Why Witch loot drops suffer so badly - the stats:

I recently created a thread which proposed a debate about a more class-biased loot drop system. This mainly fell on deaf ears and suffered greatly from people 'yelling' RNG IS RNG, even if some weren't yelling :). A few people understood the issue and a few agreed that the system would benefit from a small makeover.

I talked about stats a lot in that thread, but was trolled for not providing 'exact' stats. Well, I've put together some 'exact' stats which I believe to be representative of my experience as a whole since Open Beta began.

The premise of the mechanic being that RNG is not actually RNG at the moment, but rather something quite different. So here goes:

(Please bear in mind that my sample size for exactitude is these four screens, but my reason for doing this sample in the first place is that the following stats were my experience throughout my progress so far and actually represent 1000s upon 1000s of undocumented loot drops)

I took a Level 28/29 Witch carefully through four Monster Level 21 screens - Broken Bridge, Church Dungeon Upper Floor, Western Forest, Broken Bridge, killing everything and opening everything. I have 42% IIQ to adequately counter any over-levelling penalties. My IIR was 24% to encourage improved-item-version sampling, in that, would increased finds of one particular item increase my chances of finding Superior, Blue, yellow examples of the same item.

I found a total of 433 items in the four maps.

That's an average of 108 items per map.

And this is how RNG works. the exact drops per map were 108,126,109,90. So two were bang on statistical average and two had a variance of almost equal proportion above and below the average. The RNG did not provide me with 1 item in one map and 215 items in the second map. That is not how RNG functions. RNG functions on a plain of probability/chance within the confines of the rules/mechanics which govern them and this does NOT produce 'any result possible', even less so when averaged over many samples.

Now, of those 433 items, 50 were Currency items and 2 were Skill Gems.

The currency system is NOT pure RNG. The currency system is 'weighted' RNG whereby the lower the value of the currency, the more probability/chance there is of it dropping. It would NOT be possible for all 50 of those drops to be 50 Exalted Orbs. This is because the RNG is functioning under a set of rules which produces the following 'pattern' that I got from the four screens:

Scrolls of Wisdom - 22
Portal Scrolls - 8
Armourer's Scrap - 6
Orb of Transmutation - 4
Blacksmith's Whetstone - 3
and then any other orb I found was found only in the singular, no other orb had a significant and noticeable impact worth stat'asising. I feel sure that everyone who reads this will agree that this is exactly their long-term experience with Currency RNG.

As for the Gems, I have no idea what the RNG is for them, whether they are part of the Currency table, the Equipment table or whether they have their own table.

Moving on to the Equipment RNG:

The top 10 items found were:

1. Scroll of Wisdom - 22
2. Scale Doublet - 10 (3 blue, 1 yellow) STR/DEX level-Required 17
3. Scholar's Robe - 10 (1 blue) INT level-required 18
4. War Sword - 10 STR/DEX level-required 20
5. Chainmail Doublet - 9 (1 superior) STR/INT level-required 21
5. Infantry Brigadine - 9 (1 superior) STR/DEX level-required 21
5. Recurve Bow - 9 (1 superior) DEX level-required 20
8. Portal Scroll - 8
9. Iron Staff - 7 (1 blue) STR/INT level-required 20
10. Iron Sceptre - 7 STR/INT level-required 18

And from this table we can see two more areas where we have 'weighted' RNG. Firstly, there does appear to be additional input to ensure that Chestpieces are more common than any other item. If I had continued the list down to the top 20, it would include all the other Chestpieces between level-required 17 and 22, all bar one stat'ing between 4 and 6, all of them being statistically relevant. The one that wasn't was another STR/INT one. Sholar's Robe stats so high because it's the only INT Chestpiece in the 17-22 level-required range out of 10 available Chestpieces, so some other classes still have better drops in reality.

No other items, apart from weapons, stats as strongly as Chestpieces, and by a noticeable margin. Like-wise a similar 'weighting' is applied to the latest weapon for each class.

The other noticeable 'weighting' of RNG is that the items which drop on a Monster-level more closely match those of the Moster-level. It is possible to get lower level item drops, but they have a reduced RNG probability/chance. You will note the required-level in the top 10 an, now, here's a list of the top 10 item-no-shows by required-level:

1. Level 1 - 10
2. Level 3 - 7
3. Level 2 - 4
4. Levels 4,5,7,10,14,23 - 3
10. Level 9 - 2

All other no-show items are 1 or 0 missing items for the entire level of items. Effectively, the older the required-level, the less chance of it showing up on the Monster-level. 23 only appears because it's the 21+2 and requires bosses.

So, for Equipment Drops, we already see two instances of RNG 'weighting' which no-one really cares about (unlike the currency weighting) and that offer us a much more enjoyable gaming experience.

Now I can begin the rub for the Witch class:

Up to required-level 23 there are 194 Equipment Items that have a chance to drop. The class distribution of those items is as follows:

Str - 27 (14%)
Str/Dex - 34 (17.5%)
Str/Int - 24 (12.5%)
Dex - 31 (16%)
Dex/Int - 24 (12.5%)
Int - 19 (10%)
Neutral - 35 (18%)

So you can see that the Witch is the most under-dropping character before you even analyse the 381 actual Equipment drops, which were as follows:

Str - 53 (14%)
Str/Dex - 91 (24%)
Str/Int - 55 (14.5%)
Dex - 66 (17%)
Dex/Int - 38 (10%)
Int - 33 (8.5%)
Neutral - 43 (11%)

As you can see, the increased RNG weighting from the previous stats is not only heavily boosting some stats, that are already destined to drop more related items, but is also having a significant impact on the last two classes and neutral items. Instead of the INT class getting 1/6 of all drops, they, in fact, are only RNG'd to permit 1/10 of all drops, but, on top of that, in reality, only get 1/12 of all drops as their RNG mechanic stat. The worst of all classes (closely followed by Shadow).

And this is not all. Not by a longshot.

Once I remove the Chestpieces and Weapons and Neutral items we can clearly see the probability of finding other, equally important, armour types.

Total Helmets, Shields, Boots, Gloves available (35% of all Equipment items) by class:

Str - 13
Str/Dex - 10
Str/Int - 10
Dex - 13
Dex/Int - 10
Int - 12

And, for the first time, the INT class does not come last and 'should' have a best possible RNG result. So what were the results:

97 items found (25.5% of all Equipment items)

Str - 15 (4%)
Str/Dex - 17 (4.5%)
Str/Int - 15 (4%)
Dex - 23 (6%)
Dex/Int - 17 (4.5%)
Int - 10 (2.5%)

So when I say that I'm a bit grumpy and think I'm getting hard-done-by with my Witchy drops it's not because RNG is RNG, it's because that's what the current mechanics 'ensure' I get. Hence I would like a debate about class-weighted loot drops.

If I farm four screens then I want the Scholar's Robe to have 10 found with 3 of them blue and one yellow, not a freakin Scale Vest. Once I have a nice Chestpiece and Weapon, I don't want to farm four screens on an RNG probability of 2.5% interest, before we've even estimated the RNG probability of one of those 2.5% turning blue or yellow or having ideal links.

When people complain of 'crap drops', it's not a vague figment of QQ'ing imagination. It's what the 'rules' for the RNG provide. It is not 1/6 equal interest for all classes and it's not even 1/20 for four of the most 'interesting' items 'needed'. And this is at a relatively low level where the diversity of drops is still relatively contained.

I'm not suggesting a major switch in game mechanics or game principles. Weighting already exists. Even just a slight push towards class-weighting would have a big happiness impact without a big notice in the change to a mechanic.

If anything, if you want the game to be 'more' RNG then I would even be happy with a complete removal of the level-required weighted RNG if it was simply replaced by a similar class-required weighted RNG. I'm not suggesting that all a class can find is their own gear, but rather, that they should be more likely to.


CRONK.

"

fsg wrote:

you seem to have misunderstood on fundamental piece of game design here

there are no classes

Yes, there are. In many, many respects.

"

Duskbane wrote:

433 items is nowhere near a large enough sample size to rule out randomness. Anyone who has taken a statistics class can tell you that. Come back at about 433,000 at the minimum.

Yes it is.