Consumer Reports tests speed vs MPG & comes to false conclusion about Honda Insight 2
 

When I saw that Consumer Reports had recently done a comparison of the fuel consumption of various cars over a range of speeds (55, 65 & 75 mph), my first thought was: "good, more data to add to the pile."

Here's their data:

http://forkenswift.com/album/7-cr-orig-mpg.gif

All well and good, right? You could make some strong, logical conclusions from those figures. Right?

Apparently not if you're Consumer Reports!

Here's what they concluded and decided to highlight about the 2nd gen Honda Insight:

That statement is misleading at best, and outright wrong if you examine the numbers in a meaningful way.

If we accept that fuel economy is the relationship between fuel used and distance travelled, then the Insight's absolute increase in fuel consumption caused by increasing speed actually lies closer to the middle of the pack of the vehicles tested.

You can see this if you use litres/100 km (or you could use gallons/100 miles):

http://forkenswift.com/album/7-cr-mpg-lpck-change.gif

So the Insight may have had the biggest numerical MPG drop of the vehicles tested, but that's not the same as the largest change in fuel economy of the fleet.

It's pretty clear that in terms of energy consumption, emissions, and money, the absolute costs of increasing speed (ie. most meaninful in the real world) are largest in the vehicles we would expect (the bigger ones with the largest engines).

This is a glaring example of how MPG can be a misleading/confusing metric for consideration of fuel economy: I can only assume the writer saw the Insight's apparently "massive" 15 mpg drop and it drew her focus.

It is very interesting how information can sometimes be presented, which is factual, but not necessarily correct in the respect of the right question, rather than a question. The data they presented is technically correct, but the way in which it is presented does make it appear that the insight is a "worse" performing vehicle and the way it is presented implies that it will give the user the worst value in mpg, which is completely not true.

Very interesting example of totally correct, misinformation.

I would have said her statement was correct if she'd said the Insight experienced the "largest drop in MPG" rather than "fuel economy".

...they're all "...LIES, damn LIES, and STATISTICS..." , said Humorist Will Rogers.

...you'd figure they'd look at the "rates-of-change" or something, NOT the absolute numbers.

Well, if you ask me the conclusion is not necessarily wrong. The Insight did have the biggest relative percentage drop in fuel economy. This just means that as speed increases its fuel economy goes down at a faster rate than other cars tested.

even at 75 mph the insight is still blowing the competition out of the water!!!
 

I don't like consumer reports anyway because they almost always vote for the ones that are crap.....

Consumer reports already decided to the new Insight is crap in a previous issue. Now they are just trying to reinforce that point of view. You should email them on this.

We could quibble about who had the biggest percentage drop, depending on how you compute it, blah blah blah.

But I did notice that the Insight still returned the best fuel economy of the vehicles in the test cohort, even at 75 mph.

I would also note that I'm pretty sure I could beat the Insight with my 04 VW diesel at 75 mph, and maybe the other speeds as well. Admittedly, that's using hypermiling techniques, while they are following some "normal" protocol.

Finest regards,

troy

You're right: relative to its first value (at 55 mph), the Insight had the steepest change in fuel economy at 75 mph (regardless of whether MPG, L/100 km, or gallons/100 miles is used).

But I still feel that absolute (vs. relative) fuel economy figures are more meaningful/useful to the average person in a comparison like this. The writer has decided that the slope of the graph is more significant than the actual fuel used in the change from 55 to 75:

Is the slope of a graph more "significant" to the average consumer than the actual change in fuel consumption? Hmm...

---

tjts1: they have comments under this Consumer Reports piece. Since I posted this, someone has brought up the same topic. (Not me!)

Commenter says:

You should go fix her wagon.

...the Insight goes "...furthest..." on a gallon is way too simple for unsophisticated readers (duh) to read or understand. What?

http://forkenswift.com/album/7-cr-orig-mpg.gif

...it's simple, really: you accentuate the POSITIVE and they'll accentuate the NEGATIVE. So, which is it, poor representation of the facts or biased writing, or both?

on the upside... people like pictures charts and graphs. anyone that should be allowed on the road should be smart enough to see that the insight gets the best mileage at each given speed vs. the competitors listed.

the few who thoroughly read the article will almost undoubtedly come to the conclusion that the writer is a fool.

Edmunds ran a mileage test and got these results:
http://hiwaay.net/~bzwilson/prius/edmunds_010.jpg

Then Consumer Reports gave us the Honda Insight mph vs MPG for ranges that I'd already been tracking for the Prius NHW11, NHW20 and ZVW30:
http://hiwaay.net/~bzwilson/prius/edmunds_020.jpg
Note that I do not consider Consumer Reports conclusions terribly helpful; not when we have three Prius models to compare to the Insight.

An Australian magazine also ran a head-to-head mileage test:
http://hiwaay.net/~bzwilson/prius/WAtoday_010.jpg

Finally, there are the EPA results from individuals:
http://hiwaay.net/~bzwilson/prius/pri_insight_010.jpg

What I'm seeing is the speed range, 70-75 mph, is key to a higher EPA mileage rating. The ZVW30 and NHW20 Prius do and they have an excellent EPA rating. But the NHW11 does not and it came in pretty low. So too did the Honda Insight.

I am fairly certain that BSFC at high power settings is the single most important aspect of highway performance. This is something the diesels share, a high BSFC at high power settings. Of course the diesels fall down in the city but that is something they will have to address.

Bob Wilson

I can address that, any driver can address that, more efficiently than a hybrid even, with just their brain :)

Oh, wait, U.S. Consumers, nevermind...

...uh, how many KILOMETERS to a MILE? or was that LITERS-PER-FURLONG?


...my (YMMV) simplistic summary:

A) hybrids do better in CITY than on HIGHWAY, while...
B) IC-vehicles do better on HIGHWAYS than in CITY.

...and, predominately IC-vehicles don't "...run out of steam..." at >65 mph as small-engined hybrids do (notice the current, larger engined Prius doesn't follow the same downward trend of its two predecessors).

Why would high throttle settings have any impact on highway FE? When you're at a steady speed on a flat highway, your engine is only generating about 10-15 HP. That says a very low throttle setting to me.

High throttle settings would be useful in getting up to a given speed, so they would be used more frequently in stop-and-go driving. Like in a city with lots of stop signs and traffic lights.

-soD

This graph includes the power curve for my NHW11 and is not far off from the NHW20 and ZVW30:
http://home.hiwaay.net/~bzwilson/pri..._MPG_Rev_B.jpg

You'll notice that:

70 mph -> 25 hp
75 mph -> 30 hp

But that assumes flat land and most roads, even on river bottom, have small dips and rises. These variations push the power requirement up quickly and that is where the inefficiencies come it.

At 10-15 hp, I'll be in the 45-55 mph range, which is great for performance but not what the EPA nor the most reviewers are actually driving. In fact, there are dead stops in the Euro, 'intra-urban' cycle, and though I don't have the EPA profile handy, I thought it had at least one dead stop.

A good theory, it seldom works out in practice. Too much of the energy goes to heating the brake pads.

Bob Wilson

...the same reason that parachutes "work" is why going faster takes more "energy" (HP).

...similarly, air resistance approaches 'solidity' (incompressability), ie: going "faster" requires exponentially more HP as you approach Mach 1.0 at sea level.

...the fact that the FE of four of the cars remained almost linear between 55 and 75 mph means their losses were FAR less than typical.

Looks like the misinformation campaign hit its stride with the nitwit online media.
Consumer Reports tests fuel efficiency vs. speed; Honda Insight biggest loser — Autoblog

Bob - thanks for posting that info-fest. Good stuff.

tjts1 - yes, I've seen the headline & skewed Insight conclusion spreading as well. Here's another random example: Consumer Reports tests fuel efficiency vs. speed; Honda Insight biggest loser - GupShup Forums

.

The comments on the autoblog article are surprisingly intelligent. Seem like most people have a concept of gallons per mile much more so than the editor.

Thanks to friends, I've finally gotten some Jetta (maybe Golf) TDI, mph vs MPG data and updated this graph:
http://hiwaay.net/~bzwilson/prius/edmunds_020.jpg
I'll attach my spreadsheet so folks will have the data but here is how I see it.

MPG at 75 mph is the key data point. Cars that get better than 40 MPG at 75 mph will do well in the EPA, Euro and automotive press. But if their highway mileage is under 40 MPG, they will be reported as 'dogs.' Notice that my NHW11 fits in that category but the reason for my lifetime 52+ MPG is extensive used of 65 mph as my highway speed limit. But the EPA thinks it is a 41 MPG highway car. Knowing where the 'knee in the curve is,' allows an informed driver to trade off: good, fast and cheap.

Bob Wilson

That is still a low throttle setting to my way of thinking. I really don't think I understand what you are saying...

-soD

What happens is the 25-30 hp point is on a steep part of the power required curve. Any slight rise or additional drag requires the engine to run up that curve into higher consumption and even more inefficient power settings:
http://hiwaay.net/~bzwilson/prius/pri_power_eff.jpg

The areas of peak engine efficiency are marked by the small, red dots. You'll notice the 25-30 hp curve fall right under the last of them. Any higher power demands push the engine into inefficient, lower BSFC regions.

So it is a combination of two effects:

• steep power slope - so any local variance in power demand will rapidly go into much higher power regions
• upper limit of high efficiency - anything over 30 hp is going to have lower BSFC than the 25-30 hp region.

I've thought about making a cruise control device that limits the peak HP setting, just using the ICE rpm of say 2,600 rpm AND a maximum speed. The car would slow down going up a hill but it would stay in a fuel efficient mode. But it in an actual design, I would allow excursions up to 3,400 rpm to handle hills. If any more is needed, the accelerator would be used.

Bob Wilson

... and the misinformation continues to spread
 

This time at cars.com:

At Highway Speeds, Mileage Drops More for Fuel-Efficient Models - link to article
As with the AutoBlog piece, a few commenters immediately point out the flaw in the analysis. And then they're countered by people who insist on things like:

...and...

Sheesh.

It is quite funny, but I had a similar experience. Before I got my Lancer I used to drive a Neon and averaged about 200 miles per tank (back then I didn't use to pay too much attention to milage so this is as exact as I can get). One summer however I undertook a road trip to Canada and was in a bit of a rush getting home where I did 80 - 85 mph the whole way home with very few brakes on the 800 mile trip. Surprisingly, I got almost 280 miles out of my last tank on that trip. Considering how fast I was going this surprised me back then, compared to my usual highway speed of 65.

In retrospect I realize that there were a few favorable factors that helped with the big mileage jump: first it was exclusively highway driving, my regular tanks always included about 30-40% city driving and as I now know, without proper driving technique city really kills gas mileage. Second, for about 3/4 of the tank I was following a Jeep Wrangler that seemed to be in as much a hurry as me. I was following behind at a reasonable difference but at that speed, even that probably had an impact on my mileage. Lastly, going from Pittsburgh to Baltimore has a net negative elevation change of more than 1000 feet so that probably wasn't detrimental to my mileage either.

I guess what I'm trying to say is that a lot of people who are not very informed on the subject, may pull generalizations from similar personal experiences, not realizing that they're missing vital parts of the puzzle.

The comments on Consumer Report are getting worse and worse.

I posted a pointer to this chart at the Consumer Reports comments:
http://hiwaay.net/~bzwilson/prius/edmunds_020.jpg

Personally, if this has 'struck a chord' with various vehicle advocates, it will be a good thing. Maybe they'll get busy and start measuring actual vehicle performance.

Bob Wilson

paraphrasing a VERY old quotation: "...you can lead an idjiot to data, but you can't make him/her believe it."

...perceptions are an 'individuals' information filters...they only let 'in' what that individual wants to hear or see.

Gang:

We just can not trust Consumer reports as a sole source to base our car purchases on. I have read their mag off and on through the years and I don't think I even once agreed with them on products that I knew something about. It's a cool idea to have a service that does not supposed to except advertising, but that just opens the doors to kick-backs, just like we have in politics. I am NOT accusing Consumer reports of such practices, but I often scratch my head in wonder as to why and how they conclude their winners from their losers. Sometimes things are good because they are heavy and sturdy, then something is good because it is lightweight and flexible.... I just want to scream when I read their review of products that I already know the details of. I can not think of a single time I agreed with them.

^ Agreed. Totally.

The thing about places like Consumer Report is that they regulate the mass movement of people from inefficient cars to efficient cars (and other things). If all fronts of the media came out and told us that the insight was the best thing since sliced bread then we'd more or less all go out and buy one. That would leave the oil producers in serious trouble. And we can't have that now can we?:rolleyes:
So what they do is appeal to the sheep in society to keep driving what they are and keep driving the way they are. "Don't hypermile, it' dangerous" they'll say. And the sheep will follow.
We are in a minority and i fear will continue to be, due to the power of the media.
My 2 cent....

ollie

Disagree. Totally.

I do not feel the people at that magazine really care about what they call a winner or loser, and I do not feel they care what is good or bad for people or the environment. I have my suspicions as to what REALLY motivates them, but I will not accuse them of what I can not prove.

$$$$$$ (oops, stuck key!) :D

I'm going to stand up for Consumer Reports. They are not perfect but they are much better than nothing. Just because a product they panned works fine for you does not prove they are wrong about their ratings. Not every panned product fails, not every car is totally unreliable or reliable for that matter.

Claiming that their motivation is money is silly considering they don't sell advertising.

They ain't perfect but they are better than the paid infomercials that pass for reviews in many other publications.

I maintain that at best they aren't much better than nothing and at worst they are worse than nothing.

Hey Metro.

Looking at the chart, I must reach the conclusion that at some speed, the Yaris will actually get better mileage than the Insight. (The lines are converging.) Now, granted this will happen at probably 100 mph (if the Yaris can even reach that speed).

...what "bothers" me about Consumers' Report is their touted reputation for being "...accurate & fair..." -- fair (or fare?) depends upon your 'hidden agendas' and whether you're good (or bad) about disguising them...accurate (vs. valid?) is truly debatable given such poor articles as this one (Insight-II).

...and, sadly, too many consumers simply assume "...if CR said it's good, it must be true!" So much, for 'consumer' UN-peer reviewed publications.

...bad science? bad mathematics? bad writing? or bad editing? Regardless of the reason, it's still bad!

In this case the DATA they generated is probably accurate and useful. The headline grabbing conclusion they printed is misleading.

Moral to the story, if you are going to drive 75MPH all the time, buy a Yaris, if you want to get there cheap.

regards
Mech

Well, considering the Insight is still beating the Yaris by 2.5 mpg at 75 Mph I'd say nope... I'd still pick the Insight...