Set out cups in rows: 2 cups in row 1, then each row has 3 more than the one before. The counts are 2, 5, 8, 11, … Each count has an address, its position n, and the rule an=3n−1 turns the address into the count: row 10 has 3(10)−1=29 cups. You can also describe the pattern step by step, a1=2 and an+1=an+3, but then reaching row 10 means walking through rows 1 to 9.
Order is part of a sequence. The lists 2, 5, 8 and 8, 5, 2 contain the same numbers but are different sequences, because their first terms differ. That is why a sequence is a function whose inputs are the positions 1, 2, 3, …: plotting (n,an) gives separate dots, not a continuous curve. The long-run question is what the dots do as n grows: an=n1 gives 1, 0.5, 0.33, 0.25, … and converges to 0, while 3n−1 grows without bound and diverges.
A sequence is not a sum. Adding the first four terms of 3n−1 gives 2+5+8+11=26, the total number of cups in four rows. That total belongs to a different sequence, the running totals 2, 7, 15, 26. If a problem asks for the nth term or how the terms behave, work with the sequence; if it asks for a total, it is a series.
Think of it like this. A sequence is like the numbered lockers along a school hallway: each locker number holds exactly one item, the order is fixed, and a rule tells you what is inside locker n without opening the others. Asking what is in locker 10 is a sequence question; asking for the total value of lockers 1 to 10 is a series question.
Core idea. A sequence assigns a term to each position 1,2,3,… by a generating rule; terms are listed, not added.