MKBHD's blind smartphone camera test ended with a mid-range phone on top. In the standard photo category the Google Pixel 7a beat every flagship in the field. Marques Brownlee posted the podiums himself: standard went to the Pixel 7a, then the Pixel Fold, then the OnePlus Open, out of what he described as a few million votes.
If you are holding an expensive phone, that reads like an accusation. It is not one. The useful question is narrower: what does a blind vote actually measure, and is that what you thought you were paying for?
What the test is
The format is simple, and the simplicity is the point. Every phone shoots the same scene. You are shown two of those photos side by side, with no idea which camera took either, and you pick the one you prefer. Repeat that a few million times and an Elo rating — the ranking maths borrowed from chess — sorts the field. Three categories: standard, low light and portrait.
Brownlee published the outcome in The Blind Smartphone Camera Test Winners! and summarised the podiums in a post on X, which is where every ranking quoted here comes from. The field was twenty phones according to the outlets that covered the results; his own post lists only the top three per category.
Blind is the honest part, and it is not a small thing. Take the logo away and brand loyalty goes with it, which is more than most camera comparisons manage.
A vote is a preference, not a measurement
Consider what the format does to a photo before anyone judges it. Both images are scaled down to sit side by side on a screen, often a phone screen. Resolution, fine detail, noise texture, how cleanly an edge holds together — those are the first things to disappear under downscaling and compression. What survives at that size is the processing decision: how bright the picture is, how much contrast it carries, what colour the sky came out, what the skin looks like.
So the vote measures something real. It measures which rendering of a scene people prefer at a glance. That is worth knowing. It is not the same claim as which camera is better.
A second structural detail: you have to choose. No pair offers "these are both fine". A difference too small to matter still produces a winner, and a few million near coin flips stack up into a ranking that looks decisive.
Why a cheaper Pixel is built to win this
Google applies much the same look across its phone line. An A-series Pixel and a Pro Pixel produce images that resemble each other far more than the spec sheets suggest, because the processing is the shared part and the sensor is not. At a size where processing is most of what you can see, the cheaper Pixel is showing you nearly everything the expensive one would.
Everything a flagship spends on runs the other way. A bigger sensor, a better lens, stabilisation and a real telephoto pay off at full size, in bad light, at distance, and in motion — conditions that do not fit inside a small side-by-side pair. That is the explanation that fits the result, not something the test isolated: a vote ranks outcomes, it does not attribute causes.
The same results show where the expensive phones still win
Read the other two categories and the picture completes itself. Low light went to the iPhone 15 Pro. Portrait went to the Pixel 8 Pro. Both are flagships, and both categories give hardware more room to show.
Then notice what was never on the ballot. No zoom category. No video category. Telephoto reach and video are where a large share of a flagship camera budget goes, and neither got a single vote. A test cannot rank what it never asks about. That is not a flaw, it is a boundary — and boundaries only cause trouble when a headline crosses them.
This was not a one-off. The previous edition of the same test also ended with that year's mid-range Pixel, the 6a, on top — that one from the coverage of that edition, not from the post above. Two runs is a pattern in the format, not proof about the phones.
What to do with this if you are buying for the camera
If you already own a flagship, nothing here says you chose badly. It says the advantage you paid for lives in the situations this test did not put to a vote, and you already know whether those are your situations: distance, dark rooms, video, cropping in later.
If you are about to spend on a phone "for the camera", the result is more useful than the headline. A mid-range phone with good processing will hand you pictures that look right on a screen, which is where the vast majority of your photos end their lives. Pay the flagship difference for the harder cases, or do not pay it. The one thing not worth doing is paying it because a ranking said flagship, since a ranking is a format before it is a verdict — and that holds when the ranking comes from a lab instead of a crowd, as the gap between Tensor on paper and Tensor inside a phone keeps showing.
The format explains this result better than the result explains the phones, so how the blind test is actually built is worth reading before quoting its ranking at anybody. And the harder question sits underneath both: what "best camera" even means once you accept that a vote and a measurement answer different questions, and that only one of them is the question you have.
Last updated September 22, 2026