Might be asking for a heaping as it's complete conjecture based of personal experience, for a slightly crazy idea.

I finished grinding my Aldor Exalted. Being the obsessive person I am , I noted down the rep/hour for my various grinding sets. Originally to try different patterns to try to optimise.
I'm fortunate enough that I a) live 10 mins from my workplace so can grind in my lunch hours and b) am 1 hour behind server time so miss the rush.

I'm also one of those people who can make do on 5 hours sleep a night, so often grind into the early hours.

After half a week, I began to 'feel' a difference. Lunch time/prime time evening felt similar, but the graveyard shifts felt so much longer to rep with despite having zero competition. Thought I was going mad, or sleep deprived or something so for the last four days I kept a careful note and kept to the same pattern as much as possible - competition allowing.

Here's what I got....

Lunchtimes (13:00 to 15:00 server): 1575 Aldor rep per hour, with a SD of 259.8
Primetime (19:00 to 1:00 server): 1425 Aldor rep per hour, with a SD of 461.7
Graveyard (1:00 server onwards): 1075 Aldor rep per hour, with a SD of 142.9

Seems my data at least backed up my feeling, not getting any madder just yet.

At the same time and afterwards, I ground Consortium with Zaxxis Insignias. Didn't feel the same there, and some hasty calculations showed no loss rep/hour with them.


That got me thinking what possible reason could there be outside of personal data variance for this phenomena. It occured to me that Zaxxis Insignias are soulbound, Sargeras marks/Fel Armaments are not. I definately felt more 'lucky' with armaments outside the Graveyard shifts.
Perhaps then, this might be a anti-botting measure. Who's online at 2 am server? Hardcore, shift workers and bots. First group will pretty much suck up whatever you throw at them, second are a minority and the third you wish (as Blizz) to dis-incentivise farming contested mobs if not remove completely.
It should be noted my g/hour did not significantly alter depending on shift. I neglected to keep a note of green/blue per hour and as an enchanter they were all dissed so did not effect g/hour. But with the average cash drop from a mob between 1/10th to 1/20th the value of a Mark - that's not where the farmers money is in grinding these mobs.

It occured to me that if I were Blizz and I intended to maximise the time sink of faction grind on non-soulbound tokens - stealth nerfing the drop rate during times farmers/botters are active and 'honest' players are not might be a valid strategy.



OK. It could just be personal experience, and variance swinging against me. But there was eight hours of data for the graveyard shift, I would have expected a higher SD if it were just my data.
Has anyone else come across or 'felt' a similar disparity over time for drop rates?

(If so, salutory warning - grinding rep at 4am in the morning may make you come up with conspiracy theories to relieve the drudgery)

That's pretty interesting data and poses some more questions. Two more things I would do, would be:

1) Keep record of mobs killed per hour and see if you get a correlation between that and your rep. Obviously there should be. I know you said you kept the same grinding pattern, but we can't really quantify that and it could be a source of at least some of the variance.

2) Do it more! A larger sample size as well as making a plot between (Rep/Mob) vs. Server Time would probably be the most accurate indication of whether you are actually on to something significant.

It seems crazy that they would change drop rates based on time of day, and if it was true it would be a big deal and cause a huge uproar from anyone who plays at unusual hours. But that's interesting data, I guess we need people like you to get data on this to find out if its happening.

Yeah, in order to be a little more conclusive it might be a good idea to gather a much larger sample of data from several sources by having several other people keep track of their rep/hour and different times of the day. Might be interesting, but personally it seems a little silly if it is indeed true. How widespread is proffesional farming since the expansion?

Sadly purely anecdotal but also something i kept wondering about for quite a while. Ever since the day of farming elemental fire for ragnaros attempts (later on elemental earth for huhuran, only recently armaments and marks etc) for me the magic mark would be 11 am server time. I would get utterly unlucky with drops before, not get any for 30 minutes straight, then suddenly after 11 i would get loads in a row.
Like i said, purely anecdotal evidence, but somehow something that i kept noticing again and again.

My understanding is that something along these lines exists for certain types of fish being randomly caught, such as stonescales, nightfins, and sunscales. Has that actually been verified at some point? If they've done it with one type of "drop", it proves they at least are willing to do something of that nature and would suggest it would make sense to investigate further.

Two words: Stonescale Eel

Yes those are two words. They also don't prove anything related to mob drops. I don't see anything here but random loot in action. You'll probably see the same thing farming the same mobs in the same area for the same 3 hours a day over 3 weeks and recording each 3-hour period.

Just as a side note, "Bassed" in the title should be "Based."

I used to fish heavily for nightfins. And I do believe the poster was trying to point out that it is possible for the game to determine a drop based on time. If you fish the same spot then you will notice that during the day you will get sunscale salmon and at night you get nightfin. Whether or not fishing is the same coding as actual mob drops, I don't know. Which I suppose is your point. My guess is that the coding would be very similar. I think the poster is on to something, because it appears very possible. If you can perceive the time of day being a factor in one form of drop then you can perceive time being a factor in another. It makes sense to me that the game designers can have 2 loot tables for a mob determined first by time of day then by item. And I too would like to see some proof of it.

Or, the original post should have been about fishing droprates.

This does show the capability, but doesn't support it as far as intent. The fishing drop rates appear to simply be flavor for fishing. Sunscale->Sun, Nightfin->Night. There are also seasonal fish, such as squid (is it called Winter Squid?).

That supports the fact that Blizzard wanted to add variation to fishing drops, but has little to do with attempts to curb farming/lengthen rep grinds by conspiratorial drop changes.

It'd be nice to see more study, but it just feels like "You've got to kill every mob that starts with 'a.' Like 'a ghoul.' Then dance four times at midnight...then AC will spawn."

I've been farming Marks of Sargeras in Shadowmoon Valley off Mo'arg Weaponsmiths and Wrathwalkers using a very typical Pounce->Mangle->Shred combo with 2700AP and the Shred idol from Heroics badges. They basically die before my Prowl 10-second rule finishes, and by doing some maths using the prowl timer + travel time as a gauge, I conclude that I rack up an average of 4 kills per minute.

I've consistently been getting 25-35 marks per MOTW/Omen of Clarity buff, which correlates to Wowhead's drop rates of 25%. (4 kills a minute x 30 minutes x 1/4 drop rate = 30)

Again, anecdotal, but on the other side of the spectrum to what you're experiencing.

(p.s. I've noticed I'm more likely to get a Fel Armament when clenching my teeth and my butt and thinking of one while clicking the body, but that's very, very undeniably placebo at work.)

Did I say it was conclusive?
First line says 'conjecture' and 'personal experience'.

All the stats serve to do is confirm that my 'feeling' that over the graveyard shifts of those four days I was indeed lower than at other times. It is not demonstrative of the global trend.

Now to get that, yes much more data would have to be gathered. Am I going to spend the next three months grinding the same mobs at four different time slots a day? Like hell I am. Even should I do so - any charts posted would get Woo-suk'ed

But I find it hard to believe the Stonescale Eel phenomena was observed from one poor sod fishing continuously for three months. I've already asked a couple of guildies who also grind Aldor/Scryer at those times and the response was one 'Now that you mention it...' and one 'Dunno'.
You flag it up, people watch for it and if it bears out for enough people it gains acceptance.

Regarding mob kill rates, g/hour stayed constant. If anything, as the graveyard was back to back versus disrupted by competition at other terms that'd be a retarding factor on any loot seperation.
And I'd love to say bassed was a cunning reference to the fish related Eel case, but sadly the truth is I checked the body not the title for typos.


Edit: 8 hours in graveyard. 10 in prime. 3 in lunch.
Yup. Nowhere enough to state anything about global drop rates.

To reiterate, that's not the point. I'm not trying to prove this is the case. I'm offering up for discussion the possibility it is, and what ramifications and reasons it might have. Along with notifying other people to watch for it to confirm/deny.

Yes, it's called winter squid. It's the agility buff food. This whole thread reminds me of the old "Molten Core conspiracy drop rate" threads. However, time based drop rates sound more plausible based on outside factors such as anti farmer sentiment. There was no real reason to suspect the drop rates in raid instances. This is the only point to why I would agree for the OP to continue testing. I think that Blizzard might actually be listening to the anti farmer forum trolls.

Actually, if that was 8 separate 1-hour measurements for graveyard and 10 separate 1-hour measurements for prime, then you'd have enough for it to be statistically sisgnificant.


Was it? Or was it 4 two-hour measurements at graveyard and 4 2.5-hour measurements at prime, in which case it won't be enough.

This is really interesting and significant news if it's true. There's no real concrete evidence yet in this thread but it actually sounds sorta plausible to me, moreso than "no paladins in raid = paladin loot!!" theories. Blizzard takes their economy pretty seriously

Chances are there's nothing behind this and drop rate has nothing to do with primetime, but personally I might just jot down date and frequency when I farm things from now on, not that I'll ever remember to.

The best bet at confirming or denying this is contacting the data collection sites and ask them if it's possible to add the collection of time looted in their data collecting programs. As this is all background information, easily sortable by a computer and covers an enormous amount of data, it shouldn't be hard to come up with conclusive results within a month after this going live.

Actually, I semi-believe this is probable, but more of based on drops per hour alotted. I remember when we had some server issues in molten core and the (main) server kept restarting while the instance servers stayed up, yet for some reason the raid ID kept resetting it's time. Meanwhile, every kill we were getting BOE's, we rounded the night off with 19 pieces of tier 1. Another week things were hectic and Mal'g kept restarting while I was farming Essences of Life, and everytime it'd restart I'd get 2-3 more essences. At the end of the day, the AH was flooded with all sorts of mob-farmed goodies.

No true proof yet, but if I ever come across a time where the servers are whacked again I'll go for data recording.

Heh, time based drops would certainly explain why some guilds would get 2-3 thunderfury drops in a row, and why another would never see any.

Then again, same dead horse, new beating stick.

What does SD stand for? I missed that.

I've always had the impression that a mob's loot is generated when the mob is spawned, NOT when it is killed. This might only be in instances (change that to "when the instance is created, mobs and their loot are created"), but would make just as much sense outdoors.

In this case, differences at different times of day would still be linked to other player activity; even if noone is within sight of you, during prime time someone, somewhere, is always killing similar mobs with the same loot table. Items with an intended percentage-based drop rate should technically work out the same, however, which would go against your findings. I'd like to know if you did this several different days to help eliminate the possibility of odd streaks or some additional variable.

I came, I saw, I confused matters more. A job well done.

As was mentioned earlier, keeping track of kills/hour would probably help more. You'd want to analyze expected rep per kill rather than per time. Possible theories I've come up with are that with a dynamic respawn rate, as it gets late at night, there are fewer mobs spawned due to only 2-3 other people being in the zone along with you.

For the lunchtime grinding, the higher number could be due to a short spurt of play without having to wait for respawns at all. I can't imagine you get that much time in once travel and consumption of a sandwich are factored in.

Lastly, looking only at Primetime vs. Graveyard, it's notable that even if you consider the Graveyard sample completely accurate (SD of 0), it's within a single standard deviation of Primetime's mean value. That's considered statistically insignificant. Graveyard shift is also less than 2 standard deviations away from Lunchtime's mean. Simply put, you need a larger sample to decrease the mean standard deviation, and you still need to analyze other factors. This data alone suggests there's not a significant difference.

Foxery: SD is "standard deviation"

I ground exalted Scryers the second week the expansion was out, and definitely saw 'streaks' but since those were some pretty hard play sessions time of day seemed to have little to do with it. Was approximately 2,000 rep an hour at any time. There were hours with far less, but again they were just interspersed in there. Overall it took ~10 hours of grinding.