Ok as it would appear that (sadly) Thirsty and the good doctor aren’t posting any more
There are two major factors at play, Condition and Maturation
Maturation takes time; it’s different for every style of beer and varies according to the storage conditions.
The science of maturation is very complex but can be broken down into a couple of rules of thumb that might help, these are generalisations and there will be exceptions. Most notably Hefeweizen, the banana flavour is very unstable and drops off rapidly so they tend to be best young. American style hop bombs, where the main focus is the hops, again they drop off fairly fast and you aren’t really drinking them for their malt or yeast flavours so they tend to peak fairly early.
Rules of thumb: -
Beer matures better in larger volumes. Not faster
Beer matures better colder, but it takes longer
The bigger the beer (higher O.G.) and the darker, the more it benefits from longer maturation.
Fully mature beer can (and will) taste very different if the condition (amount of dissolved CO2) is wrong for the style.
To understand condition you need to understand equilibrium put simply; at any given temperature and pressure there will be an exact amount of dissolved CO2.
Increasing amounts of carbonation will make the beer more acidic, increase the prickle on the tongue (this changes your perception of flavours), lift more aromas out of the beer; reduce the apparent malt flavours and body of the beer.
As a general rule the bigger and darker a beer the lower the desired carbonation, again with exceptions notably some Belgian styles.
If you visit Braukaiser you can find the equation and at the foot of the tables a rough guide to ideal levels of dissolved CO2 for common styles. This can vary from 3 to 9g/L of dissolved CO2 at a serving temperature of 4oC this is a pressure range of roughly 10-225kPa. The idea that one pressure suits all beer styles is to say the least misguided.
From a practical point of view where it gets very interesting is in a domestic fridge there can be (due to thermal layering) a 5oC difference between the top and the bottom of a keg, if you were serving at 70kPa and the bottom of the fridge was at 2oC the top of the keg could be at 7oC the amount of dissolved CO2 could vary by 0.8g/L, more than enough to make a significant difference to the taste of the beer.
The answer is to put a small fan (old computer fan) into the fridge to break up the thermal layering, so that the temperature is the same throughout, which of coarse doesn’t address the question of maturity, and to serve the beer at the appropriate temperature and pressure for the style.
It is important to have a good quality stable regulator, there are lots of cheap ones on the market that I wouldn’t trust my beer with.
What is probably happening is, as the keg gets lower the temperature of the beer is falling and the dissolved CO2 is increasing, and odds on the beer is actually maturing and it is in fact ready to drink. If the last glass tasted best, you probably started drinking the beer too soon.
Mark
For some reason I can’t insert hyperlinks so for the carbonation tables
http://braukaiser.com/wiki/index.php?title=Carbonation_Tables
PS – for those unable/unwilling to embrace metric 1 Vol is about 2g/L of dissolved CO2, 100kPa is roughly 1 Bar or 14PSI. g/L is a better way to look at dissolved CO2.