Hmmmmm, don't know if I like the new raid mechanic:

[Stanley_Nicholas]

The probability of actually drawing 0 raid items if you fill all 12 positions is almost null.

Actually no, there is a rather significant chance for nobody in a 12-man group to pull a raid item. 11.2% chance. And in that same 12-man party, there is only a little over 60% chance that 2 or more raid items will drop. Someone calculated it all out in the announcement thread, it's very interesting to see:

At first read I liked the Idea and then I did some maths OMFG what are you doing Turbine.

So we are going from a 100% chance of getting 2 items raid loot to

11.22% of getting nothing (full raid)
88.78% of getting 1 item (full raid)
61.87% of getting 2 items (full raid)

Yes i know its 1 in 6 great so I personally have a 16.67% chance of getting an item or 83.33% chance of bending over and taking a beating from the turbine nerf stick.
.
Lets look at the chance of everyone in the group pulling NO LOOT!
.

01 MAN RUN = 83.33%
02 MAN RUN = 69.44%
03 MAN RUN = 57.87%
04 MAN RUN = 48.23%
05 MAN RUN = 40.19%
06 MAN RUN = 33.49%
07 MAN RUN = 27.91%
08 MAN RUN = 23.26%
09 MAN RUN = 19.38%
10 MAN RUN = 16.15%
11 MAN RUN = 13.46%
12 MAN RUN = 11.22%
.
Now lets look at the chance of pulling at LEAST one item of raid loot.
.

01 MAN RUN = 16.67%
02 MAN RUN = 30.56%
03 MAN RUN = 42.13%
04 MAN RUN = 51.77%
05 MAN RUN = 59.81%
06 MAN RUN = 66.51%
07 MAN RUN = 72.09%
08 MAN RUN = 76.74%
09 MAN RUN = 80.62%
10 MAN RUN = 83.85%
11 MAN RUN = 86.54%
12 MAN RUN = 88.78%
.
Now lets look at the chance of pull at LEAST two items of raid loot.
.

01 MAN RUN = 00.00%
02 MAN RUN = 02.78%
03 MAN RUN = 07.41%
04 MAN RUN = 13.19%
05 MAN RUN = 19.62%
06 MAN RUN = 26.32%
07 MAN RUN = 33.02%
08 MAN RUN = 39.53%
09 MAN RUN = 45.73%
10 MAN RUN = 51.55%
11 MAN RUN = 56.93%
12 MAN RUN = 61.87%
.

Lets loot at the chance of pulling multiple items for a 12 man raid
(# Items@Percentage Chance)
.
00 @ 11.22%
01 @ 88.78%
02 @ 61.87%
03 @ 32.26%
04 @ 12.52%
05 @ 03.64%
06 @ 00.79%
07 @ 00.13%
08 @ 00.02%
09 @ 00.00% (7/515090)
10 @ 00.00% (1/1048576)
11 @ 00.00% (1/8388608)
12 @ 00.00% (1/8388608)

Or even a six man raid
(# Items@Percentage Chance)

00 @ 33.49%
01 @ 66.51%
02 @ 26.32%
03 @ 06.23%
04 @ 00.87%
05 @ 00.07%
06 @ 00.00% (10/463459)
.
.
*Please feel free to let me know of any mistakes in the above more than happy to fix..
** Please note all figures are subject to my own stupidity...
*** Also most figures are AT LEAST not EXACTLY x Items!!!

Edit: The math is sort of accurate. It overestimates the chances of getting loot in some places.

[Blackwolfe]

In addition, the primary warded chest creates the list of characters that are allowed to loot the chest at the time chest spawns. This list contains all the members of the raid at the time the list was created. Characters not on the list do not have treasure generated for them when they open the chest and may not have items re-assigned to them.

BTW, The announcement doesn't say that you have to be "in the raid instance" to get raid loot generated for you. It says that it checks who’s in the raid party when the chest spawns. So they make the Reavers exploit a little harder, but still just as doable by a dishonest guild. I hope they thought of this and that the announcement was just worded this way.

[akla_thornfist]

as soon as you beat the raid the chest spawns so anyone added after that will not be on the list, i dont see how they can cheat that.

[Beherit_Baphomar]

I have little issue with new raid reward system. In fact, it seems there will be greater opportunity for more raid items rather than less.

But will the raid items drop based on class like regular loot? Should raid items, as opposed to regular loot, be fully randomized independent of your class?

Items from chests have never dropped based on class. Thats end rewards. Items in chests are completely random. As I am sure the 'warded' chests will be.

I dunno, I think I like this idea. Its better than what we have now. Everyone has an equal chance at something and its better to 12man instead of shorting it.

I dunno tho, not sure as of yet.

[CSFurious]

i think everyone deserves a chance to go on a raid, and the previous system discouraged that so this change is good for the player who raids once or twice a month vs. the catassers who raid every raid every 3 days, maybe now they will teach others their ancient-DDO secrets because they will not be worried about 2-manning or 4-manning raids that are designed for 12 players to go on (not necessarily needed) as this game is about playing with others (at least to me)

[kaishaku]

Scale the chance based on the number of people eligible to
pull items from the chest, such that:

* 12 member parties have the above-mentioned 1 in 6 chance
* 2 member parties have a greater chance, perhaps 1 in 1

For example, the chance could be 1 in party_size/2.

Adjust the formula as desired. Scaling for party size seems
simple enough and seems to address the complaints I've seen.

[Mhykke]

i think everyone deserves a chance to go on a raid, and the previous system discouraged that so this change is good for the player who raids once or twice a month vs. the catassers who raid every raid every 3 days, maybe now they will teach others their ancient-DDO secrets because they will not be worried about 2-manning or 4-manning raids that are designed for 12 players to go on (not necessarily needed) as this game is about playing with others (at least to me)

I don't see how making a full raid encourages more people to go on raids.

Under the current system, what's the difference b/w having a full 12 person raid, and having 2 six man raids? Well, besides the fact that you get twice the named items?

By the same token, if I'm looking for a raid group to join, why is getting into a 12 person raid easier than getting into one of 2 six man groups? I'm still looking to fill one of 12 slots, no?

[oronisi]

The probability of actually drawing 0 raid items if you fill all 12 positions is almost null. The same can be said for drawing 12 items with all 12 positions. The greatest probability is 1-5 items and then it starts to descend again. Honestly, this is a very very excellent innovation. Once in awhile there will be bad luck, but that will be basically 5 raids out of 100 probably scoring terrible luck if they fill all 12 positions.

All in all, this new method should enable characters to max out their wish-lists earlier and make it possible for new players to score give-aways (or new characters to run raids in lieu of the maxed-out character).

I believe this new system will also strengthen guild and inter-guild relations. Awesome work Turbine.

As for prestige -- there are plenty of quests out there that would be much-more impressively solo'd or two-manned than a raid. That Tower in Delera's Graveyard on Elite for one would be amazing if solo'd.

Actually....there's an 11% chance that a full raid group will walk away without a single raid loot at the end of the raid.

That's far from almost nil.

[Allorious]

I must admit, after reading all the posts I lke the current system better. Not that I like our current system, its just better than the chance that no raid loot would exist at all. I'll take guranteed 2 pieces of loot over the possibility of none. That being said I REALLY like the token idea. Collecting a token for completing the raid and after you have collected 6 tokens, or whatever, turning them in for a piece of raid loot from the NPC who gives you the raid quest. Like others have said, it doesnt matter, this is the way Turbine wants it...for whatever reason.

[Nirvana]

Just in case someone doesn't understand the math...


If one person get 1/6 chance to get named item, then he has 5/6 chance to get nothing.

Let Pn(k) denote as probability of getting k named item(s) from n-man raid.

**Pn(k) is simply a binomial distribution, its distribution density function is

Pn(X) = (n C k) * (1/6)^k * (5/6)^(n-k)

where (n C k) means n chooses k.

So 1-man raid, you only have 16.67% chance to get 1 named item.
P1(0) = 5/6 = 83.33%
P1(1) = 1/6 = 16.67%

For 3-man raid which is current 'ideal' size of raid,
P3(0) = 5/6 * 5/6 * 5/6 = 57.87%
P3(1) = 3C1 * 1/6 * 5/6 * 5/6 = 34.72%, where 3C1 = 3
P3(2) = 3C2 * 1/6 * 1/6 * 5/6 = 06.94%, where 3C2 = 3
P3(3) = 1/6 * 1/6 * 1/6 = 0.4630%

For 12-man raid, using the above equation, we have:
P12(0) = 11.22%
P12(1) = 26.92%
P12(2) = 29.61%
....
P12(12) = 0.00000000045939%



I don't think the math quoted by Stanley Nicholas is correct, but it doesn't matter. We all can see that the future raid loot system sucks so bad.
For 12-man raid, you have 11.22% chance to get NOTHING from the chest, 29.61% chance to get 2 items.


Note: when I say Nothing, it means 0 named item. Who cares about the regular loot which you can get anywhere else in 2mins.

[Aspenor]

BTW, The announcement doesn't say that you have to be "in the raid instance" to get raid loot generated for you. It says that it checks who’s in the raid party when the chest spawns. So they make the Reavers exploit a little harder, but still just as doable by a dishonest guild. I hope they thought of this and that the announcement was just worded this way.

It's not really any harder. You can't swap out characters, plain and simple. If you're not in the party at the time the chest spawns (whenever that is, be it the beginning of the raid or the completion) then you don't get loot, period. There is no way to swap characters and receive loot.

[Vorn]

I noted on the other threads that I kind of like this idea, from a casual player's perspective who hasn't gotten much in the way of raid loot. I'd really prefer the token thing--a grind with an end in sight fits my personality better than the grinding lottery we have now. It just appears to me that my chance to come along on a raid, if even as a filler, has improved (or maybe not....I doubt the change will really affect folks who are comfortable doing it the way they've been doing it and have a good system--now it will just take longer). I do also like the idea of improving the chances of named items dropping under 'hard' or 'elite' to help out the really good players.
Just my 2cp.
Cheers!
V

[Gol]

The probability of actually drawing 0 raid items if you fill all 12 positions is almost null.

Actually, it's an 11% chance. That is, better than 1 in 10 raids will produce no loot at all.

[Stanley_Nicholas]

I don't think the math quoted by Stanley Nicholas is correct, but it doesn't matter. We all can see that the future raid loot system sucks so bad.
For 12-man raid, you have 11.22% chance to get NOTHING from the chest, 29.61% chance to get 2 items.


Note: when I say Nothing, it means 0 named item. Who cares about the regular loot which you can get anywhere else in 2mins.

Yeah you're right, parts of what I quoted actually aren't correct, though they still give the same impression. It turns out the probability of a full raid group getting raid loot is overestimating the likelihood by more than a factor of 500. I was at work when I quoted it, and only checked a couple of the numbers.

[Blackwolfe]

It's not really any harder. You can't swap out characters, plain and simple. If you're not in the party at the time the chest spawns (whenever that is, be it the beginning of the raid or the completion) then you don't get loot, period. There is no way to swap characters and receive loot.

Just because you’re in the party does not mean that you are in the instance. Hell the guild can be in the raid party and PvPing in the Lobster while someone solos the quests finishes and calls the rest in to get their "hard earned" winnings. Or someone with two accounts and computers can have a ghost member waiting just outside the raid to enter when it is completed, only to switch the raid loot to the first player.

I mean this I looked at this for two minutes and came up with these exploits. I guarantee that the people who will do this have all ready figured out 10 more ways to continue to exploit the raid.

[Kistilan]

Actually, it's an 11% chance. That is, better than 1 in 10 raids will produce no loot at all.

The overall production of items should increase. I guess I didn't make myself clear on that. There is a possibility of 12 items dropping. In the end this will create less-runs for most raiders. I'll take a Zero run once in awhile to see a pay-out of 5 items a few times.

11% isn't quite accurate. But less than 10% is null enough for me. That's 9/10 of the time there's at least ONE item. Probability spectrum incorporated, the chances of 0 items with a full 12-man is very negligible. Again, I'm no mathematician, but I haven't seen any formulas yet that strike me as correct. I have had up to Calc 2 in college and the probability equations were not my favorite, but if someone were to produce the equation I think we'd find about a 5-9% chance of no items (not 11%).

[Gol]

The overall production of items should increase.

If you believe in fuzzy math or count the "substitute" loot you get 5 times in 6, ok. If you believe in statistics and logic and only count raid loot, you are oh-so-wrong. Sorry, math wins in my book.

11% isn't quite accurate. But less than 10% is null enough for me.

It's actually 11.2% for no raid loot at all.

I have had up to Calc 2 in college and the probability equations were not my favorite, but if someone were to produce the equation I think we'd find about a 5-9% chance of no items (not 11%).

I'm 2 classes short of a Math degree (Abstract Algrebra I & II), and aced all 3 of my statistics classes.

Probability of event N happening in X number of consecutive distinct trials is P(N)^X. So, if N is the odds of "no loot", P(N) is clearly 5/6 = 83.333% chance, and X is 12 for the number of people in a full raid. 83.333%^12 = 11.216%.

Happy now?

[Vorn]

Nirvana gives the equation above. It's a simple binomial distribution. The 11.22 is what I get on my calculator for 12 members, collectively pulling zero items, with p=1/6 of getting an item.
Not sure about all the other numbers though...this old math prof is too sleepy to keep pushing buttons.

Cheers!
V

[kaishaku]

A 10 minute brute force examination agrees with the 11% number provided.

Here are the results of code which approximates the system being discussed.

   0 pulls - 11.19%
   1 pulls - 26.89%
   2 pulls - 29.69%
   3 pulls - 19.70%
   4 pulls -  8.88%
   5 pulls -  2.85%
   6 pulls -  0.67%
   7 pulls -  0.11%
   8 pulls -  0.02%
   9 pulls -  0.00%
  10 pulls -  0.00%
  11 pulls -  0.00%
  12 pulls -  0.00%

A 10 pull or better is "1-in-a-million" or better.

I think the mechanics of small vs large raid parties are more
important than how many times you will see 0 special items.
In the end, you will see the same number of items as before,
but only with 12 member parties.

I think there is a simple solution...

Scale the chance based on the number of people eligible to
pull items from the chest, such that:

* 12 member parties have the above-mentioned 1 in 6 chance
* 2 member parties have a greater chance, perhaps 1 in 1

For example, the chance could be 1 in party_size/2.

Adjust the formula as desired. Scaling for party size seems
simple enough and seems to address the complaints I've seen.

Each member of a 6 man team might have a 1 in 3 chance.
Each member of a 4 man team might have a 1 in 2 chance.

[SableShadow]

i think everyone deserves a chance to go on a raid, and the previous system discouraged that so this change is good for the player who raids once or twice a month vs. the catassers who raid every raid every 3 days, maybe now they will teach others their ancient-DDO secrets because they will not be worried about 2-manning or 4-manning raids that are designed for 12 players to go on (not necessarily needed) as this game is about playing with others (at least to me)

I recommend a thorough reading of "Atlas Shrugged" by Ayn Rand.