Something I find quite interesting towards the end of an iteration is that if there is a choice of two stories to pick up then the project manager will nearly always press for one which can be completed within the remaining time in order to get the points total for that iteration higher.
This might even mean that we pick something which is lower priority because we might not finish a higher priority but larger story in time.
This would mean that the points for that card would not be recognised until the next iteration which would mean that in the current iteration the points total would be lower than expected.
Most of the time this doesn't make a great amount of difference and if it helps take some pressure off and create an impression of 'success' then it seems a reasonable trade off to make.
It does still seem less than ideal to have that type of decision dictated by such a limited metric though.
Dermot pointed me to a blog post by Tim Ross titled 'Are burndowns evil?' where he discusses this in more detail and although I agree with the points Tim makes, it often seems that a product owner ends up with the impression that the project is 'failing' or that we are 'behind' if the velocity 'target' is not being met each iteration.
It seems to take product owners who are new to a more agile approach a bit of time to get used to the idea that achieving the points total isn't the most important thing to focus on and that it's more useful to focus on other things which actually give them value.
I've worked on projects where we've got to the stage where the points total is a side show but it takes quite a bit of time of consistently delivering before we can get that level of trust.