Hall of Lame - worst Unique rolls ever found

@nephalim

That sword is pretty rare with those rolls.
I think that then speaking about damage intervals equation should be more complex.
The chance should be found through average damage number and this differs a little.

lets just say about that fire damage mod.
minimum interval is 10 and maximum is also 10.
your rolls are 1 and 7.
possible variations is 100 as u wrote.

But here there it starts to differ-
7 from 100 should mean that there is only 6 lower possible rolls but it isn't true.
there are more low combinations such as 1-6; 1-5; 1-4; 1-3; 1-2; 1-1; plus 2-5; 2-4; 2-3; 2-2; 2-1; plus 3-4; 3-3; and etc.

And same goes for upper rolls. For now its difficult to say the outcome. I'm a bit short on time, will come up for precise result after 5 or 6 hours, have stuff to do.

Ok, im done with those side stuffs.
So here are the chances-

for phys roll- 3%
for fire roll-
7-1; 6-1; 5-1; 4-1; 3-1; 2-1; 1-1;
6-2; 5-2; 4-2; 3-2; 2-2; 1-2;
5-3; 4-3; 3-3; 2-3; 1-3;
4-4; 3-4; 2-4; 1-4;
3-5; 2-5; 1-5;
2-6; 1-6;
1-7.
->28=((8-1)^2+(8-1))/2 28/100=0.28 or 28%
for cold roll- 6+1=7; ((7-1)^2+(7-1))/2=21%
for chaos roll- 1+2=3; ((3-1)^2+(3-1))/2=3%
for lightning roll- 2/21
for attack speed- 2/11

The end result- 0.03*0.28*0.21*0.03*2/21*2/11=9.16*10^(-7) or 0.0000916%

So yeah, its still a clear winner :D

Edit: if we would consider about general item quality, the equation will be different again, for instance: lets say item has 10 mods, all of them are average roll at 50%, so the most average item should have 50% chance no matter of number of mods (and with current rules its not 50% but 0.98%). With general quality, item is allowed to have such values as 90%, 10%, 30%, 70% which average is 50%. well maybe not average, its quite some digging.
But its out of topic rules.

I've made a google doc - gonna put it in the first post. :D
Also, I added a "calculator" (pretty primitive, but does the job). If you have trouble using it, tell me. Gonna write a short "user manual". :D

That Rebuke is just amazing.

I like what Andrius is trying to do, but I didn't quite understand it. What I do know is the equation OP uses is favors items with lots of min and max rolls, but doesn't favor items with many low average rolls. What I mean is that there were some non-legacy Rebukes I saw on the market with lower DPS than mine, but still wouldn't have scored as well.

Here's an example of what I mean:

Rebuke A:
Adds 20-32 Physical
Adds 20-32 Fire
Adds 20-32 Cold
Adds 20-32 Chaos
Adds 1-51 Lightning
11% Increased Attack Speed

DPS = 259.74

2 * 2 * 2 * 2 * 2 * 2 * 2 * 2 * 1 * 2 * 2 = 2^10 = 1024 points


Rebuke B:
Adds 19-31 Physical
Adds 19-31 Fire
Adds 19-31 Cold
Adds 19-31 Chaos
Adds 1-70 Lightning
20% Increased Attack Speed

DPS = 289.38

1 * 1 * 1 * 1 * 1 * 1 * 1 * 1 * 1 * 21 * 11 = 21*11 = 231 points


So the higher dps weapon scores better (lower points = better), even though I'd consider the lower dps one "worse" in actual use. So it depends on what you define "worse" as, since some rolls (usually attack speed) might have more weight than others when determining better or worse.

While this is very interesting point, and it reality judging the roll is more than what I'm doing, it is very hard to make rules (and calculations) to determine "worse" roll. Especially when you are comparing 2 different Unique items.

Also, in some cases, judging the rolls is impossible, because we're entering zone of critical strike chance, which is better if character using the item profits from it.
Calculating weapon's DPS that way becomes almost impossible to valuate, thus I think we should continue with a little primitive and not really "realistic" way of calculations, yet clear one. :P

Unless someone has an idea how to include dps in the calculations, without making the formula overly complicated.

Can't show, because I already rerolled it, but I bought Ventor's Gamble from Catarina, it had +3% quantity, -34% rarity, +1 health, and all resists were below -20%.

It's probably sufficient to not to put weights on mods, since there are armors, rings, belts, unique mods. And every mod is differently used by each character.

Item's chance to have those rolled mods vastly differs from average mod roll chance which can be described as general quality.
The more item has variable mods the more difficult to get a bad rolled item in general.

Its really up to politics of thread creator to decide what rules to take. I can only throw a bone.
lets just take those chances from my previous post-
3% 28% 21% 3% 9.53% 18.18%
chance to roll lower for each one at a same time is 0.0000916% (this one is used in op)
The average roll of this item (which can be called general quality) is (3+28+21+3+9.53+18.18)/6=13.78%
To roll this item with 13.78% general q. is (0.1378)^6=0.00068597% (6 mods) (well actually it should be a bit lower, this one is highest one possible, to get actual chance requires some more work)

The last number differs from first because there is a lot of roll combinations which every combination leads to same general q. and this one rolled isn't the most probable one.

bumptydity bump

Alright, I know that a worse (better?) Ventor's Gamble is already on the list. Still want to share this treasure before I waste some divines on it.



If I counted it right, it's around 0,301. Not record-breaking, but still pretty bad.

After several divines, I finally got it! A beyond-terrible, mind-numbingly bad, one-of-a-kind Ventor's Gamble!

If my calculations are correct, it's approximately 0,192 perfect (I have no idea how to count it using your method - I tried, but my result was 3,4025203480675410579640752275007e-5)



Sorry for necro, but this thread deserves to live on.