Rate my back — six zone scores, not one body average

You can have your back rated from a single photo, and BackMaxx is the only app that rates the back specifically instead of the whole body. Prop up your phone, press the shutter, turn around and let the 5-second self-timer fire. An AI vision model reads that one frame and returns a Back Score from 0 to 100, a rank from Newbie to GOAT, a score for each of six back zones, a score for each of eight back muscles, and a training plan built from whichever came back lowest. The photo is analyzed transiently and never stored on a server.

What it means to have your back rated

A back rating is an estimate of trainable muscle development, made from what a camera can see from behind. It is an estimate, not a measurement. Nothing is being weighed, calipered or scanned in three dimensions.

The model judges visual signals. How far the lats flare out from under the armpit. How much ridge sits between the shoulder blades. Whether the upper traps slope into the neck or sit flat. Whether the rear delts round the shoulder off from behind. Whether the erectors show either side of the spine at all. Whether the left half of the frame matches the right.

Those are the same signals a coach reads across a room. The difference is that a number holds still. An opinion you got in a gym doorway cannot be compared to the one you get in eight weeks; a repeatable score in the same light can. That is the whole argument for putting a figure on something as slippery as how developed your back actually is.

How BackMaxx rates a back

One photo goes in. Fifteen numbers and a rank come out: six zone scores, eight muscle scores, one Back Score. Then a plan, ordered by what scored worst. The full scan sequence runs photo, analysis, zone map, plan, and it does not ask you to type anything about yourself first.

The six zones split the back the way it grows rather than the way an anatomy chart splits it. Each one carries its own score, so a single photo can put you in one band for width and a much lower one for thickness.

The movement counts come from the app’s 77-exercise corpus. The first figure is exercises whose primary target is that zone; the second is every exercise that touches it as a primary or secondary target. Upper Back is the widest-served zone in the library at 55, because almost every row and pull collects it on the way past. Rear Delts is the narrowest at 16, because almost nothing collects them by accident. Symmetry has no exercises of its own at all: 8 movements touch it as a secondary target, and every one of those 8 is a single-side movement.

The plan that follows the scan is ordered by those same numbers. Your two weakest zones lead the first session, the movements are pulled from the corpus by zone tag, and the list is filtered down to the equipment you actually own. 46 of the 77 exercises work at home, and 17 of those need no equipment at all.

The Back Score, and the nine ranks

The zone scores roll up into one Back Score from 0 to 100, and that score falls into one of nine named bands. The bands are deliberately uneven. The bottom one is 38 points deep, the top ones are six or seven, because the distance between an untrained back and a trained one is enormous while the distance between Titan and GOAT is a detail contest.

A low first score is not a bad outcome, it is a wide runway. Every rank has its own breakdown of what a back at that level tends to look like and which zone usually moves first.

The eight muscles behind the six zone scores

Zones are what you see. Muscles are what you train. BackMaxx scores both layers, and the map between them is not one to one.

Five of the six zones own at least one muscle. Upper Back owns three, which is why it is the zone most often dragged down by one component you cannot isolate by eye. Symmetry owns none.

That right-hand column is why the muscle layer exists. A low Upper Back score tells you to row more. A low Lower Traps score sitting inside that same Upper Back score tells you rowing is not the answer and Y-raises are. The eight muscle pages each go further than a table can.

What the rating measures, and what it does not

The Back Score rates trainable muscle development. That is the entire scope of it. It is not an attractiveness score, not a health score, and not a statement about anybody’s worth.

Specifically, it does not estimate body fat, does not diagnose anything, does not rank you against other people’s scans, and does not know your bodyweight, your lifts or your training history. It cannot know any of that. It has one frame of one back and nothing else.

What it does is give the same back a comparable read twice. Scores are visual estimates built for training motivation. A 41 is a starting line and a piece of information about which zone to open your next session with. It is not a verdict on you.

Why a back needs its own rating and not a whole-body score

Every AI physique rater currently on the market scores the whole body and hands back one number. That number is an average, and averages are where backs go to hide.

The back is the one region you cannot audit yourself. Chest, arms and abs get inspected in a mirror daily. The back gets seen in badly lit photographs a few times a year, usually by accident. So it is exactly the region where a whole-body average fails you most: a genuinely lagging back travels comfortably inside a decent overall score and nothing ever tells you.

A back is also not one quality. Width and thickness are built by different movements, vertical pulling for the first and horizontal rowing for the second, and one score cannot say which of the two you are short on. Six can. A photo that returns Pro for Lat Width and Starter for Upper Back produces a plan led by rows. Reverse those two numbers and the plan is led by pull-ups. Same total score, opposite training week.

There is a practical consequence as well. A whole-body score lurches around when you diet, because abs and vascularity move fast and drag the average with them. A score built from lat flare, blade separation and trap slope barely reacts to a week of eating differently. What it reacts to is pulling volume, which is the variable you were trying to read in the first place.

What one photo can’t tell you

The limits are worth reading before the score is, because they tell you which parts of the number to trust.

• No depth. A camera flattens the back. Upper Back thickness is inferred from shadow and separation, never measured, so a quarter-turn changes the read.
• No body-fat separation. A lean back with modest muscle can read higher than a bigger back under a softer layer. Dryness looks like development in a photograph.
• No side view. Lat Width is judged from behind. The flare that shows in a front lat spread simply is not in the frame.
• No history. The model cannot tell whether a Starter score is week three of training or year five, and the plan it writes is the same either way.
• No fix for a crooked photo. A rotated torso reads as asymmetry. Symmetry is the one zone that is as much about your feet on the floor as about your muscle.

None of that makes the rating useless. It makes it a repeatable read rather than a measurement, which is what tracking needs anyway. Shoot the same way every time and the limits stay constant, so the change between two scans is real even where the absolute figure is soft.

What happens to your photo

The photo is analyzed and not kept. It goes for analysis, comes back as scores, and is never stored on a server. Progress photos you choose to keep stay on the device, in the app’s own timeline, where the before-and-after comparison happens locally.

There is no account. No email address, no password, no social login, and no ads. Because there is no account, there is also nothing to sign into from a second device, which is the tradeoff, and for photographs of your bare back it is the right one. The privacy policy sets out exactly what is and is not retained.

How to take a back photo that rates accurately

The score is only as repeatable as the photograph. Five things matter, roughly in this order.

1. Prop the phone, never hold it. Mid-back height, lens level, about two paces away. Press the shutter, turn your back to it, drop your arms and let the 5-second timer run. A held phone tilts, and tilt is what makes one lat look bigger than the other.
2. Put the light in front of you, not above. Overhead light pours shadow into the gap between the shoulder blades and flatters Upper Back for free. A window facing the phone lights the whole back evenly, so separation shows up only where it exists.
3. Bare back, arms down, no pose. No shirt, no towel over the shoulder, no shrug, no spread. Relaxed is the standard, because relaxed is the only pose you can reproduce identically in a month.
4. Square your hips and feet to the phone. Rotation is the single largest source of a Symmetry score you do not deserve, in either direction.
5. Same spot, same light, same time of day. Mark the floor. Two scans shot in two different rooms are not comparable, and comparing scans is the entire point. There is more on this in the guide to back progress photos.

How often to re-rate

Weekly. That is the cadence BackMaxx is built around, and it is about the shortest interval at which a change in the score is muscle rather than water, food, pump or lighting.

Rate more often than that and you are rating the photograph. Back muscle does not visibly move in three days; shadow moves in three hours. Rate less often than monthly and you lose the thing a score is for, which is catching a zone that quietly stopped moving while you trained the ones you can see.

The scans that carry the most information are the ones bracketing a program. Each focus program runs three to six weeks, which is long enough for a single zone score to shift.

Run one end to end, re-scan, and read the zone you targeted before you read the Back Score. The total moves slowly, because it is an average of six things and you only pushed one. The zone score is the earlier signal, and it is the one that tells you whether the plan worked.

Find out where your back actually stands

One photo, a Back Score out of 100, six zones and eight muscles ranked — then a plan built on whatever came out lowest.

Download on theApp Store Download on the App Store

Free to download. iPhone, iOS 26 or later.