Estimating in hours equals implied precision, which equals implied 100% accuracy. The common result of this is a behavior of adding extra time to compensate for unknowns. People don’t want to be held accountable for an estimate when they don’t have all the data needed.
Larger stories with greater complexity take more time to produce an accurate estimate. How much value is there in taking the effort to give precise estimation for work that will not be done for several months.
Points equal implied imprecision as they are relative only to each other as determined by a specific team.
Teams are able to estimate the complexity of their work (we know we can do it but we need to figure it out) and the risk (the amount of stuff needed to get done, and or, we are not sure if we can even do this)
The most commonly used point system is the Fibonacci sequence 1,2,3,5,8,13,21,34,55… This often modified for simplicity to 1,2,3,5,8,13,20,40,60,100…
Why not use a simple 1,2,3,4 or A,B,C,D?
Fibonacci gives us an exponentially increasing series which drives home the bigger gaps as items get larger. The difference between a 1 and a 2 is much more important than the difference between 13 and 14. Time is not wasted trying to determine if it is a 9 or 10 or 11, if it is larger than what the team calls an 8 and smaller than what the team calls a 20 then it is a 13. As the story gets closer to being accepted into an iteration it is further broken down and refined and the accuracy of the estimate improves.
Note: T-Shirt sizes XS,S,M,L,XL…is a good relative estimation method, however, it is more difficult to total work measured this way to produce an understandable velocity for forecasting.
Using Story Points allows the team to estimate the near term work with a high degree of confidence, the deeper you go into the backlog the confidence drops, but the team still has a good idea of the effort involved without spending an inordinate amount of time breaking down a story which may never be done due to changing priorities.