Back to Durability!

So I have a new and promising durability solution, and of course it was an age-old solution.

The fumble.

As I said in the last post, part of the problem with the system as it stood was that it needed to happen every day and that equipment was breaking as soon as the 1% showed up (in cases where the equipment actually lasted longer than a week).

Though I had remarked on potentially using successive rolls before, I didn't really give it much thought until I was looking over the Tao of D&D's use of break rolls on equipment: on a 1 on your d20 attack roll, the weapon has a chance to break based on its quality: 1d4 for crappy weapons versus 1d20 for "mastercrafted" weapons.

I had been using it for a while, but forgot about it during all the time I hadn't actually sat down and played D&D. My DM brain was getting rusty.

So I decided to incorporate the fumble into some simulations - on a 1 in 20, roll a d100 and compare to the break chance - and boy-howdy did it work. I added one special caveat: if you don't break your weapon when you roll a d100, add the d100 roll to your total stress. I suppose this should maybe be changed logically, as you would expect to want the biggest roll possible, so maybe I'll change it to be add 100 minus the roll if the players can manage it (although I might be able to digitize it out of their way).

The results were very positive: I tested 1000 total resistance, 10000 total resistance, and 20000 total resistance.

Resistance: 1,000
Average # of Days Lasted: 76.76
Standard Deviation # of Days Lasted: 44.66 (min was 2, max was 285!)
Average Break Chance when Item Broke: 37%
Standard Deviation of Break Chance when Item Broke: 0.16%

Resistance: 10,000
Average # of Days Lasted: 240.73
Standard Deviation # of Days Lasted: 121.79 (min was 8, max was 782)
Average Break Chance when Item Broke: 14%
Standard Deviation of Break Chance when Item Broke: 0.07%

Resistance: 20,000
Average # of Days Lasted: 336.57
Standard Deviation # of Days Lasted: 170.57 (min was 30, max was 1123)
Average Break Chance when Item Broke: 10%
Standard Deviation of Break Chance when Item Broke: 0.05%

The standard deviations of days are admittedly a little insane: the data seems really spread out. On the other hand, of the items in the 20,000 bracket, over 60% were within one standard deviation of the average (i.e. between 166 and 507). I can live with that, RNG gods.

The one extra caveat I'll need to add is that 1 on a d20 would work plenty well for weapons, but what about other things, like armour, backpacks, books or bananas?

At this stage, I am considering using a 20 on an ability score roll with an associated item as incurring a break roll (e.g. swinging from a rope), but there's no ability score to eat a banana or wear armour. What does one do?

Well considering what causes wear and tear on equipment - stress from fighting, falling or stretching too far - I could call for a d20 roll in those situations, but that starts feeling like DM fiat when I need to personally adjudicate every kind of scenario, and DM fiat leaves a bad taste in my mouth.

I could have players roll for whatever's on their bodies when they fall below 0 HP, but again circumstances should dictate whether or not a roll is necessary.

I could make each attack hit a semi-randomly determined part of the body and have that item roll its resistance to the blow (1d20, then the 1d100 on a 1). But then there are way too many rolls going on for each attack.

At this point, I am going to implement the complex rule for weapons and keep the break roll limited to situations that call for it. Items will naturally still take daily stress while out in the wilderness, so they'll still break if they ever hit their total resistance.

Now time to implement!