This is an exercise I've thought about a few times. Let's say you have the 1st Overall and are about to happily draft Auston Matthews, but then I offer your my team's firsts a year from now, three years from now, and five years from now. Would you take that deal? Let's say this is the Oilers and the Devils talking. The Oilers trade us Matthews, but what are they getting back. Our 1st next year could easily end up being a top 10 pick, but could just as easily end up around 20-22 range. 3 and 5 years from now is almost impossible to predict. What if we were to say instead that it will be our pick this year, next year, and the year after? That's much more attractive because then you have more of an idea of where the picks will be, who you can draft with those picks, and you will reap your rewards sooner.
Most people would probably trade the 1st overall for three guaranteed top 10 picks, but are far less likely to do it for three picks that may or may not be top 10.
I disagree with Jim's analysis that the Stempniak trade is bad because the 2nd is next year, but he's not wrong that picks now are more valuable than picks later. You have more information on where in the 2nd round your pick will be and more information on who will still be available in that area. Yes the draft may be deeper or more shallow, but that is at worst offset by the greater information.
Think about our 2nd round choices recently
2010: Merrill debut November 2013
2011: none
2012: Severson debut October 2014
2013: Santini likely debut next year
2014: Jacbos likely at least 1 year away
2015: Blackwood won't even be on the radar for the big club for another 3-4 years
The general consensus is that these are all better than average prospects for 2nd rounders, and even still outside of Severson it takes them at least 3-4 years to become NHL players. Adding or subtracting a year to that development time is huge, and so a pick being in next year's draft over this year's is certainly a significant factor into value.