The time has come to post to this thread.
I first noticed the loss of my 12th capacity bar yesterday, though it could have disappeared sometime within the last couple of weeks.
It's a little disappointing because I've only put 1200 miles on the car (odometer 14700,) bought last December, so to me it's nearly new.
I don't have, or have access to, Leafspy or equivalent, so the usual caveats/complaints about not having true charge parameter values apply.
I've been keeping the state of charge around 50-60% calculated by either: GOM reading divided by GOM reading at 100% charge (which I've only done once in order to get a nominal reading at "full" charge) or; by using the 'number of hours to 80% charge' subtracted from 17, divided by 17 and multiplied by 0.8 (formula adapted from viewtopic.php?f=38&t=7079
The average daily change in state of charge is only 17% because the car is driven so little, 10 to 15 miles per day.
I'm trying to limit capacity loss by keeping the SOC as low as practical to control calendar loss and by keeping the daily change in SOC low to control cycling loss.
However, it is a model year 2012 which was put in service November of 2011 so calendar aging really does apply.
Using our local time of day pricing for off peak power ($0.0896/kWh), my average energy cost is $0.0194 per mile. This compares to (is 28% of) our other car (2008 Nissan Altima 2.4L ICE getting 33mpg at $2.30/gal) at $0.07 per mile. Quite acceptable, ignoring eventual battery replacement and assuming the government doesn't claw my savings back to replace, or more than replace the gas/road maintenance taxes I'm not paying.
I used a Killawatt meter initially to gauge actual power draw from the house mains (pretty constant at 1.34kWh per hour) for the first couple of weeks before reading horror stories about overheated outlets. Subsequently I've been using an assumed draw of 1.34kWh per hour to compare my average mi/kWh (3.6) to that reported by the car (4.9). If the car is accurately reporting energy out of the battery, this means my charging efficiency is about 70% (calculates to 73% but I don't believe that last significant digit can be justified, so I round.) The method for keeping the SOC lower than 80% is to take the number of miles to add to the GOM to get to about 50% SOC, divide it by my empirical miles added per hour of charge (about 6) to get the number of hours to charge. This takes less than a minute while shutting down in the garage.
I haz sad.
But I'm going to concentrate on things that aren't there:
Lack of maintenance.
Lack of concern for gas prices or stops at gas stations.
Lack of noise, vibration and harshness.
Lack of oil leaks and the accumulation of dirt and dust on leaked oil.