Visually, allowing 1 to 5 stars inflates the way ratings look, because the first star is always present, even for the worst products.
Imagine if it was pie chart or a bar chart: the lowest rating would look like 20%, not 0%.
Or imagine a product where 50% of people gave it the lowest rating (1 star) and the rest gave it the highest rating (5 stars). This would be presented as a 3/5 star item, which visually appears quite a bit better than its real rating (50%).
That's a consequence of the UI pattern - when you use the same widget as indicator and input, you need some way to represent "no vote yet" - plus, it would be hard to click/tap "no stars". It could be done better but alas, minimalism is the name of the game.
Visually, allowing 1 to 5 stars inflates the way ratings look, because the first star is always present, even for the worst products.
Imagine if it was pie chart or a bar chart: the lowest rating would look like 20%, not 0%.
Or imagine a product where 50% of people gave it the lowest rating (1 star) and the rest gave it the highest rating (5 stars). This would be presented as a 3/5 star item, which visually appears quite a bit better than its real rating (50%).