MArcomage

Free multiplayer on-line fantasy card game

Please log in

Razorhelm on 19:51, 1. Jan, 2011
Has anyone else noticed that the same rare seems to repeat several times in sequential games? That is, I seem to get on a run and pull the same rare three of four times in a row, in different games, and sometimes my opponent draws that same rare right after me without having played shapeshifter or some similar mirror card. Could there be a glitch in the program such that a card you have not drawn in a while from a certain hand begins to "catch up"?
Lord Ornlu on 19:56, 1. Jan, 2011
Yup, happened to me with Koshei and Dark Horde
dindon on 22:38, 1. Jan, 2011
If you go on the message boards for any game that uses a lot of chance, you will always see topics like this. By nature, we often see patterns where none exist. Our caveman brains were designed to pick patterns out of noise, and false negatives were a lot more costly than false positives. A false negative could mean death (from eating a poisonous mushroom, or getting mauled by a mother bear, or whatever) whereas a false positive just meant we developed a needless superstition.

Yes, random number generators are often biased or flawed, but those flaws usually don't manifest themselves in obvious ways like this. Which of the following looks more random?

1) 00101100000000111111
2) 01101001011101001010

The second one, right? The first one is a random string of 0's and 1's I generated using Python's random module, which uses the Mersenne twister algorithm. The second one I fabricated myself to appear superficially random (i.e. to alternate between 0's and 1's frequently, but without being a strict alternation).

See also: http://en.wikipedia.org/wiki/Clustering_illusion

tl;dr It's a bug in our brains, not in the code.
Razorhelm on 22:52, 1. Jan, 2011
I understand, but these are numbers from one to fifteen, not just zero and one, so the odds of random generation of one of those numbers coming up three or four times in a row and shortly thereafter one of the other numbers coming up three times in a row are much greater, i.e., much longer odds, than the string your generated in your example. I am not denying it may just be the a tiny window into the infinite variety of number combinations, which is why I asked if others have had the same experience. If so, it may be an indicator of a corrupted program rather than a series of random events.
dindon on 01:01, 2. Jan, 2011
The chance of getting the same card of a given rarity 3 times in a row is 1 in 225. The chance of 4 times in a row is 1 in 3375. Given the large number of turns played every day, it's not surprising that this happens.*

Unless someone has been doing statistical analyses of their draws, anecdotal evidence isn't going to be particularly compelling here because people are more likely to remember the outliers (the rare times when they drew the same card several times in a row) and not give much thought to all the more "normal" draws.

* These calculations ignore the little probabilistic nudges that are programmed into the game, such as the decreased probability of drawing a card that's already in your hand.

Edit: This was discussed a few months ago as well. I was equally skeptical back then too (surprise!). Mojko posted a link to the RNG that the game uses, as well as the card drawing code, so if you really want to convince yourself that there are no tricks up the games sleeves, you can take a look at the inner workings.
Lord Ornlu on 03:51, 2. Jan, 2011
Ofcourse, dindon is right on this. It's just peculiar. It happens to me all the time but then again it could be that the times this happens leave me with a deeper memory than the times the system "functions normally" and it gives the impression that it indeed happens very frequently.
DPsycho on 04:27, 2. Jan, 2011
Gotta give props to Mojko. People do report this once or twice a year, and he looks into it every time, always confirming what he already knew: The system is still random.
dimitris on 17:40, 27. May, 2011
This is, of course, not a technical matter, but rather a conceptual one. That is, a deck has undefined number of cards (virtually infinite), so Lady Luck plays her tricks more easily and nastily :)
Damalycus on 20:16, 27. May, 2011
Today I was complaining about my enemy (AI) playing 3 quick last hope cards in a row. He just did the same trick, with three quick pegases.

What is the possible chance of that and previous combined?
I think it's just because I spam "execute enemy move" too fast.
Mojko on 06:07, 28. May, 2011
Speed of execution of enemy move has no effect on cards you draw. He just got lucky that's all ;-) Note that playing 3 Pegasi in a row isn't that improbable.
Damalycus on 11:35, 21. Jul, 2011
Random my butt >_<

2 games in a row. First draw - rare. Same rare.
It's like 6% for drawing rare.
6% for drawing 1 from 15.
and X% for doing it in two consequential games.

This thing has some kind of "memory effect". Like mine pseudorandom engines usually do :D.

*whine, whine, whine*
Mojko on 18:46, 21. Jul, 2011
I'm just going to quote short dialog from my favorite movie.

Barbossa wrote:
What? Impossible!!!


Jack wrote:
No, just improbable.
Unexpected on 18:54, 21. Jul, 2011
mathematics sucks, my draws are kind of magic
... and Jack Sparrow always tells true.