Consider that the characters wake up in the morning feeling rough and beaten - as most of us feel when we wake in the morning. They've slept on the hard ground, possibly under the stars; or perhaps on the unpadded wooden floor in a inn's common room, a more typical night spent in a town than the Howard Johnson version of a D&D inn that is usually supposed. It is all right; it takes a few rounds of stretching, scratching, consuming and walking about to get the kinks out and get ready for a day's adventure.
Suppose we say, however, that as the day moves along, the characters tire. Not very much, not at first - and certainly less if the players remain sentient and in comfort. Everyone gets tired, however, even those who rest. And as they tire, they lose their glorious, inviolable stats somewhat; a little bit at a time, by the hour.
It isn't as though this sort of thing is new; and it is obvious it has proven spectacularly successful in a number of video games. We live, we work, we swing a weapon a few times, we get tired. The premise exists . . . the only question is, how tired?
We can break down the day by hours to get a sense for this. The party rises at seven, gathers their camp and gets out on the road by eight. Presuming a 20-mile (32 km) walk per day over eight hours (with two hours of rest between), by nine they've covered 2.5 miles (4 km). Not to forget, they are loaded with equipment and weapons! This isn't the merry walk or jog down to the downtown core a commuter might take in the morning. This is slogging in armor over poor roads or wilderness - the distance covered might be much less, depending on the terrain, as little as 2 miles a day or 400 meters an hour.
At ten the party rests, largely from hunger and not from being footsore. They set off again at ten-thirty, stopping around noon for a worthy meal. They're off again at one and so they tramp until four. Those three long hours need a break, the last before the light is lost and camp must be set up. So they walk from four-thirty to six, perhaps seven if they want to walk for nine hours that day and really make some time. The sun will go down soon, however, and setting up tents, making a meal, caring for the animals and possibly building a bit of a perimeter is no easy short-time task.
I can see a simple evaluation for this built out of the Fibonacci series. Copying from wikipedia, the series runs:
1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89 . . .
The party sets off for their first two hours and by the end of that, their stats and hit points are reduced by 1%. Note, one hour of work and two hours of work by this system reduce the character's stats by the same amount.
We don't count fractions of hours, so the hour and a half before lunch (3.5 hours total) count as a loss of 2%. Hell, that's nothing. It means a character with 100 hp is reduced to 98. Piffle.
It starts to add up, however. That walk between one and four, three hours, that begins to rack up the numbers. By one-thirty, the character has walked four hours and is down 3%; by two-thirty that's 5% and by three-thirty, 8%. That's enough to reduce a strength of 18 down to 16.5. By this time I would expect the characters to be asking themselves, do we really want to walk another hour?
Four o'clock in the afternoon still counts as 6.5 hours; it only takes another half hour of walking to drop everything 13%. That last walk between the afternoon rest and camping may not seem like much, but quitting at six o'clock still means everyone's stats and hit points have been reduced by 21% across the board.
Think about that. That's enough to render all a character's precious power stats ordinary and transform a few dump stats into dangerous levels. Some characters, such as a paladin, ranger or monk, would actually lose enough stats to no longer meet the class requirements! Thus the paladin might not have the energy to lay on hands, gain a +2 against attacks or cure disease. A druid may lose the ability to cast spells.
Does the party really want to walk another hour? That would drop us all 34%. Ouch. Maybe we could rest?
The resting would matter, wouldn't it? Logically, a longer rest in the middle of the day would reduce these effects, right? I agree. But I feel it would reduce the effects the same way they're gained. In other words, one hour of rest would restore 1%, as would two hours of rest. Meaningful gains from rest would require many consecutive hours of resting . . . for, I would argue, setting off again would compile the total amount of work done in a day. It would not, as some might suggest, return to zero.
I feel I should make that clearer.
Our party pushes on from eight until noon without a break, reducing our energy by 3%. We take an hour and rest, gaining back 1%. Take note - a second hour of rest would not gain another 1%! That first two hour period takes it away and adds it precisely the same. We'd have to rest three full hours to gain back 2% of our loss.
It is three and we've rested for three hours. We're down 1% below normal. We set off and walk an hour. Now, do we lose 1% in that hour walked? No!
We have worked 5 hours, giving us a total loss of 5%. We have rested three hours, giving us a total gain of 2%. 5 minus 2 = 3% below par. Another hour of work would drop us to 6% overall and another hour would drop us to 11%. It is five in the afternoon and once again, despite the rest, it may be time to think about stopping for the day.
As I said, I have no intention of implementing this rule. I still think it is elegant.