It all started when I re-read the Mileage Database 2.0 page's summary display:Should the GH mileage databases include the "starting distance / starting MPG" figures when calculating one's car lifetime MPG average, or should it stick to providing an overall average based solely on the individually recorded tank readings? Or should it let you choose to include any starting MPG when it's really available and accurate (and Jason can re-code it that way)?

The reason for my poll is simple: not everyone may have kept up with their "car lifetime average" prior to using the GH database and therefore may make a [sometimes pretty off] guesstimate, leading to skewed results… and the larger the "starting distance", the bigger the potential skew.



Example: I'm assuming this discrepancy happens quite often (although probably not as extreme as my example), especially with those entries where the "starting distance" is 80% of one's total miles recorded to date… as in 2-3 tanks (or even 0 tanks!), yet already at >10K miles traveled.

As a matter of fact, it even skews the listing of averages when sorted "by distance (farthest)".... at least in the TCH database, where all entries are still relatively quite new. The averages for "farthest miles traveled" are prone to interpretation, as some total miles recorded did not truly start at "0".

 

I have noticed folks who come in, enter one huge mileage and then we never hear from them again. IMHO, the records that should count are those from vehicles whose entries are currently being updated. The rest should go to a 'legacy' or 'historical' vehicle summary.

As for the 'first tank,' it often comes in looking pretty poor because: (1) owner didn't get a chance to baseline processes, and (2) owner didn't have a clue about the vehicle performance characteristics.

I've been around long enough that taking out my first tank would have no effect. But at one time, it delayed my 'star' by three weeks.

BTW, I think a hypermiler star should come from multiple tanks, not just one. It should be earned and evidence of earning it should be evident in the database.

Bob Wilson

 

So what constitutes a hypermiler?

I know this is what the database states: "Hypermiler! The driver of this car has achieved "hypermiler" status. To qualify, a car must have higher fuel economy than EPA estimates over more than 3,000 miles. The EPA city/highway figures are weighted for each car. If no percentages are specified, 55/45% is default."

For a TCH I assume that means 3000 miles over 39.1 MPG, but I have seen some people that have a star where they don't have 3000 miles worth of tanks over 39.1. (ntguru has an average of 39.0 MPG, and has no tank data inputed). I also notice that there are some people that had stars who no loner have them. These people had 2000 miles worth of tanks above 39.1, but had stars at one point in time.

Right now, I am one tank away from star status, and I am working dilligently to get my status, but my current tank is at a display of 38.3 MPG which translates to around 36.5-37 MPG. So I don't think I will be able to get it on this tank. Much more of these terrible shifting winds, and I don't think I will get it for a while (head winds on the way to work, then wind shifts and a head wind on the way home, I need a new job going the other direction).

Back to the original topic, I personally think only actual tank data that is inputed should count, and I don't think the display should even be available for a data input point. I know my display can be optimistic, so the only true measure of how much fuel you are burning is how much you are actually putting into the car. I don't have a problem if you wait to input all tank data every 6 months, as long as it is factual data and not I think I put in 12 gallons.

 

I agree with all you said. The mileage should only be based on calculated not displayed values. When I look at the database and am trying to figure something out like the effect of temperature on mileage, I discard data from those who use only the display. My displayed value has always been high.

 

My original objective for this poll/thread was to discuss whether mpg averages being computed/displayed were as accurate as they could be (realizing that the db has many other "accuracy" impediments, one of which is the legitimacy of the entries themselves)... but since the topic has expanded into this potential "thorny" area, let me throw in my $0.02 as well:Just some random thoughts, that's all. In the end, however, my main question still remains outstanding: do "starting miles / stand-alone mpg" figures have any place in the GH databases?

 

There are alot of flaws but I don't know that there is a solution. 1 tank entries should probably be excluded. As should inputs from MPG displays. I know my display is at least 2mpg higher than my actually economy figures. I think the only way to get an accurate measure is to allow people to input ONLY 1 tank at a time and ONLY allow them to enter the information in terms of gallons of fuel and miles travelled. Of course, it will greatly reduce the amount of information available but it will increase its accuracy.

 

When I first found this site, I think I had about 4,000 miles on my FEH and had not recorded my mileage other than in my head or on gas receipt I've lost. Now I have +31,000 miles and only a possible 3 or 4 tanks below EPA during my first summer with the A/C running. Everything has gotten better with all the tricks and techniques I've learned, and my weather is much better than most recording their mileage here.

There were times I thought about starting to log my mileage, but feared I wouldn't keep up on it or starting it at an unfair time during the winter for me in South Florida.

Any recommendations? Is there any point in starting now?

GaryG

 

I started on what has become a huge spreadsheet after about 2500 miles. I keep track of every tank and every mile and my wife thinks I'm loony.

I like to keep track of how I'm doing overall and I compare it to how I'd be doing in the V6 we almost bought.....right now I'm saving about $55/mo in gas and will hit my break-even point in Feb 2008. I just like knowing.

I entered my data from a starting mileage of around 2600 miles and have since added about 13 tanks over the next 5000 miles. I think any accurate addition to the DB is valuable to the community. I also think tracking it can help you keep an eye on your own performance and how the weather affects your MPG. It will be nice to do a year to year comparison to see how I am doing.

 

"optional" is not quite the way i would like to see it. i would like an MPG datapoint for previous ownership (pre-purchase MPG?), and then "MY" lifetime MPG, possibly followed by the car's entire lifetime. for instance: whan i bought my insight it was showing 60.1 LMPG. since i have owned it i have averaged 67 MPG, yielding a total LMPG of 62. it would be nice to see all three values. my .02!

also, i think 67MPG would make me a hypermiler.....

 

Can a moderator go back and question those who post on the mileage to make the data more meaningful?

For example, some have reported that they drive at +50% of the speed limit. I'm sure that based on the mileage they achieve that they are not going 90 to 100 miles per hour.

There are other descrepancies as well that could be ironed out so that we who study the info to learn would have more reliabel information.

It would also be useful to add if the terrain is flat or hilly since this has impact on the mileage achieved.