I have to disagree strongly with your position that E10 only affects mileage by 1 or 2 %. While that may be the case in your particular application, I can tell you from my personal experience that I have lost as much as 20 % in fuel economy / mileage on my 2000 Acura Integra. I have years of data and could tell you to the date when the "summer blend" started being pumped in Southern California. My mileage would drop literally in one fill up from over 30 MPG down to 27 or so, and the next few fill ups would drop it down to 25 or 26 MPG where it would stay for the duration of the summer blend season. In So. Cal. we were subjected to E 10 for Air Quality Management District reasons for about half the year, generally April through October, about a decade before E 10 was mandated by President Bush. So I have years of data on this, the first time it happened I brought it in to the dealer because I thought it had something terribly wrong with it to cause such a drop in mileage. The techs just laughed at me and after pulling the codes and doing an inspection reported that it was summer blend gasohol and learn to live with it.
For what it is worth, in So. Cal. in my experience 76 gas is some of the cleanest realtive to fuel filter replacement schedules. That being said I've been running Exxon Mobil for the past year and a half or so and I've not noticed any significant performance or filter replacement changes.
Regardless of what the techs say (they rarely have a clue about what they're talking about) that 10% drop you recorded was caused by something else mandated by the air quality nazis. While I and basically everybody else measure 3% or less difference, it boils down to a simple matter of energy content. To lose even 10% on mileage by adding 10% ethanol to gasoline, ethanol would have to have zero energy in it (i.e. no different than adding water). What you experienced is something else mandated by the terrible people in charge of California, or a problem with your specific car not adapting to alcohol in the fuel very well. It is not characteristic of the experience of the vast majority of E10 users.
I agree with Bates, ethanol kills the mileage. Driving my 1994 Escort from CT to MI via upstate NY, I could CONSISTENTLY see 40 MPG across the state of NY (MTBE additive, not ethanol at that time) after filling up there. With CT and MA ethanol gas, would see 33-34 driving the same speeds, roads, styles.
Now with NY with mandatory ethanol blend, I get the same consistent 33-34MPG all the way from CT to MI.
It's a tossup and can be determined what's best for your particular vehicle by trial and error, using a particular brand of gas consecutively over multiple tanks to find out. This dilutes out or flushes out any residual additives you have remaining in your tank from the previous gas company.
In reality, certain vehicles will run better on a particular brand, certain driving styles will get better mileage out of a particular brand. But probably not much.
As documented i this article, even Consumer Reports dropped a formal test because the day to day air temperature and humidity changes accounted for more mpg change than the gasoline brand.
E85 naturally will increase fuel consumption over 85 Regular. Ethanol has less energy density than regular gasoline, so in order to put out the same amount of power it has to burn more fuel. For E85 vs regular gasoline it's not uncommon to burn up to 35% more fuel.
Any Ethanol will affect your fuel mileage no matter how small a percentage. The question is on the smaller ethanol blends, will you notice?
I don't think gas brand makes that much difference in FE/MPG. More important to me is difference in quality, e.g., top tier gas standards that exceed gov't requirements. I just discovered www.toptiergas.com and was surprised to see Mobil absent from the list, a brand I've used for years and one that I assumed was in the top tier. I may stick with Shell and/or start using Chevron more.