In order to reduce the amount of times you have to update your number of torches (etc), and also to introduce some random uncertainty, some people suggested an alternative method. Instead of putting "10" in the number of torches you have left on your character sheet, you could put "1d8". This means that every time you need a new torch, you roll 1d8.
If the result is anything other than a 1, then you don't have to update your character sheet at all. But if you do roll a 1, you change it from 1d8 to 1d6. And if you roll a 1 on the d6 in the future, it goes to 1d4. And once you roll a 1 on the 1d4, you're out of torches.
Again, the point of this is to reduce the amount of times you have to change the number of torches you have on your character sheet. Also, it introduces tension - you're several levels into the dungeon, and you only have 1d4 torches left; will it be enough?
This is all fine. I like this system. But it does have a small problem, I think.
Let's say you have 1d4 torches left, and you buy 2 "units" of torches to go from 1d4 through 1d6 up to 1d8. Each "unit" (read: die) you buy costs the same. But, and here's the (slight) problem: going from 1d10 to 1d12 is more valuable than going from 1d4 to 1d6, because the 1d12 only has a 1 in 12 chance of depleting (which is good), but the 1d6 has a 1 in 6 chance of depleting (which is worse).
In other words, it always makes sense to buy as many torches as you can (if you ignore encumbrance, I guess), because the last "unit" you buy will be more valuable than the first (read: have a smaller chance of being depleted).
So, one way of changing this would be to flip it around: Having 1d4 torches is the MOST amount of torches, and when you roll a 1, you switch "upwards" to 1d6 instead. Then, when you finally get to 1d12 roll a 1 on that, then you're out of torches.
This doesn't solve the problem of each "unit" of torches you buy is valued differently, but at least then the least valuable ones you buy are the ones you buy last, instead of the ones you buy first. This introduces diminishing returns on buying loads and loads of torches, which has some positive effects. But the problem with this approach is that it's less intuitive ("What? Having 1d12 torches means I have LESS than 1d4? That makes no sense!").
So, what do you think? Personally, I feel like I'm overthinking this and that I should just go with 1d4 being the lowest amount which is more intuitive, and the problem with that which I'm describing above is so small as to be insignificant.
Additionally, I want to abstract this further and say that instead of "torches" in your inventory, it should just say "supplies" which includes torches, arrows, and other similar resources. The point of this would be to reduce clutter on your character sheet.
Right, I see your point - this system would be frustrating for people who want to make sure they have enough stuff. But even if you're just buying one "unit", you know you'll always get at least one use out of it - because even if you roll a 1, you still get a torch - but it was the last one, so now you don't have any additional ones. But even if you buy lots of units, you could still run out very quickly with a string of bad rolls. So yes, I agree, that's a problem with this system.
In defence of the stochastic approach, torches don't all last the same amount of time; the stochastic reduction of torches reflects the fact that sometimes your torch is a PoS and burns out after only 15m and sometimes it's brilliant and lasts an hour (or whatever).
Yes, I agree. Uncertainty creates tension. If you know exactly how long your torches will last, then you can theoretically calculate exactly when you have to start heading for the exit to make it in time. Without knowing how long your torches will last, you'll have to make a choice whether to risk delving deeper or heading out. That's an interesting choice, with real consequences.