LA County Restaurant Grades: Why 96% Get an A (and How to Tell Them Apart)

Walk down almost any commercial street in Los Angeles County and you will see the same letter over and over: A, A, A, A. Green card, white letter, top corner of the window. After a while your eye stops registering it. That is the practical problem with LA County's restaurant grading system. It was built to inform diners, and it does — but the signal it sends has flattened out, because nearly every kitchen that stays open earns the top letter.

None of that means the system is broken or that the letter is meaningless. It means the letter is a floor, not a ranking. The information you actually want — how close a place came to losing that A, and what the inspector wrote down — lives one layer underneath, in the numeric score and the inspection report. Here is how the scoring works, why the distribution clusters at the top, and how to read past the letter.

How LA County scores a restaurant inspection

Los Angeles County uses a deduction model. Every inspection starts at 100 points. Each violation the inspector documents carries a preassigned point value based on how much public health risk it represents, and that value comes off the 100. Whatever is left at the end is the score, and the score maps to a letter that gets posted in the window:

• A — 90 to 100 points. Described by the county as generally superior food handling practices and facility maintenance.
• B — 80 to 89 points. Generally good practices and maintenance.
• C — 70 to 79 points. Generally acceptable practices and maintenance.
• Score card — 0 to 69 points. No letter at all. The facility posts the raw number instead, so the public sees exactly how low it went.

That last tier matters. LA County deliberately declines to issue a letter below 70, on the reasoning that a "D" or "F" would be read as a category when the honest answer is a specific number. A 68 and a 41 are very different kitchens, and the score card format preserves that difference.

Two more details worth knowing. First, the deduction structure was overhauled in 2016 specifically to make serious violations weigh more heavily, so a modern LA score reflects severity better than an older one did. Second, the placard colors changed. The county decided in 2022 to recolor the cards and rolled the change out gradually as inspectors visited facilities: A cards went from blue to green, B cards went from green to blue, and C cards went from red to yellow. Older cards are still in windows around the county. If you have a memory of "blue means the best," that memory is now out of date — and a blue card today means a B.

The A-inflation problem

Here is the number that reframes everything. Between 2022 and 2024, LA County Public Health issued more than 120,000 grades. Roughly 96 percent were an A. About 3 percent were a B. Fewer than half a percent were a C.

Read that distribution as a diner and the conclusion is unavoidable: seeing an A in a window tells you a restaurant is inside the largest and least selective bucket in the system. It does not distinguish that restaurant from the one next door. A grade that 96 percent of inspections earn is not a ranking, it is a permit-to-operate signal.

This is not evidence of lax enforcement, and it is not a scandal. Public grading systems tend to produce exactly this pattern over time, for a reason that is basically good news: posting a letter in the window changes operator behavior. When a bad grade is visible to every customer, operators fix the fixable things before the inspector arrives, and the whole distribution shifts upward. Public disclosure worked. The side effect is that the top category got crowded.

The practical consequence is that the interesting variance now lives inside the A band, not between the letters. A restaurant that scored 99 and a restaurant that scored 90 post identical cards. Both are A's. But a 90 means ten points of documented violations, and depending on how those points were assigned, that could be one significant food-handling problem or a long list of smaller maintenance issues. The card cannot tell you which. The report can.

So the useful question is never "does this place have an A." It is "what is the number behind the A, and is it drifting."

Closures and repeat low scores

Below the grading layer sits an enforcement layer, and the two are related but not the same thing.

A facility that scores below 70 posts a score card rather than a letter. If it scores below 70 twice within a rolling twelve-month period, it becomes subject to closure. That is the structural consequence baked into the numbers.

Separately, inspectors can close a facility immediately when they find an imminent health hazard — vermin infestation, no hot water, sewage backup, no functioning refrigeration, and similar conditions that cannot be corrected on the spot. These closures do not wait for a score to accumulate. A restaurant can be shut down on a Tuesday, correct the problem, pass a reinspection, and be open with a fresh A by the following week.

That sequence is the single biggest reason the window card is an incomplete picture. The card reflects the most recent graded inspection. It does not carry a closure history, it does not flag whether the last visit was a routine inspection or a reinspection after a problem, and it does not show a pattern of repeat violations for the same issue across multiple visits. All of that is in the record. None of it is on the card.

One more geographic wrinkle: posting rules are not uniform across the county. Most of LA County — unincorporated areas plus the cities that adopted the county ordinance — requires retail food facilities to post a grade. A handful of cities never adopted it, including Avalon, Bradbury, Hidden Hills, La Habra Heights, San Marino, Sierra Madre, and Signal Hill. Pasadena and Long Beach run their own health departments entirely. So "no card in the window" does not automatically mean something is wrong. It may just mean you crossed a city line.

Compare the numbers, not the letters

If you want an actually useful read on a restaurant, work with these four habits.

Find the numeric score. California requires the most recent inspection result to be available on request, and LA County publishes inspection results in a public database. The number is the comparison unit. Two A's are not equivalent.

Read what the deductions were for. Points lost to a cracked floor tile and points lost to improper cold holding of raw protein are not the same risk. The report categorizes violations. Scan for anything involving temperature control, handwashing, cross-contamination, or vermin evidence — those are the categories tied most directly to foodborne illness.

Look at the trend, not the snapshot. A restaurant that has posted 97, 96, 98 across three visits is a different operation from one that posted 98, 91, 90. The second one may still show an A today, but the direction is the story. Direction is only visible when you have more than one inspection in front of you.

Do not compare across jurisdictions. An LA County score and a New York City score are computed in opposite directions — LA deducts from 100, NYC adds penalty points where lower is better. Seattle and King County use a four-tier placard averaged over multiple inspections with no letters at all. A "90" in one system has no relationship to a "90" in another. Comparing them produces nonsense.

That last point is exactly why we built FoodCodeScout. Inspection data is public, but it is scattered across county and city systems that each score differently, publish differently, and update on their own schedule. FoodCodeScout pulls the underlying records — score history, violation detail, closure events — into one place so you can see the number behind the letter, and the pattern behind the number, without learning four county formats first.

The A in the window is a starting point. It tells you the kitchen cleared the bar. What it will never tell you is by how much, or whether the margin is shrinking. That part is public. It just takes one more click.

About the author — Rick Jenkins is the founder of AI Revenue Forge. ARF builds vertical-specific AI virtual receptionists for service businesses in HVAC, dental, medspa, real estate, home health, credit repair, and pawn shops. Headquartered in Charlotte, NC. Part of Jenkins Worldwide Enterprises.