I'm startig to think Wizards broke the shuffler in one of their updates.

Between all the "waiting for server" screens, the client crashing for non-english setups, and the overall memory usage, it wouldnøt be unlikely they managed to break the randomizer in one of their previous updates.

A few examples.

4 times in the last 20 matches have my mono-white 22 land deck pulled 7-8 lands in the top 12 cards.
At least once per day my 26 land Simic ramp deck will pull 3 lands in the top 18-20 cards. On average that should happen once in every 3000 matches, yet it happens daily for me.

Both are fringe situations which should happen extremely rarely, yet they happen daily.

I see on stream that I'm not the only one getting screwed or flooded to the extreme.
CGB is running 28 land decks, considering 20 land decks, and he's still getting screwed regularly. Merchant is getting f**ked over by the randpomizer so often, he's coined the term; "Getting Magic the Gatheringed".

... To be fair I also see it happening to opponents. They're stuck on 2-3 lands while I cruise to an easu (if joyless) victory.
MGA feels less and less about skill, and more and more about praying to RNGesus.

I've been feeling like I've been getting mana-screwed quite often lately, but, as usual, there's confirmation bias at work. We can't assume the shuffler is "broken" without first knowing how it works, what inputs it takes, etc.

True, it could be broken and it could be truly random, we don't know, but sometimes it just helps to rant ... even in writing.

Yah but what a boring, overranted rant.

This is the first time I’ve heard the shuffler is br0ken

Well, now you know

if you were curious you could probably keep records of the games you play and if its even remotely as biased towards low and high land games as you're suggesting it should be apparent even within a relatively small sample. If you're playing 50 games and you have multiple games in that set where the number of lands you draw are 1 in a thousand or less then its probably much more reasonable to assume the game is biased in determining how many lands you get than that its just a statistical anomaly.

considering how common it is for people to feel like their results are too anomalous too often to be random in games where the results are definitely random, its more probably the case that your mind is just playing tricks on you.

Welcome to Magic!

Mana screw & flood are easily the worst parts of Magic, and there are lots of other CCGs which attempt to fix it. You could try one of them - in fact come to think of it, off the top of my head I'm not aware of any other CCG which has mana flod and mana screw.

And you were boring all the way back in the Microbless and Wizards_Sean days on the official forums, but you don't see me prancing around with a need to talk down to other people.

--

As I said; 4 times in the last 20 matches have my mono-white 22 land deck pulled 7-8 lands in the top 12 cards.

It has happened a further 2 times in the last 9 matches (since I made this thread).

I know random **** can happen, but this is a fringe scenario which should happen on average aaround once every 3000 matches, and here I am watching it happen in roughly 20% of my matches.

Considering the amount of errors we've seen in the MGA client over the past 2 years, it's not entirely unfeasible that Wizards actually broke the randomizer.

I could probably do tests with 24 land decks, and this would only happen in 1% of the matches, but that would only indicate that the radnomizer can't handle 22 land decks ... For all I know that is exactly the issue I'm running into here.

are you actually recording all the information though? humans aren't great at remembering things and having the specific information would probably be useful in verifying exactly how anomalous the results are.

Yeah Wuster, YOU’RE boring

I'm sorry I had the audaciosty to imply your precious WotC aren't perfect BB, but hey you keep that **** going.


Yup, I use a tracker, and then plot in land draws in an excels sheet after matches.

« Audacity »

I’m not a Wizards apologist tho. I think they do some things well, and other stuff not so well. Like any company or person, really.

Except Cucho, he does everything perfect and is literally dreamy.

If I'm calculating it right, probability of drawing 7-8 lands in top 12 cards with a 22-land deck is slightly bigger than 8%.

Also none of the statistics you've said happen once every 3000 matches. The odds of drawing 3 lands in top 18 card in 28-land deck hits 50% at about 300 games.

can i see the spreadsheet?

It's been too long since we'd had an RNG discussion, the good ol' days

Have u considered that you're just getting really lucky getting all this bad luck?

Meh... I’m not getting involved.

I don’t even play the game. Ya’ll have fun.

We do know how the shuffler works - it was explained.

For each card in the deck, the game runs a rand() function and assigns it a number. Once all cards have been assigned a number, they are placed numerically in the pile. Done.

The rand() function is rather basic at this point, and is of course not truly random. However, the psudo-random numbers used in any programming language are so big, that the human brain will not be able to discern any difference. It's comparable to the function used when encrypting data.

To boil it down, if you would be able to predict an actual outcome from a rand() function correctly, you would have to be able to guess the next number created. Say, 1 in 4 billion. And that happens for each card in the current library-pile each shuffle. So 60 predictions. Technically, you don't need to predict the actual number, just the numbers relative position. But still.

Now - it will still feel funky in unranked bo1 when they use the "run it twice and select the most playable hand based on some criteria" thing. Your odds of getting a more playable hand just doubles.

Other than that, I feel your frustration. Having my opponent on 2 life, with 2 slitherwisps in play, 3 counters in hand and dying 6 turns later because I could only draw lands 6 turns in a row. With 28 flash-spells in the deck in total. GAAAAAWRRGGG.

Or getting to diamond 1, 1 game away from mythic and losing 8(!) games in a row, 3 of which are total floods, 2 completely mana starved, 2 to opponent pulling out a combowin out of nowhere(actually, pulling it out of their deck, but that is SO random!) and 1 game because I become slightly frustrated and make 3 missplays in the same freaking game.

And remember that the shuffler does not account for the fact that a card is a land or not. It just randomizes it. Make a deck with 60 plains, all different art. Let the shuffler have at it. It will still mix those lands up as mindlessly as it does anything else. It's just hard to see with lands.

Or go with another old theory on these boards - take a deck you've just played. Go into the editor, replace any 4-of with another 4-of. Save the deck. Next game you play with the deck, you'll have a very high change of getting 2-3 copies of THAT card in your opening hand. So make sure you do that with your combo-pieces or ramp-cards.

I don’t play either but this is the first Shadowcran-like poster we’ve had in a while so you should jump in on the fun! He even went after Wuster! Who goes after that guy?! Lol

I don’t know. Wuster was right too. This rant has been done too many times.

It’s especially silly since everyone has the same shuffler, so even if it is broken, it’s broken for everyone else too. No advantage gained or lost, so seriously who cares?