I figured I should add to my own discussion.
My current team have been using estimations from previous sprints to gauge what each point value should represent. Literally going through card after card until we come to a unanimous agreement on what a 1, 2, 3 and 8 point story looks like. Keeping a confluence page as a reference to what our baseline stories are.
My team prior had a very different approach. We took a single story and broke it down into the smallest deliverable pieces of work we could manage. We decided what was the easiest task and considered that to be 1 point. We would then estimate based on how a new card compared. A card double the effort would be 2, and a card double the effort of that card would be 3 and so on.
We're a place where coders share, stay up-to-date and grow their careers.
We strive for transparency and don't collect excess data.