EPIC Estimation : T Shirt Sizing - (XS, S, M, L, XL, XXL, XXXL).
We can do epic estimation with non linear series. When this epic is broken down to user story and estimation, let's not do math, because it's not necessary that sum of user stories points equal to epic should be equal to epic point.
User Story Estimation : Non Linear Series : Fibonacci sequence ( 1, 2, 3, 5, 8, 13, 21, 34, 59, 93 .. ) OR Rounded Fibonacci sequence (1,2,3,5,8,13,20,40, 60,100). While estimating user story we also need to consider three factors - Complexity, Effort and Unknown ( uncertainty ).
Why Non Linear Scale ? We are good in relative sizing, we can do better detailing of the object which is near to us when compared to the object which is very far.
Linear Scale :
2 Series (0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28,30)
4 Series (0,4,8,12,16,20,24,28,32,36,40,44,48,52,56,60,64,68,72,76,80,84,88,92,96, 100)
Even if you take 4 Series, see how many buckets you have from 20 to 100. If I have to do estimation on linear scale, then I will look for more details of the object even it is big in size.
Non Linear scale estimation is faster.