Too little of a bad thing
There's an old saw that goes "Be careful what you wish for. You might get it." Consider for example the on-resistance, the Rds(on) value, of a power MOSFET where a vendor will say that under some particular operating conditions, the Rdson of some particular part will not be greater than some particular value. This is all well and good.
However, what will happen in your power supply design if the vendor finds a way to make the power MOSFETs differently so that the Rds(on) value is very much lessened from its allowed maximum? Maybe that might not be all well and good.
Consider the following simulation result as an example.
A client asked me to do a worst case analysis of the waveform for the input power line current of his new power supply design. He was especially concerned about a limit for the peak value of input current. I made him quite unhappy because if I made the happy assumption that his power MOSFETs might have very much lower, very much better, Rds(on) values than their datasheet called out as a worst case maximum, that his input pulse current values could rise to unacceptably high numbers.
The MOSFET vendor had a specified limit value for how bad, how high, the Rds(on) value might be, but there was no limit for how good, how low, the Rds(on) value might be.
I lost that client.