Is a 4.7-star toy good?

It is average. 67% of the well-reviewed toys measured here reach 4.7 stars or better, and the median is exactly 4.7. Star averages compress into a narrow band near the top, so they separate almost nothing. Read the review count first.

How many reviews should a toy have before I trust the rating?

The median toy in this set carries 3,464 reviews and 83% carry at least 1,000. Below roughly 200 reviews an average is still moving, and a handful of early buyers can hold it anywhere. ToyByAge will not list a product under 200 reviews for that reason.

Why does a 4.9-star toy sometimes look worse than a 4.5-star one?

Because the sample decides the meaning. 4.9 stars across twelve reviews is a claim made by twelve households; 4.5 across three thousand has survived thousands of children, birthdays and washing machines. The lower average carries far more information.

ToyByAge data

A 4.7-star toy is average, not exceptional

Answer first

Across 258 well-reviewed US toy listings measured on August 30, 2026, the median star rating is 4.7 and the median review count is 3,464. 67% of them sit at 4.7 stars or above, so a 4.7-star toy is ordinary. The number that separates products is how many people rated it, not the average.

The measurement

Median star rating: 4.7
Median review count: 3,464
At 4.7 stars or better: 67%
With 1,000+ reviews: 83%

258 distinct US listings, each counted once, measured on August 30, 2026.

Why the star average is nearly useless

Ratings on a large retailer do not spread across five stars. They pile up between roughly 4.3 and 4.9, because people who dislike a product mostly return it rather than rate it, and because a listing that drifts below 4 stops selling and disappears. The visible range is therefore about half a star wide, and inside that range the differences are mostly noise.

That is what the 67% figure above means in practice: the badge most shoppers treat as a mark of quality is held by two thirds of the field.

How to read a rating instead

Look at the review count before the stars. Under about 200, the average has not settled and can still move by several tenths.
Compare like with like. A 4.6 with 5,000 reviews beats a 4.9 with 40, every time.
Check the date. A high average on a listing that has not sold recently describes a product that may no longer be the one in the box.
Ignore the difference between 4.7 and 4.8. It is smaller than the error in your own guess about the child.

This is why every ToyByAge list shows the rating and the review count together, with the date each was checked, and why the published floor is 4.4 stars across at least 200 reviews rather than a star target alone. The full rule is in how we choose gifts.

Method, and what this is not

The sample is the ToyByAge catalog on August 30, 2026: 258 distinct Amazon US listings, deduplicated across lists. Every one already clears the 4.4-star and 200-review floor, so this describes the shape of ratings among products that already pass a quality bar — not across all toys sold. Poorly rated products are absent by construction, which is exactly why the averages cluster so tightly.

Cite it as: ToyByAge, “What a Toy’s Star Rating Means”, August 30, 2026, n = 258.

Common questions

Is a 4.7-star toy good?: It is average. 67% of the well-reviewed toys measured here reach 4.7 stars or better, and the median is exactly 4.7. Star averages compress into a narrow band near the top, so they separate almost nothing. Read the review count first.
How many reviews should a toy have before I trust the rating?: The median toy in this set carries 3,464 reviews and 83% carry at least 1,000. Below roughly 200 reviews an average is still moving, and a handful of early buyers can hold it anywhere. ToyByAge will not list a product under 200 reviews for that reason.
Why does a 4.9-star toy sometimes look worse than a 4.5-star one?: Because the sample decides the meaning. 4.9 stars across twelve reviews is a claim made by twelve households; 4.5 across three thousand has survived thousands of children, birthdays and washing machines. The lower average carries far more information.