Every phone on the site's phone cards, from sub-$200 Android phones to foldables and rugged phones, on one absolute scale. Four objective dimensions read the specification sheet of the model and storage configuration we recommend; three editorial dimensions judge the camera, everyday speed and software, and battery life under independent tests.
| Dimension | Kind | Weight | Reads |
|---|---|---|---|
| Promised software updates | objective | 15% | os_update_years (years) |
| Battery capacity | objective | 10% | battery_mah (mAh) |
| Display refresh rate | objective | 5% | display_refresh_rate_hz (Hz) |
| Storage | objective | 5% | storage_gb (GB) |
| Camera | editorial | 25% | editorial rubric |
| Everyday speed and software | editorial | 20% | editorial rubric |
| Battery life under test | editorial | 20% | editorial rubric |
Reads os_update_years (years) from the specifications of the recommended variant.
| years | Score |
|---|---|
| 0 | 0.00 |
| 2 | 2.00 |
| 7 | 10.00 |
Why these anchors: Measured and published. The years of operating-system and security updates the maker commits to, counted from launch — how long the phone stays safe to use. 10.0 at 7 years, the longest commitment a major maker publishes (Google from the Pixel 8, Samsung from the Galaxy S24); 2.0 at 2 years, the old Android norm that several budget brands still promise; 0.0 when the maker commits to nothing. Between them the scale is a straight line, 1.6 points per year of support, because each further year is worth the same to an owner. Measured on the cards: 46 stated values, 7 years 19 times, 2 years 10 times, 3 years 7 times, 6 years 7 times, 8 years twice (which keeps 10.0).
Sources: blog.google (read 2026-09-28): “We’re committing to providing seven years of software support for the Pixel 8 and Pixel 8 Pro, including the latest Android operating system, security updates and ongoing Feature Drops.”; news.samsung.com (read 2026-09-28): “Lastly, in Samsung’s commitment to a longer and reliable product lifecycle, Galaxy S24 series now offers seven generations of OS upgrades and seven years of security updates.”
Reads battery_mah (mAh) from the specifications of the recommended variant.
| mAh | Score |
|---|---|
| 3000 | 2.00 |
| 4400 | 5.00 |
| 5000 | 7.00 |
| 6000 | 8.50 |
| 7500 | 10.00 |
Why these anchors: Measured proposal. Milliamp-hours compare phones fairly because every current phone cell shares the same lithium-ion voltage. 5.0 at 4,400 mAh: nine in ten phones on the cards carry at least this (n = 148; tenth percentile 4,400 mAh). 7.0 at 5,000 mAh, the most common capacity on the cards (35 times). 10.0 at 7,500 mAh, the largest battery in a flagship-size phone on the cards (RedMagic 11 Pro, OPPO Find X9 Pro; OnePlus 15 has 7,300); rugged phones with oversized packs (8,800 mAh and more) keep 10.0. 2.0 at 3,000 mAh, below the smallest compact phones the cards list.
Reads display_refresh_rate_hz (Hz) from the specifications of the recommended variant.
| Hz | Score |
|---|---|
| 60 | 5.00 |
| 90 | 6.50 |
| 120 | 8.00 |
| 144 | 9.00 |
| 165 | 9.50 |
| 185 | 10.00 |
Why these anchors: SPEC §3.3's own example scale, the same one gaming_tablet-v1 uses: a display refresh rate means the same thing on a phone as on a tablet, so it is scored on the same absolute scale. Measured on the cards: 136 stated values from 60 to 185 Hz, 120 Hz 102 times. A small weight, because at 120 Hz almost every phone on the cards already sits at 8.0.
Reads storage_gb (GB) from the specifications of the recommended variant.
| GB | Score |
|---|---|
| 64 | 2.50 |
| 128 | 5.00 |
| 256 | 7.50 |
| 512 | 9.00 |
| 1024 | 10.00 |
Why these anchors: Measured proposal. Most phones take no memory card, so the storage bought is the storage kept, and it differs between variants of one phone — which is why SPEC §3.3 reads it from the variant we recommend. 5.0 at 128 GB, the smallest storage stated on the cards (n = 19 stated values, 128 to 512 GB); 7.5 at 256 GB, the most common (13 of 19); 10.0 at 1 TB, a tier only flagship phones are sold in; 2.5 at 64 GB, the budget phones that fill up within a year of photos and updates.
| Score | Descriptor |
|---|---|
| 10 | Among the best phone cameras available by independent reviewers' consensus, in low light, zoom and video alike. |
| 8 | Excellent photos and video in most conditions, with a real telephoto. |
| 6 | Good photos in most light with a usable ultrawide; limited zoom. |
| 4 | Good daylight photos; weak in low light and no useful zoom. |
| 2 | Soft, noisy photos in anything but daylight, and unreliable focus. |
Why this dimension: The feature buyers compare most, and one no spec-sheet number predicts: sensor sizes and megapixels say little about the photos. Judged from independent reviews of the recommended model, with the evidence stored.
| Score | Descriptor |
|---|---|
| 10 | Flagship performance that stays fast under sustained load, with clean, well-supported software. |
| 8 | Fast in everything including demanding games, with clean software. |
| 6 | Smooth in everyday use; demanding games need lower settings. |
| 4 | Adequate for calls, messaging and browsing, with occasional stutter. |
| 2 | Noticeably slow in everyday use, heavy bloatware or ads in the system software. |
Why this dimension: Chip speed and software quality together decide whether a phone feels fast for years. Judged from independent benchmarks and reviews rather than from the chip's name alone.
| Score | Descriptor |
|---|---|
| 10 | Two full days of typical use in independent tests, among the longest of any phone. |
| 8 | Lasts into a second day of typical use. |
| 6 | Lasts a full day of heavy use. |
| 4 | Lasts a day of typical use with little to spare. |
| 2 | Does not last a day of typical use in independent tests. |
Why this dimension: Capacity is not runtime — an efficient chip and screen go further on the same milliamp-hours, and Apple states no capacity at all — so the runtime a buyer asks about is scored separately, from independent tests, never from the maker's claim.