Without having done any specific tests, my best
measured power figures (ok, adjusted for the apparent reading error) from "just riding around" are
5s: 15.66
1m: 8.11
5m: 5.62
20m: 4.59
60m: 3.98
Although I didn't get my power meter until recently, analyzing the power calculated from some of my rides, I knew that 5m power was my greatest strength and used that to my advantage in my race. I didn't need power in the race, I just needed to know to look for opportunities where I could "hit it" for 5 min.
I already knew my sprint sucked but I was really surprised about the 5 min power.
Yeah, you'll get better at testing, and you'll probably improve your 5s and 1m numbers. It's a bit like bowling or golf, in that there are a lot of things that can go wrong during these two tests. Letting up 1s early is a big one. Also, going back into a 5m test knowing your previous power, you can pace yourself a bit past it, and you might get improvement just from efficiency in pacing (I front-load my 5' tests with a lot of 600W-700W kind of work).
Since we're not lucky enough to be strong all-rounders, it's really important to know where the profile peaks are so you can leverage those durations. Another interesting approach is to interpolate the profiles for all durations, then test on both sides of your peak. Doing this, I think I'm a bit stronger relative to others at :50 than I would be at 1:10. So, if you took a look at your 4' and 6' numbers, then maybe 4:30 or 5:30, you might be able to bias your race strategy a bit.