Guild Wars 2

Every day I see another thread about this. Every time, it is some person who salvaged about 10 items.

Luckily, I do save a lot of my data from various actions in this game, and I normally don’t release it publicly, but I will to help dispel this misconception.

All of these salvages were done yesterday, 11/23/2012. All were on rares level 70 and above, but all below level 80. All were done with a Mystic Salvage kit. To my knowledge, the basic salvage kit gives the same chance for ectos. I do my salvages in waves, and I will show the individual blocks.

16 Salvages – 20 Ectos
16 Salvages – 13 Ectos
11 Salvages – 9 Ectos
11 Salvages – 5 Ectos
17 Salvages – 15 Ectos
12 Salvages – 13 Ectos
15 Salvages – 13 Ectos
14 Salvages – 12 Ectos
16 Salvages – 11 Ectos
13 Salvages – 9 Ectos
16 Salvages – 16 Ectos
15 Salvages – 14 Ectos
8 Salvages – 6 Ectos

180 Salvages total , 159 Ectoplasms = 88.3% rate.

Most of my data from up to one month ago shows roughly a 0.9 ecto/salvage rate, and I have salvaged roughly 10,000 rare items to date.

So please, before you decide to ask this question for the 25th time in the last 5 days, ask yourself: Is my sample size sufficient? Are my results unlikely to be the result of random chance?

For reference, I also have an enormous database of forge upgrade rates, and other various data points on most of the actions in this game that use a RNG.

So far, the only noticeable change I see has been in the Cursed Shore rare item drop rate. I use Grenth runs as a standard, and I understand that it’s not a steady state (variables in length of run, quality of play, quality of party, etc.), but the # of rares per run has been cut to ~25% post-patch. This is a large enough change to provoke my curiosity, but I do want to collect more data before I make this assumption. Grenth is currently bugged on IoJ, and I haven’t been able to collect data for several days.

Allegedly, the upgrade -> precursor rate has been upped, but I have insufficient data points to support or contradict ANet’s claim. I don’t really assume things in the patch notes to be true: many of the things such as the champions “guaranteed” to drop items have been patently false. My confidence in ArenaNet’s ability to properly implement their changes is lukewarm at best, which is why I do prefer to collect my own results.

Just because it works for you, doesnt mean it works for everyone else.
I for one got almost always 1 ecto per salvage on average a few weeks back, now its more like every 10th attempt gives me one ecto.
Last two days i got 0 ectos back to back from all my rare items, something that didnt happen before.
So grats to you for your 88,3% chance, unfortunately thats not the same rate i was getting.

I’d challenge you to collect your data over several hundred salvages, rather than estimating. I would say most of these cases tend to be observer bias. I have indeed run into strings of 5-10 salvages with 0 or 1 ectoplasms. This can happen, but 10 or 20 salvages is not a representative sample size.

I should like to think that this % chance applies universally, because the remaining conclusion would be that different players have different ectoplasm salvage rates, which is something that would be highly unusual.

For the record:
I didnt come to my conclusion because i went to the forums, i went there AFTER i noticed something fishy myself.
I see a big difference when i use my BL-kits though, where the rate is the same as before.

Fact is, we dont know the exact values behind the rng system, and a lot of people are complaining about this.
Some maybe lucky and get same results as before, others are obviously very very unlucky (or bugged out).

I just remembered that my mystic kit was made (forged) with another character and i couldnt equip it with any other char although it said accountbound.
Since the last patch i can do exactly that.
Maybe that has something todo with it?

I doubt it.

This is not to say I rule out the possibility of a bug. This game has many, in varying flavors of frustrating. You could be right.

I only mean to say that my doubts are generally assuaged by data, and I do not like to start a ruckus until there is some hard evidence that there is an issue in the first place.

Track your salvages starting now. If 200 salvages later, you’re still getting like 1 ecto per 5 rares (0.2), message me about it, I certainly am very interested to know. You have no reason to lie to me. Tell me the results and if they’re still odd, I’ll look into some more ‘strange’ variables on my end (level of the item within the ecto-eligible range, individual salvage kit variance, variance of using a kit crafted by another character, things that shouldn’t matter but could be tripping some kind of bug)

Statistically speaking, to get an accurate value, you’d need to salvage at least ~500 random or sequential samples to call it sound. Then you can achieve roughly 95% accuracy with 99% confidence (and hoping I did my math right).

LFk seems to have done that, and kudos to you. Actually, with binomial distribution, about 13,179.4 salvages are required to give a 99% accurate and 99% confident reading of the salvage rates. Though, I highly doubt you recorded your 10k salvages, or that you even managed to count to 10k manually, if you really did, then your answer would be the only credible one and would likely equal to that of the one written in formula by the developers.

If you haven’t salvaged that much and talking about how it’s this or that, you need to go salvage more.

So, if you want to compare and say method X works better than Y, or it works worse than before, you need to record ~500 samples with one method X/time without bias and then repeat with the method Y/new time. Only then, we can credibly say it is better or anet changed it or something.

I have 9,541 recorded attempts.

It’s actually not as bad as you might think, not to mention I didn’t even decide to start counting until I had done a great deal.

As you can see from above I do my salvages in waves. When I did more of this, the waves were in groups of 25-35 salvages per wave. Sometimes 40+. There really are only a tad more than 300 recorded data points for the ectoplasm salvage category.

It would be a great deal more harrowing if I started considering the level of each individual item, instead of just grouping them into “70-79”. THAT, would be impressive if i reached 10k.

Hi,

It seems that you believe that basic salvage kits have the same rate of salvaging ectos as the other higher quality kits such as mystic kits or black lion salvage kits. That was my belief as well but I do not have any evidence showing so. I was wondering if anyone here would have any evidence regarding the quality of salvage kit versus rate of salvaging ectos because some (or rather most) believe the complete opposite but without confirmation from a dev or substantial evidence I cannot completely rule out one or the other.

As quoted from my question in another thread:

Simply put, does a Black Lion Salvage Kit warrant you a higher chance (50% as advertised) of obtaining ectos or are ectos not considered in the “rare material” spectrum at all and hence you’d have the same chance of obtaining ectos regardless of quality of salvage kit (e.g. crude vs black lion)?