It's really an 8% difference because of the lower energy density of E10, and 3 is a small number of times to repeat, especially with how inaccurate your measure of fuel consumed per refill is. You could easily overestimate by a gallon range 3 times out of 3 (either randomly, or maybe because shell pumps happen to shut off later). Someone else might underestimate by a gallon 3 times, assume their fuel economy is the same, and not get involved in discussions about it.

But again, even if you did this 1000 times on your car, it would not be sufficient to draw any conclusions about the effect of ethanol as an additive on fuel economy, because you have not eliminated possible systemic bias in measurements, have not used a range of vehicles, and have not controlled for other possible differences in the fuels, or alternatively have not directly measured their energy density.

I pointed out earlier in this thread that this has been studied by NREL.

http://feerc.ornl.gov/pdfs/pub_int_blends_rpt1_updated.pdf [ornl.gov]

They found a decrease in fuel economy of 3.68+/-0.44% at 95% confidence for E10, which is consistent with the ~3.5% decrease in energy density for the fuel.

Adjusting for fuels is relevant. If you get a less than 5% mileage drop using E10, this is to be expected, because there is less energy in the fuel. It's not because your car is using it poorly.

If they are mixing crappier gas with the ethanol because the mix allows them to meet the minimum octane rating, then this could easily explain a lower than expected mileage. If you are using 91 octane E10, your car might not even run on the gas the ethanol is mixed with. OTOH, if if the "100% gas" really is additive free, they might be mixing in isooctane, which would explain a higher than expected mileage. This is different than high-octane fuels, which don't actually contain extra octane.

Ethanol, as an additive, is something that most cars will benefit from, relative to the gas it is mixed with. It increases the octane rating and reduces direct emissions of pollutants. Currently, it replaces additives like MTBE (which in turn are what eventually replaced lead). I'm not sure if ethanol is better environmentally than MTBE, but I'm fairly sure it's better than lead additives. Isooctane might give you better mileage, but it also might increase carbon monoxide emissions relative to other additives.

I agree that there are downsides to using ethanol as an additive; the data just doesn't show that it reduces fuel economy itself.

I also agree that ethanol, as an alternative fuel given America's supply and technology is dumb. And there is certainly some political pressure to use more of it as an additive as a way to promote more use as an alternative fuel, and to subsidize the corn industry.

A tank is a very imprecise measure of fuel. Did you at least record exactly how many gallons each refill (rather than estimating from the tank size), and make sure the fuel gauge was in an identical spot, while stopped on flat ground, before each refill? Otherwise, perhaps you have a regular commute with 30-mile intervals, and just refilled when it was sort of near empty.

If you filled the tank 1/2 gallon higher with E0 (due to a difference in the nozzle), and went 1/2 gallon lower on the E0 (due to commute patterns and/or effects on the fuel gauge), that would explain the difference.

Regardless, even if your results were consistent enough to give statistically significant differences in miles per gallon (after taking into account energy density), that only shows that it was unlikely to be from differences in things like temperature, tire inflation, driving pattern, etc.

So I again refer you back to experimental results using fuels controlled to make sure they are in fact comparable. And again, the difference with your experience could be specific to your car, or a measurement bias, or a difference between the fuels aside from the ethanol content.

That doesn't surprise me.

Absence of damage in one engine is not sufficient to estimate the amount of ethanol. But they would probably get in very serious trouble if they went way over consistently.

It definitely will, unless the electricity generation is 100% coal.

The main difference for manufacture, etc is the battery. Here's a paper on that:

http://dx.doi.org/10.1021/es1029156

It basically says that the share of energy and other environmental costs for battery manufacture is small compared to the savings from using European electricity instead of an ICE. For example, the life cycle greenhouse gas emissions are more than 30% lower. If you substitute Australia's energy mix (basically the most coal-heavy around), you still get 6% lower life-cycle emissions from the electric vehicle.

I have a few concerns about the poll. The biggest one is that the percentages only add up to 75%.

I think people who thought their mileage decreased would be much more likely to answer that poll.

And I think it is unlikely that those who answered the poll know how to measure such things properly. Some might even be comparing their observed mileage to the what was advertised on the sticker.

Some of the comments make it clear that they are comparing mileage years ago with E0 to now with E10, when their car is older. And at least one comment shows that they are using an octane boosters which I believe is not supposed to be done with E10.

So I think it's really hard to draw any conclusions from it.

That's right, assuming that it's actually 10% ethanol rather than "up to" 10% ethanol, and that the other 90% is the same stuff as the 100% gasoline.

It is possible that, for whatever reason, your car did worse with a switch to E10, or better after switching to E0. Could be just your car, and not the model. Or maybe the ethanol dissolved some junk from your fuel system. It is also possible that something else was different about the test. I'm guessing none of the measurements were very precise. It's even possible that pump nozzles were different lengths, or that the different density makes your fuel gauge read empty earlier.

But it's really likely that there are other differences between the two types of gas besides the 10% ethanol.

The plural of anecdote is not data.

It is also likely that few if any of these anecdotes involved comparable fuels and adjustments for the reduced energy density of the E10 fuel.

You mean someone who can get comparable mixes and run controlled tests... Like NREL?

http://feerc.ornl.gov/pdfs/pub_int_blends_rpt1_updated.pdf

They found a decrease in fuel economy of 3.68+/-0.44% at 95% confidence for E10, which is consistent with the ~3.5% decrease in energy density for the fuel.

I would argue that their tests on 16 vehicles are much more reliable than comparing unknown amounts (only counted the number of miles to get near empty) of unknown fuels (one of which might have about 10% ethanol), in unknown driving conditions using one vehicle, even if it is just one study without peer review.

Now, there is certainly evidence that the manufacture of ethanol consumes as much or more fossil fuel than the energy content of the ethanol. But that's the cost (along with the resulting additional emissions) we should be comparing to the tailpipe emissions reductions from Ethanol blends.

They are serious... And don't call me Shirley.

There are a couple of problems with the above analysis. First, the calculations involving random event probabilities are wrong. For example, the probability of getting heads exactly once when you flip a coin twice is 50%, not 100%. Second, lists of possible wait times are averaged without weighting them by their probabilities. For example, the average of 10 minutes 90% of the time, and 4 minutes 10% of the time is 9.4 minutes, not 7 minutes.

If 3 customers arrive over the course of a couple of minutes, there is a 25% chance of an issue with one of them, an 8% chance of an issue with 2 of them, and an issue with all 3 is extremely rare. In any of those cases, the smoothly flowing lines will start to back up until the issues are resolved.

No, if issues are randomly distributed, you have a 59% chance of no problems in your line (0.875^4), a 28% chance of 1 problem (0.41 * 0.875^3), a 10% chance of 2 problems, a 3% chance of 3 problems, and a 0.4% chance of 4 problems. Since these cases each take 4, 8, 12, 16, and 20 minutes, respectively, the average wait is 6.3 minutes. If you simplify and say that on average there are 0.5 problems, you still get 4 minutes plus 50% of a 4 minute delay, which is 6 minutes.

At length 8, you have a 34% chance of no problems, and a 66% chance of at least one problem. It will take 8-40 minutes, and the average is 13.8 minutes, not 20 minutes. In the simple case, you have on average one 4 minute issue, plus 8 minutes of normal wait, for a 12 minute typical wait.

Yes, on average 1.5 issues divided across 3 cashiers is 0.5 delays per cashier, which gives the same 6 minutes as the 3 separate lines.

No, your average is around 12 minutes just like the separate lines, but you are more likely to have a wait closer to 12 minutes, and less likely to have an 8 minute wait, or a 40 minute wait. Even though there are on average 3 issues, sometimes there are more, and sometimes there are less.

So yes, a single line makes you have wait time closer to the average more often, and reduces the likelihood of a very long or very short wait. But it does not reduce the average, nor change the best or worst case. You can't magically make the cashiers process more purchases per minute with a different line ordering.

But the point is, when one line is moving faster, more new customers get in that line. And therefore, you, as an individual customer, are more likely to have a shorter wait. Yet on average, you have an average wait.

Take the 8-person lines. About 34% of the time, one of them processes 8 people in 8 minutes (and 8 new people get in that line). 26% of the time, there is 1 issue and it only processes 4 people (so 4 new people get in that line). And 40% of the time, there are at least 2 issues, and only 2 people get through in 8 minutes. So in 8 minutes, 4.5 new people per cashier got in line, about 60% of them got in the fast line, 22% in the medium one, and 18% in the slow line. This is correct because we are talking about the probability of the last person in line having a certain experience throughout their wait, and combining that with how many people get in that line in each case.

Therefore in this example, you have a 60% chance of being in the fast line, and only an 18% chance of being in the slow line, just by getting in the line that just finished processing a customer.

More people go through the fast line, so you are more likely to be in a fast line than a slow one. You are also more likely to wait longer the times you are in the slow one. And on average, you will wait an average amount of time. Of course if there are many lines chances are there is at least one faster one.

Other than side effects, a single feeder line won't change the average waiting time. But it will make the first people to get in line get to the register first, which matches one way of looking at fairness, and also it optimizes the maximum wait time.

Let's try a simple example. Line A moves at 1 minute/customer, B moves at 2 minutes/customer, and C moves at 3 minutes/customer, then you have a 55% chance of being in line A, the fastest one. This is because in 6 minutes, 11 total customers make it through the line, and 6 of them were in line A. 1/3 of the lines are the fast line, and 1/3 of the people at the register at a given moment went through the fast line, but 55% of people go through the fast one.

And you're in the fastest line exactly 1/3 of the time (clock time, not purchases) if the lines are the same length, or more than 1/3 of the time if people are smart and accurately estimate the total wait time in each. And on average, you wait an average amount of time.

But the fast line processes more people per minute. So you spend a disproportionate number of waits in the fast line, and a disproportionate amount of time each visit to the slow line.

And on average, you spend the same amount of time in each line. And on average, you spend an average amount of time in line.

However, if people are smart and predict which line is moving faster (in a case where the delays are predictable, like a line of people with packed carts and a wad of coupons in their hand), that line will be longer and the others shorter. Then you are still in the fast line more often, but the wait time in each line would typically be more equal, and thus you also spend much more time in fast lines.

So any way you look at it, you are more likely to be in a the fastest line than the slowest (i.e. better than 1 in N chance, where N is the total number of lines). You're just more likely to remember being in the slow one.

But one single feeder line is more fair, even if it isn't more efficient.