2002 Impreza fuel economy worse than normal

DD#1's trusty rusty 2002 Impreza wagon has had a recent drop in gas mileage. She had been getting 22-23 mpg pretty consistently, but it seems like now she's getting more like 18.

• 2.5TS automtatic
• 125,000 miles

We did recently put a set of wheels / tires on it sourced from a similar vintage Forester. They are about 5.5% larger in circumference, so that would put apparent fuel economy down to about 21.5 mpg just based on odometer error, without any difference in actual performance. But the change seems worse than that.

Also, although she doesn't have a long commute, she has switched routes away from the freeway to avoid construction. Keep in mind this part of her commute is maybe 5 miles each way that is now via surface roads with lights. Not a huge difference either.

I gave the car a once-over yesterday. No odor, no apparent leaks, no CEL, no actual or pending codes, all readiness monitors OK. I drove it and it seems pretty normal for a 15-year-old Subaru. It's not missing or hesitating. It's no ball of fire with the taller tires on it, but it's nothing notably bad. I maybe detected a slight hesitation to shift up from 3rd to 4th, but it didn't do that on the return leg, so it may have just been the terrain or driving profile.

At this point my only guess is to throw some O2 sensors on it and see if they help, but that's a shot in the dark. I know these are at least 50k old because we bought the car with about 70k on it and we haven't changed them. They may or may not be original. Again, though, no set or pending codes according to my OBDII scanner.

Any other thoughts? Thanks.

How's the PCV and valve cover vent hoses? Cheap to check and it's not like even more crankcase pressure is going to help MPG.

In reply to ssswitch:

Probably original. I can replace those pretty easily and inexpensively. Thanks.

Don't really want to, but I should probably throw a set of plugs on it, too.

Heavier wheel/tire assemblies plus a greater number of complete stops and accelerations wheel eat into the milage susbsatntially, as well.

In reply to WildScotsRacing:

Yeah... I was just unconvinced that it would be that large a difference. Overall fuel economy is down about 20%, of which maybe 4-5% is directly attributable to odo error (it had slightly oversized tires on it before the switch). Thanks!

Might be worth taking some vacuum measurements at idle, too. Maybe the cat is clogged?

When I used to measure such things, we would easily see +/- 20% swings in fuel economy out of a pool of 20- 30 drivers, all nominally driving the same route. Differences in acceleration, % over the speed limit, etc.

I could easily see a route difference contributing that much if the driver does lots of jackrabbit starts, and the difference between freeway and stop and go is significant.

The key to making comparisons relevant is to limit the number of variables. Have her go back to the freeway for a week and mathematically correct for the mileage issue from the odometer reading and then compare.