Stephen at System sans Setting has posted an idea about PCs losing experience between campaigns. This sounds good; here is a slight variation on the same idea, aimed at D&D 5E (my current system of choice).

What if characters just lost experience at a fixed rate when not actively adenturing? Let’s say an inactive character loses 25% of his XP for each year of “extended downtime” (basically no adventuring nor any related activity). What would this mean for a character retiring just after reaching level 20?

Year XP Level
0 355000 20
1 266250 18
2 199687 16
3 149765 14
4 112323 12
5 84242 10
6 63181 9
7 47385 8
8 35538 8
9 26653 7
10 19989 6
11 14991 6
12 11243 5
13 8432 5
14 6324 4
15 4743 4
16 3557 4
17 2667 3
18 2000 3
19 1500 3
20 1125 3
21 843 2
22 632 2
23 474 2
24 355 2
25 266 1

So our 20th-level veteran adventurer would be level 18 after one year, level 6 after 10 years and back to level 1 after 25 years. The exact figures can vary depending on the starting XP total (several of the XP amounts in this table are quite close to a level threshold, either just below or just above) but the decrease should always be similar.

If you overlook the two years at level 8, the decrease in experience is faster at higher levels, which is consistent with real life. I am an average bass player at best, but I can stop playing for a couple of months with little ill effect, whereas many virtuosos can feel a loss in proficiency if they stop playing for more than a few days.

This is for full downtime; a character who keeps training without actually adventuring should probably lose XP at half the above rate, so 25% every two years. Say our 20th-level fighter retires at 30 years old but keeps training regularly; he would still be level 4 at 60, and only reach level 1 at 80, which feels right to me.

What do you think, would you use this kind of rule? Do you find them better or worse than the ability scores changing with age, as in some older versions of D&D?