Battery Mentor turns real usage into honest rankings. Here is exactly where the numbers come from and how we keep them anonymous.
Every number on Battery Mentor is aggregated from real devices running Battery Guru, our companion Android app used by over a million people. Battery life, screen-on time, charge time and temperature always come from real devices; we never use lab benchmarks or manufacturer claims for them. Spec sheets are used only for hardware details such as battery capacity.
Reports are submitted anonymously and only after a full charge or discharge session, so each data point reflects genuine everyday use.
We store a compact distribution per hardware variant, not individual sessions. Each session is projected to a full 0-100% figure as it arrives, counted into that distribution, and then discarded. One reading is one measured charge or discharge session: we count sessions rather than uploads, because one upload often carries both.
Each hardware variant publishes the medianof its sessions - the typical result, not the average - and prefers sessions from healthy batteries once enough of them are established. A median is not dragged upward by a phone left untouched in a drawer for three days. A phone's headline figure then combines its variants' medians, weighted by how many discharge readings each one has.
Rankings and comparisons only include devices with at least 100 measured sessions of the kind being ranked, and figures inside a plausible range (battery life between 3 and 96 hours). Screen-on time can never exceed battery life.
No accounts, no names, no emails, no location, no advertising IDs. We do not store per-session histories tied to a person, and raw submissions are discarded after they fold into the averages.
The result is a dataset that is useful in aggregate but cannot be traced back to any individual.
Measured stats are encrypted in transit and at rest. The sensitive metrics are sealed in an encrypted blob and only decrypted to render a page, never exposed in bulk.
For the full legal detail, read our privacy policy.
Each figure is a median of measured sessions. With few sessions, one odd phone or one idle weekend can move it a lot.
| Discharge readings behind the figure | Devices | Low (5th percentile) | Typical | High (95th percentile) |
|---|---|---|---|---|
| Under 10 | 602 | 3h 30min | 17h 30min | 69h 0min |
| 10 to 29 | 759 | 4h 30min | 16h 30min | 57h 30min |
| 30 to 99 | 1,102 | 5h 30min | 20h 30min | 64h 30min |
| 100 to 299 | 1,030 | 6h 0min | 17h 0min | 43h 30min |
| 300 or more | 1,778 | 7h 30min | 15h 30min | 30h 0min |
The same tiers with battery size held fixed:
| Discharge readings | 4,500 to 5,499 mAh phones | Share of figures above 48 hours |
|---|---|---|
| Under 10 | 193 | 7.3% |
| 10 to 29 | 221 | 5.4% |
| 30 to 99 | 350 | 7.4% |
| 100 to 299 | 375 | 0.8% |
| 300 or more | 917 | 0.0% |
Among 4,500 to 5,499 mAh phones, 7.4% of figures backed by 30 to 99 readings are above 48 hours, against 0.8% at 100 to 299 and 0.0% at 300 or more. That is why rankings and comparisons wait for 100 readings.
Computed from 5,271 devices on 2026-10-04
Do bigger batteries last longer in everyday use? Here is what phones with at least 100 discharge readings show.
| Battery | Phones | Typical battery life | Typical screen-on time | Middle half of screen-on time |
|---|---|---|---|---|
| Under 3,500 mAh | 199 | 12h 0min | 5h 0min | 4h 0min to 6h 0min |
| 3,500 to 4,499 mAh | 439 | 12h 30min | 6h 0min | 5h 30min to 7h 0min |
| 4,500 to 5,499 mAh | 1,280 | 14h 0min | 7h 30min | 6h 30min to 9h 0min |
| 5,500 to 6,499 mAh | 259 | 18h 30min | 9h 30min | 8h 30min to 11h 0min |
| 6,500 to 7,499 mAh | 145 | 24h 0min | 11h 30min | 10h 30min to 13h 0min |
| 7,500 mAh and over | 108 | 29h 30min | 14h 30min | 12h 30min to 16h 30min |
7,500 mAh and over: only 6% of this band have a published capacity; it likely includes tablets we could not identify.
From 4,500 mAh up, each larger band shows more typical screen-on time: 7h 30min, 9h 30min, then 11h 30min.
Below 4,500 mAh, typical battery life barely differs between bands (12h 0min and 12h 30min).
These are medians across groups of phones, not a promise for any single phone. Larger batteries often come with other differences (newer Android versions, different chips, different owners), so this shows what the data looks like, not what battery size alone causes. Capacity comes from published spec sheets where we have one; otherwise from what devices report, with impossible values removed.
Computed from 2,430 devices on 2026-10-04
Tablets and phones with similar batteries behave differently, so mixing them would push tablets up battery rankings and skew comparisons.
| Group (6,500 to 9,000 mAh) | Devices | Typical battery life | Typical screen-on time | Typical peak charging temperature (devices) |
|---|---|---|---|---|
| Phones | 189 | 24h 0min | 12h 0min | 39.6°C (188) |
| Tablets | 137 | 27h 0min | 9h 0min | 34.6°C (122) |
In the same battery range, tablets peak about 5.0 °C cooler while charging, and give less screen-on time for their battery size.
We identify tablets by name (Tab, Pad), model code, a screen of 7.5 inches or more, or a tablet-shaped screen when the size is unknown. Game handhelds and Chromebooks are left out of phone and tablet rankings. Charging temperature also depends on the charger used and where a device is used, so treat the temperature gap as a pattern, not a rule.
Computed from 326 devices on 2026-10-04
Want to add your device to the dataset? Install Battery Guru and your anonymized usage helps everyone make better decisions.
Get Battery Guru