Thanks. Glad you like it. I think there are still small problems with data when you, for instance, override the LMPG and then undo it. I'll check things like this tomorrow, but if I did it well (and I believe I did), it should be much easier to comprehend what it's looking for, no?

I like the new graphs and model comparisons! Fun to see.
Here's something you may not have thought about... I entered in data from last year (Ocober, november 2003).
Those data points show up on the graph as if they are going to happen in the future (in 2004).
Not sure how this can be resolved, but thought you'd like to know!

Eric,

Right now it's not set up based on the year it happened, because I never ask for this information. Rather, it's a trend based on the month. Is there a better way you propose I display the data?

Actually, once I thought about it, it could be very appropriate to overlay each year's monthly data. You would then see the trend for all March's together. That may be o.k.

I don't know how people feel about not being able to compare this year against last year, though. Hmmmm.

If it were possible and would not take a ton of work I think I would like to see each year overlayed on the same graph. I'd gladly go back and add in the year to the data. The ideal would be for each year to be plotted with a different colored line.

Easy for me to say what I'd like, but it could be difficult to program.
Eric

Eric,
I COULD program it, but I feel the more changes I make, the more irritated people are getting.

I would say leave it for now, until/unless there is a higher demand to change it.

Look at the main graph. I added sample size. You think this will solve some problems?

Yeah, I just noticed that change. I think it will be helpful for people.

I've changed the main graph's calculations to simply mean the weighted mileages for the respective cars instead of weighting them based on distance. I feel this is more accurate.

Car A driving 100,000 miles at 100mpg + Car A driving 1,000 miles at 50mpg should be 75mpg, not 99.5mpg for that model

That makes more sense. I've made calculations of my own with out taking into mileage, and, as you note in your example, it can make a big difference.