Exaggerating, Cherry Picking, Low Sample Sizes, and RNG

I have over 1200 all resistance and over 1500 all resistance with a barb buff and elements do full dmg to me, REALY BLIZZARD? REALY?

-Forum Poster

I have been farming [Act 2 goblins] for 2 hours now. I have seen him maybe 10% of the time, rather than 30% of the time. My growing sample size is suggestive of a nerf; couple this with the messed up drop rates atm and I am thinking wasting my time doing something else would be good.

-Forum poster

Yep [the Act 2 goblin]’s basically gone now as of this last maintenance. I ran 15 minutes at a time to sample and he never showed up once at either Alcarnus or Ancient path. Previous to this maintenance I would almost have an antire bag full of blues and rares to vendor in that same 15 minutes span, courtesy of the loot muppet. Now he simply is not there.

-Forum poster

After 300 hours or playing I found 1 lvl 18 legendary barb belt. woohoo! Most legendaries and set pieces are being sold by Blizzard. Most of the good items in the AH(s) are made by Blizzard to sell to you to make money, duh.

-Forum poster

Zero i63 items for 2 days straight in act 1.

-Forum poster

What’s with the total inability to use logic and basic math these days? Why do peoples’ arguments need to be exaggerated and made one-sided to the point of a religious debate? Why do forumgoers choose to look at only the facts that support themselves and ignore all the facts that contradict them?

All of the quotes above, which I took from real Diablo 3 Battle.net threads, fail to make any good logical argument that Blizzard should even bother spending the time to read, let alone address. They spread false conspiracies about the game that many other forumgoers are likely to believe, and then hop along with. They are the makers of self-perpetuating myths.

Example 1

I have over 1200 all resistance and over 1500 all resistance with a barb buff and elements do full dmg to me, REALY BLIZZARD? REALY?

In this example, a forum poster jumps on the myth that Blizzard made a drastic nerf to the resistance stat. As far as I could dig down, I found the initial ground-zero thread to be one that quoted numbers of damage taken with and without armor. Upon examining the numbers carefully, however, anyone can see that the resists are working just as before, with about a 5-10% difference that can easily be attributed to exaggeration on the original poster’s part. The significant nerf featured in the conspiracy theory is simply nonexistent.

Yet, a significant amount of posters replied that they had noticed the same thing, even when the numbers didn’t even show any change to the way resists work. So why did the poster in the example feel as if he was taking full damage? Well, he didn’t post how much vitality or armor he had, which are both important in determining one’s effective hitpoints. He also failed to mention which things specifically were killing him. A plague patch won’t do much damage, but standing on top of an Arcane Sentry will. How do we know he is not just using a confirmation bias to be noticing high-damage affixes after the patch and only choosing to remember low-damage effects before? Or he could have just had bad RNG with dodging and shield blocking. There are many factors at work in computing damage taken.

Example 2

I have been farming [Act 2 goblins] for 2 hours now. I have seen him maybe 10% of the time, rather than 30% of the time. My growing sample size is suggestive of a nerf; couple this with the messed up drop rates atm and I am thinking wasting my time doing something else would be good.

The goblin spawn rate in the popular areas in Act 2 was much earlier ballparked to be around 30%. This poster claims that based on his runs for two hours, the spawn rate has been nerfed to 10%.

The poster failed to give any numbers at all, so he could have just been bad at math, and mistaked something near 30% for 10% due to emotional factors and confirmation bias. Another possibility is bad RNG. It’s perfectly probable to go several runs in a row without encountering a goblin. In any 4 consecutive runs in a row, you have a 0.7^4 = 0.2401 chance to not encounter a single goblin.

Similarly, the chance of not getting a goblin in 10 consecutive runs is 0.7^10 = 0.0282. This means that if 10,000 people were to go goblin farming and do 10 runs each, we would expect 282 of them to not encounter a single goblin in the 10 runs. These people are much more likely to complain, which results in the vocal minority seen on the forums.

Example 3

Yep [the Act 2 goblin]’s basically gone now as of this last maintenance. I ran 15 minutes at a time to sample and he never showed up once at either Alcarnus or Ancient path. Previous to this maintenance I would almost have an antire bag full of blues and rares to vendor in that same 15 minutes span, courtesy of the loot muppet. Now he simply is not there.

This is written by a different poster than in example 2, but it is basically just a continued perpetuation of the myth. 15 minutes is an incredibly small sample size, and this person failed to give numbers as well. Let us see how a couple of other people have been seeing this:

Nope, seems fine to me. Soloing him is not beneficial though, as you can run 8+ instances and never have one spawn…RNG and all. Other times I have them spawn back to back.

i just got him 6 out of 10 times since i read your post… i know thats a small sample but still great % run there for me . ill do some more and come back and post.
another 10 runs 3 out of 10
another 10 runs 4 out of 10 ..
another 10 more 4 out of 10 ..
so 17 out of 40 is 40+% pop rate on goblin for me tonight…

This shows that you can’t just use such arguments, unless you are willing to look at a large number of cases. Cherry-picking the one 15-minute interval where you didn’t get a spawn is not going to refute the other 95% of cases where you get multiple spawns.

Example 4

After 300 hours or playing I found 1 lvl 18 legendary barb belt. woohoo! Most legendaries and set pieces are being sold by Blizzard. Most of the good items in the AH(s) are made by Blizzard to sell to you to make money, duh.

This guy just had bad RNG, yet he’s blaming Blizzard and is baselessly accusing Blizzard of selling good items on the auction house.

Let us see how some of the responses went:

I found 2 legendaries yesterday, total of 11 now, 3 set items 2 of them natalya’s set.

I find at least one a day, but do not sell on the ah they are not worth taking up the space for actual items that will sell.

I found 12 legendaries and one of them is green.

I have found so far 6 legendary items and 1 set item in my 150-ish hours of game play…

Example 5

Zero i63 items for 2 days straight in act 1.

Bad RNG. Plus, 2 days straight (48 hours straight) has to be a bit of an exaggeration. The Gambler’s Fallacy is also probably at work here. If people don’t get an ilvl 63 item in a while, they expect one to come along any moment because it is due to roll. Sure, that’s how the system would work if it was pseudorandom. But since it’s in fact truly random, even if you grabbed 30 rares and none of them was ilvl 63, the chance of the next item being ilvl 63 is exactly the same as before. And because the system exhibits true randomness, there are also situations that occur on the opposite side of the spectrum:

I got 2 helion bows and 1 set item from 1 elite .

Conclusion

If you want to complain about the game, use logic and maybe even some elementary math. The game certainly has many flaws, and there are many level-headed ways to offer suggestions. However, making baseless accusations accomplishes nothing.

Advertisements
This entry was posted in Forums, Players and tagged , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s