 races, and that was a waste of time (no wear, no problems). So I ended up just freshening the valve grind every 3 or 4, saving a tear-down until 10 races - and I never had an engine failure of any kind. I replaced the bearings only because the engine was apart - they actually looked and measured good enough to go another 5 or more.

You're all driving yourselves crazy over the choice of oil for no good reason. If there's a factory spec, meet it. If there's a warranty issue, comply with it. Beyond that, it probably doesn't matter AT ALL whether you run Mobil 1, Syntec, or an auto-parts-chain-store house brand (all of which are made by one of the major refiners anyway). I don't believe that there's ANY significant difference among the major refiners' products to any given spec. Use the right grade, of course. I agree with everyone on this board who endorses full synthetic over anything else. I use Syntec 5W-50 in my Z3C. But I'd sure take anybody else's synth if there was a sale and it cost me $1 less per quart - hell, I'd buy 5 cases and save myself $30 - that's a few free oil filters!

Questions, comments, or problems, please visit the Roadfly help desk.