Community Brainstorming for the Next-Generation Mileage Database

[Schwa]

On the compare page, perhaps under the hybrid chart and down the page a bit have a selection of the top averaging non-hybrid models (smart etc), and / or equivalent-to-hybrid (in the case of Honda Civic/Accord) so people can compare real world non-hybrid averages to the hybrids at a glance.

[Jason]

I've just got to say right now that we can't go in that direction. The main concentration of this mileage database is going to be hybrids and hybrids only. It simply isn't feasible from a developer's perspective to include other cars and I really don't want to draw focus away from the theme of this site. Sorry

[tcampb01]

I'd like to see the database include regional info (e.g. where were you -- geographically).

I noticed the comments on "hypermiling" in some threads. What caught my eye was a comment about it not being based on the last handful of tanks when the weather happened to be good, but rather being based on the average fuel economy for the life of the vehicle.

I can tell you that my fuel economy in the middle of the winter here in Michigan when the temps haven't exceeding the teens for the duration of the tank of fuel are DRASTICALLY different then the temps in the springtime now that the weather is in the 70s (by as much as 10mpg!!).

Thus comparing someone in sunny Southern California to someone in the frozen midwest is hardly a fair comparison.

Weather data would, of course, be ideal. But most folks probably haven't captured weather data in their logs. My own log book contains date, odometer milage, mileage this tankful, gallons, and the dashboard system's reported average fuel economy (though with the gallons burned and odometer mileage that can still be calculated after the fact). Given the date range is known for that tankful and city/state is known, it's probably possible to use a weather bereau to lookup the average daytime temps when that tank of fuel was used.

This way we know we're comparing cars to other cars driving in similar weather.

I suppose the same argument could be made for terrain (mountains vs. hilly vs. flat) and others have already made the comment that the database should try to capture whether the tank was mostly city vs. highway driving.

Regards,
Tim

[tbaleno]

I think having those variables (weather, location, terrain, etc.) should be in the database. Like someone said before it should let you sort by those fields and do sub sorts. So I can look at HCH in california, in the mountains during peak heat times. I think that would help people drill down to be able to closely compare similar experiences. It will also help people considering a hybrid to know what to expect based on their situtation.

[galaxee]

Originally Posted by howard0944

Be sure to include the gasoline 'grade' is possible. I've noticed higher MPG when using 89-octane over 87.

howard, toyota specifically mentions NOT to use anything over 87. it can screw up the starting system. (indirectly quoted from toyota hybrid tech training material...)

now back to the topic. i would also like to see calculation-method differences. mfd vs manual calculation.

[merlin180]

Jason,

>> they aren't separate. There's one for traffic and one for highway
I can enter/save a tank of gas that says "66% highway, and 66% traffic". Is that valid? - maybe I'm misunderstanding what the fields mean ?

[Jason]

"Highway" is the percentage of your driving on a freeway. "Traffic" is a subjective percentage of your driving where there's traffic that slows your driving speed, etc.

[ericbecky]

I don't think has been mentioned, and I'm not sure how we'd take care of old data, but what about adding a year field? That way we could compare one year to the next.

[russkohnken]

one of the issues that currently is confronted involves the EPA mileage estimates of all cars. Many green-minded individuals would llike the estimates to more accurately reflect reality. However, I would oppose that. For one, you can get the estimates if you drive legally and thoughtfully. For another, the difference between what you get and the estimate should signal to the driver that you could improve your mileage by changing your behavior! Behavior is the single largest factor that could improve gas mileage in this country. So let's not acquiesce to the lousy driving of many by saying that that is what they should do. Let us rather continue to say that you could do better.

[kenny]

Another vote for separate pump and dash MPG fields.
One continuous source of frustration for many has been: there is only one single field for MPG entry.
Some people enter what the number on their dash board, some enter what they calcualte on the pump.
You can "train" people all you want, but people don't do what they should do, they do what they do do.
Then even though they are told to enter the calculated number some may still put in dash number because they know others do.
The database is sort of a competition for some of us.

Having two fields will put this issue to rest for ever.
One field for dash reading.
Another field for pump calculation (or have them put in gallons and miles since last fill up and the software can do the math.)
Things should be designed to take human nature into account.
A car designer could say, "People should drive safely, so seat belts are not needed."
As we have learned good design takes human nature into account.

Also there are a few people who enter a tank or two, then lose interest and vanish.
I say if they don't enter data for 6 months, delete them.
They throw our averages off.
There is one guy named mjaminet with a civic cvt with one tank of 58.2 MPG who hasn't posted in a year.

Also I am getting better MPG now (63.5 for current tank) than I was a year ago when the car was new. (My first tank was 52.2.)
I think the current database averages tanks from ALL Mays together.
I wish it didn't do that.
If it only reflected the most recent data my numbers would be higher.
But then maybe Jason wants data for the full life of the vehicle.