This sounds like it could be an issue with the your battery's fuel guage. Lithium Ion batteries have a problem with the meter that detects how much charge is left in your device. If you don't full cycle the phone every 30 days, the meter will get worse and worse.
Full cycling involves:
- Charge the phone to 100%
- Do not charge your phone again until all the battery is depleted (i.e. until the phone dies on it's own).
This is recommended both by Apple:
Be sure to go through at least one charge cycle per month (charging the battery to 100% and then completely running it down).
... and by Battery University:
Batteries with fuel gauge (laptops) should be calibrated by applying a deliberate full discharge once every 30 charges. Running the pack down in the equipment does this. If ignored, the fuel gauge will become increasingly less accurate and in some cases cut off the device prematurely.
Final Conclusion
Given the sources and explanations below. I am officially going to do the following to optimize my battery life:
- Keep my battery as cool as possible.
- Don't worry about whether it's plugged in or not. When it doubt, keep it plugged in so it uses AC power instead of battery, unless getting hot.
It turns out that the two methods I originally posited are largely moot. The only thing that really matters is temperature.
"Temperature was the most significant factor contributing to the cell degradation, with state-of-charge (SOC) and discharge pulse length of secondary importance." (Liaw et al.2)
Furthermore, it turns out that the decay can be accurately mathematically modeled:
(See Ramadass et al.1 for explanation of terms)
However, the dominant model is that of the Arrhenius formula, which generically predicts time-to-failure as a function of temperature.
The figure below shows the capacity at various cycle counts. Just look at the capacity on the x-axis. The top graph is at 25ºC, the bottom at 50ºC.
After 600 cycles, the cooler battery had ~2x the capacity
While I could still find no evidence on the behavior of Mac power circuitry, there was helpful information on the official Dell website. Two items specifically stood out.
Q. When docked or AC adapter is plugged into a wall outlet, am I using my battery charge?
A. No.
http://www.dell.com/content/topics/global.aspx/batteries_sitelet/en/batteries_faq?c=us&l=en&cs=19#faq9
Q. Should I totally discharge, then recharge my Dell laptop battery occasionally to make it last longer?
A. No, discharging and charging does not increase the life of a Lithium Ion technology battery.
http://www.dell.com/content/topics/global.aspx/batteries_sitelet/en/batteries_faq?c=us&l=en&cs=19#faq27
However It is important to note that Apple and Dell charging circuits may be different. Although, given that Dell does this, I assume apple does as well. On this assumption, unless someone can provide sources to claim otherwise, I will assume that the Apple charging circuitry is smart enough to know this.
I encourage anyone to continue exploring this question and challenge my assumptions. Please see the sources below if you're curious for a more detailed explanation.
Sources
1 Ramadass, P., Bala Haran, Ralph White, and Branko Popov. "Mathematical Modeling of the Capacity Fade of Li-ion Cells." Journal of Power Sources 123.2 (2003): 230-40. Web. http://www.che.sc.edu/faculty/popov/drbnp/WebSite/Publications_PDFs/Web33.pdf.
2 Liaw, B., R. Jungst, G. Nagasubramanian, H. Case, and D. Doughty. "Modeling Capacity Fade in Lithium-ion Cells." Journal of Power Sources 140.1 (2005): 157-61. Web. http://electrochem.org/dl/ma/204/pdfs/0253.PDF.
[3] Ning, G. "Capacity Fade Study of Lithium-ion Batteries Cycled at High Discharge Rates." Journal of Power Sources 117.1-2 (2003): 160-69. Web. http://www.che.sc.edu/faculty/popov/drbnp/website/Publications_PDFs/Web38.pdf.
[4] Ramadass, P., Bala Haran, Parthasarathy M. Gomadam, Ralph White, and Branko N. Popov. "Development of First Principles Capacity Fade Model for Li-Ion Cells." Journal of The Electrochemical Society 151.2 (2004): A196. Web. http://www.che.sc.edu/faculty/popov/Publications/Premanand1.pdf
[5] Zhang, D., B. S. Haran, A. Durairajan, R. W. White, Y. Podrazhansky, and B. N. Popov. "Studies on Capacity Fade of Lithium-ion Batteries." Journal of Power Sources 91 (2000): 122-29. Web. http://www.che.sc.edu/faculty/white/2000studiesoncapcaityfadeofzhangharandurairajanwhitepodrazhanshkypopov.pdf.
Best Answer
This is pretty much impossible. Each program doesn't directly use battery power - they all use the CPU, which uses battery power. But every other hardware component uses power as well - the screen, GPU, hard drive/SSD, etc. And of course you're not just running one program at a time, you're typically running several, plus all the background processes for the OS and other services. This is further complicated by how the OS and CPU allocate resources - despite how it appears, everything isn't running simultaneously, the CPU just switches between tasks every quickly, giving the appearance of everything running concurrently. On top of that, modern CPUs can change the amount of power they draw depending on the tasks going on.
So the effect that each program has on battery life depends on a lot of factors, most of which aren't easily measurable. The closest you could get might be tracking the CPU utilization of each program over time, and making an estimate based on the battery drain over that period. However that would be a very rough estimate, and not very accurate for any useful purpose I can think of.