Open Settings on an iPhone, tap Battery, tap Battery Health, and you get a number: 92%, 87%, 100%. It feels precise. It feels like a fuel gauge for the chemistry inside your phone, a direct readout of how much life the cell has left. It is not. It is a marketing surface — a rounded, smoothed, conservative summary of a much messier reality, and treating it as ground truth is one of the reasons people misjudge when to replace a phone, when to worry, and when their battery is quietly cooking itself.
The claim I want to defend is narrow but firm: the “battery health” percentage on modern phones is a designed abstraction, not a measurement, and the variables that actually determine how long your battery lasts — full charge cycles, depth of discharge, thermal exposure, and time spent at high state-of-charge — are hidden from you on purpose. If you care about longevity, you need to think about those, not the number.
Start with what the percentage actually is. Apple describes Maximum Capacity as a “measure of battery capacity relative to when it was new,” and in the same support document notes that batteries are designed to retain up to 80% of original capacity at 500 complete charge cycles for older models, and 1000 cycles for iPhone 15 and later. That’s the spec. But the number you see isn’t a live coulomb-counted readout — it’s an estimate produced by the battery management system, and Apple has been open that it recalibrates over time. In 2022, Apple acknowledged that some iPhone 14 users saw the number drop unexpectedly and rolled out a recalibration that ran for weeks in the background. That’s not a gauge. That’s a model.
Samsung is worse, or at least more opaque. Until One UI 7, Galaxy phones didn’t show a battery health percentage at all — you had to infer wear from third-party apps like AccuBattery, or dig into service menus. When Samsung finally added a battery health display, it showed categorical labels — Good, Normal, Weak — before eventually surfacing a percentage. Google’s Pixel line only added a proper battery health metric in Android 15, and it too is a summary figure. In every case, the manufacturer is choosing what to show and what to hide.
What’s hidden is the cycle count, and cycle count is what actually matters. A lithium-ion cell degrades primarily through a chemical process at the electrodes that scales with how many times you’ve cycled the cell and how deeply. The U.S. Department of Energy’s explainer on lithium-ion chemistry lays out the basic mechanism: during each cycle, lithium ions shuttle between anode and cathode, and each trip causes small, irreversible losses — solid-electrolyte interphase growth on the anode, cathode particle cracking, electrolyte oxidation. It’s cumulative and it’s roughly linear in cycles, not in time.
Battery University, which is the reference most engineers actually cite, has published test data showing that discharge depth matters enormously. A cell cycled between 75% and 25% state of charge can deliver four to five times as many equivalent full cycles as one cycled from 100% to 0%. Cycling from 100% to 25% roughly halves lifespan compared to 85% to 25%. In other words: two phones with identical “Maximum Capacity: 90%” readings can have wildly different remaining lives, because one has been shallow-cycled overnight on a slow charger and the other has been drained to 5% and fast-charged back to 100% twice a day. The percentage doesn’t distinguish them. The cycle history would.
Apple started showing cycle count on iPhone 15 and later, buried under Settings > General > About. It’s the single most useful number on the phone for judging battery wear, and it’s two menus deeper than the health percentage. That placement is a choice.
Then there’s heat, which is the other variable nobody surfaces. Lithium-ion degradation roughly doubles for every 10°C increase in operating temperature above room temperature, a relationship documented across the academic literature — see this Nature Energy review on calendar and cycle aging. A phone left on a car dashboard at 50°C for a summer afternoon can lose more capacity in one day than months of normal use. Fast charging generates heat too, which is why Apple’s Optimized Battery Charging and Google’s Adaptive Charging both try to slow the last 20% of a top-up. Neither feature tells you how much heat exposure your battery has accumulated. That data exists — the phone’s thermal sensors log it — but it never surfaces.
The pattern here should be familiar. Manufacturers face a design tension: give users too much data and you get panic, warranty claims, and forum threads dissecting every 1% drop. Give them too little and you get accusations of hiding degradation, which is essentially what the 2017 “batterygate” settlement was about — Apple throttling older iPhones without telling users, then paying $113 million to settle state investigations. The compromise everyone landed on is a single friendly percentage, plus, if you dig, a cycle count. It’s a UX decision dressed up as a diagnostic.
None of this is a scandal. It’s just important to understand what you’re looking at. The percentage is a lagging indicator, smoothed to avoid alarming you, and it collapses several independent variables into one number. If your phone reads 88% after two years of charging to 100% every night and sitting in a hot pocket, that’s very different from 88% after two years of charging between 40% and 80% on a slow pad. The remaining life on those two phones is not the same, even though the display says it is.
What should you actually do with this? Three things, and they’re boring. Don’t habitually charge to 100% unless you need the range that day — turn on the optimized-charging feature and let the phone hold at 80% overnight. Don’t run the battery flat; the deep discharges are what age it fastest. And don’t leave the phone in heat, especially while charging, because that’s when the two stressors compound. These aren’t hacks. They’re just the same advice electrochemists have been giving about lithium-ion for two decades, and they matter more than any number on your settings screen.
The broader lesson is that consumer electronics increasingly present modeled estimates as if they were measurements. Battery percentages, storage “other” categories, signal bars, even the way SSDs report free space — these are all summaries of complicated underlying state, produced by firmware that has opinions about what you should worry about. Sometimes those opinions are correct. Sometimes they’re conservative in the manufacturer’s favor rather than yours.
My phone currently reads 91%. It’s been through 340 cycles, spent a summer in Sydney, and gets charged fast almost every day. I have no idea what the real remaining capacity is, and neither does the settings screen. What I know is that the habits, not the number, will decide when it stops holding a useful charge. That’s the thing worth watching.
