A teacher handing back marked tests works through the stack: for each test, read the name, hand it over. The teacher never has to decide when to stop; the stack decides, by running out. A for loop works the same way. In for name in ['Ana', 'Ben', 'Cy']: print('Hi ' + name), name is 'Ana' on the first pass, 'Ben' on the second and 'Cy' on the third, and then the loop ends because the list has nothing left.
For counting, Python uses range. range(4) produces 0, 1, 2, 3: it starts at 0 and stops before 4, which gives exactly 4 values, and range(1, 6) gives 1 to 5. A running total shows the pattern: total = 0, then for i in range(1, 6): total = total + i. The total goes 1, 3, 6, 10, 15, so the loop adds the numbers 1 to 5 and finishes at 15.
A for loop needs its sequence before it starts. 'Keep rolling a die until you get a six' has no list to walk through, because nobody knows how many rolls it will take; that is a while loop's job. A second trap: removing items from a list while a for loop walks through it shifts the remaining items along, so some are skipped. Loop over a copy, or build a new list instead.
Think of it like this. Dealing cards around a table is a for loop: one card to each player, and the round ends when every player has one, with nobody counting aloud. The players are the collection and dealing a card is the body. The picture also predicts the classic bug: if a player leaves mid-round, as when you remove items while looping, someone gets skipped.
Core idea. For loops are best when you know how many times to repeat or when iterating over a collection. The loop variable takes each value automatically.