Why?

i have often wondered this. i know that the average gets harder and harder to change as you pile on the miles, but for the life of me, i can't figure out how to get the average to go up from the initial drive after refilling, even with my best p&g technique

 

I never owned a gen 3, although I have a v.

Most cars lose a great deal of MPG as weight increases as they regen 0% of the energy they have before braking.
The Prius recovers some significant % through regen, 30%? So additional weight is not as great a penalty, as regen increases with weight.

Many of my tanks improve as I drive, my worst MPG is always within the first mile of a new tank. (you are averaging with 0)

 

I think this is because the displayed average is just a "guess", based on a count of injectors firing or some other input. Your driving over the tank likely didn't change at all - we pretty much all develop muscle memory for driving our cars a certain way.

I rarely pay much attention to the dash display. The real average for that tank (and even this assumes filling the tank to the same level on each refuel) is the calculation of miles/gallons.

 

Here’s an example from a Prime from the last 4 days of commuting:



While we owned a 2010 Gen3 for over 9 years, I never paid much attention to the intermittent effects during the trip measurement. But we averaged around 50 mpg over the 200k miles we owned it, so not bad at all.

 

Here is a typical mileage graph for one of my cars - tanks vary quite a bit. I don't get too worried about the single low tank - usually it got filled more and this is reflected in the next tank. This graph is over the life of the car, including winter in Nebraska where there is a big mileage hit, long road trips of 5,000 miles, etc.. But rarely are more than a few tanks in a row the same measured mpg. Lifetime average was 42.0 mpg, despite the variability from tank to tank.

 

I reset both my trip odos every time I fuel up. Then the first two or three miles the cons number is typically in the toilet. Reason being that the station I use is on a four lane road with a speed limit of 45 mph leaving one direction or climbing a substantial hill going the other. The down side of the hill only drops off 2/3 of the distance you have to climb up, so there is not the gain from gliding down it.

Now I do see over the course of the tank a slight increase for the first 100 miles, it levels off for about 50 miles. Then I will lose .2 or .3 for the next 70 miles. That has me showing about 1/2 tank. From then to flashing pip, I typically get a gain of .5 or .6 mpg.

I have watched the fuel sensor PID on my Bluetooth scanner and it doesn’t show any wild swings on the output from the sensor, so it has to be difference in throttle application or some other subconscious action that I’m doing.

 

I don't think it is anything you are doing, I believe it is variability in the estimating algorithm.

The fuel sensor PID is tank level, correct? Typically that is not a major input to the estimate. Fuel injector pulses is an available inout that is easy for the ECU to process.

 

Are you saying the displayed mpg on MFD always goes down after you pass the half tank? That is not the case for me. I have had latter half of the tank showing both better or worse mpg than the initial half all depending on the driving conditions. Yes, as @bisco pointed out, as you drive more on the same tank of gas, your average is over the longer distance so small change in mpg will not be reflected, but I think I have had plenty of cases that I could get the average mpg to go up after the half tank of gas is gone. It is always true that initial mpg after I reset the trip meter will be the easiest time to achieve high mileage. And if I initially get very high inflated mileage, like what happens when I fill at a gas station at the top of the hill and glide downhill first 3-4 miles, then the mpg will be also downhill from the initial high. But by the time I am done half of the tank, ~250 miles, my average mpg is fairly stable. And even in those situations, I have managed to get the average mpg to go up during the use of remaining half of the tank.

If your mpg ALWAYS goes down after half tank mark and NEVEr goes up, that must be the way you drive. Your driving condition is somehow always worse during the last half of the tank, but I have no idea why.