Jump to content

Error in the shuffling system?


Gamemaster76

Recommended Posts

Is it possible that the shuffling system is broken?

im asking because ive noticed something about the game in the last couple weeks that most of my matches have always been one sided, at first for me then it switched for my opponent.

 

I picked one deck to test this with and had been using it constantly, when i first made it i got over 15 wins in a row, at one point however i started losing constantly due to getting hilariously bad hands game after game.

 

yes the card game is about luck and these things happen but its also an application, and programs can have bugs. a non stop win streak followed by a non stop bad hand streak cant just be luck. its not that its a bad deck, i got 15 straight wins at first and now i struggle to get a win with the same deck.

then i tested with other decks that generally started me off good and now i only get the worse possible hands i could get.

 

maybe its nothing and theres nothing wrong and im just salty but i also know technology doesnt always do what its supposed to do. so on the off chance that is something wrong, i just wanted to voice my concern.

 

 

Link to comment
Share on other sites

maybe its nothing and theres nothing wrong and im just salty but i also know technology doesnt always do what its supposed to do. so on the off chance that is something wrong, i just wanted to voice my concern.

 

Mate, i feel you.

But unfortunately not much can be said as the members of this community already had numerous fights over this concern the previous months.

 

The only thing that you should keep in mind is that "The RNG works just fine" and that "if you think otherwise, then prove it".

 

So in order to save your nerves, just play the game as it and don't get into any unnecessary fights.

Just for the record, i'm a RNG-doubter.

Link to comment
Share on other sites

I wish I had time to pull up a champion deck and work out how many "perfect," "good," "bad," or "awful" potential draws there are in it, but my tech time is focused elsewhere, and there IS a lot of math involved.

 

Yes, I'm seeing a lot of bad hands, but even if there is a real problem, it doesn't make sense for it to be favoring one player over another.

 

The worst I can imagine would be a randomizing error that was based on the initial sort of the decklist, exploited by someone who had noticed it and built a deck around it. I wouldn't believe that's an actual thing though, unless someone could give a full breakdown on exactly what was going on so that anyone could create that deck.

Link to comment
Share on other sites

Is it possible that the shuffling system is broken?

Not broken at all, it just doesn't follow anything that seems viable and fair enough in the real life. For example, having a deck with 20 basic pokemon should have every 3rd card as a possible basic pokemon. However, you can still get a mulligan which is quite possible. But the possibility of getting 4 mulligans in a row, now that doesn't ever happen in real life. But this is a day to day routine for me now LoL. This however by no means suggests RNG is broken. On other cases, you also get 7 basic pokemons in the first hand itself balancing the other 4 mulligans in the row. However, that doesn't happen in real world too often either. Its just that RNG on PTCGO is quite not viable enough to be compared to real life.

 

Just for the record, i'm a RNG-doubter.

RNG is indeed not broken (if you mean in terms of fairness, but if you mean in terms of real life scenario similarities, I can prove and guarantee you its far far away from what happens in real life )

Link to comment
Share on other sites

There has to be. I have at most the longest winning streak of three. After that I get all pokemon and no energy or all energy and no pokemon (and various trainer cards). And I'm not doing any errors on my part so I'm "easy to beat" by standards but not by my choosing. If you get a long game with me, and win or lose, I'm happy for a change no matter what. It means the computer did what it should have done in the first place. My longest losing streak? By the way, 15 or so, I've lost count actually.... computer deal like described.

Link to comment
Share on other sites

RNG is indeed not broken (if you mean in terms of fairness, but if you mean in terms of real life scenario similarities, I can prove and guarantee you its far far away from what happens in real life )

I'd like to see this proof. Make it a proof though, and not a stack of anecdotes.

Link to comment
Share on other sites

You arent complaining about RNG. You are wondering if there is ,atchfixing, which i know other games do use. I dont see a reason for pkmn to use it, but qhatever. Some will say you just played better people, but that doesnt matter if your hand sucked.

 

I feel like there are times where my opponent always flips heads or whatever. I get 4x tails amd lose the game over 3 turns, etc.

Link to comment
Share on other sites

I admire some people's passion to prove that PTCGO's RNG is as fundamental as is gravity, thus shouldn't be questioned.

 

In other news, even the most consistent decks will draw awful first hands just to be proven inconsistent.

Link to comment
Share on other sites

I'd like to see this proof. Make it a proof though, and not a stack of anecdotes.

Yeah sure, make a deck with 20 pokemon (all basics) --> You will definitely have a 4 mulligan in a row start after say 6-7 matches :) [its funny enough that its either a start, 1 mulligan or 4 mulligans always]

 

Or lets say how sometimes people get 6 heads in a row and then suddenly 6 tails in a row (The malamar EX special challenge a year and a half ago)

 

If you don't count this as proof fine enough, but I apologise I don't have enough time to provide statistics of over 100 games.

 

All this happening in real life has a very minute chance, but it happens too often on PTCGO !

Link to comment
Share on other sites

Yeah sure, make a deck with 20 pokemon (all basics) --> You will definitely have a 4 mulligan in a row start after say 6-7 matches :) [its funny enough that its either a start, 1 mulligan or 4 mulligans always]

 

Or lets say how sometimes people get 6 heads in a row and then suddenly 6 tails in a row (The malamar EX special challenge a year and a half ago)

 

If you don't count this as proof fine enough, but I apologise I don't have enough time to provide statistics of over 100 games.

 

All this happening in real life has a very minute chance, but it happens too often on PTCGO !

 

PTCGO's RNG is the elephant in the room that everyone refuses to notice publicly but keeps whispering about it.

Link to comment
Share on other sites

Yeah sure, make a deck with 20 pokemon (all basics) --> You will definitely have a 4 mulligan in a row start after say 6-7 matches :) [its funny enough that its either a start, 1 mulligan or 4 mulligans always]

 

Or lets say how sometimes people get 6 heads in a row and then suddenly 6 tails in a row (The malamar EX special challenge a year and a half ago)

 

If you don't count this as proof fine enough, but I apologise I don't have enough time to provide statistics of over 100 games.

 

All this happening in real life has a very minute chance, but it happens too often on PTCGO !

 

That isn't proof, no, it's an unfounded claim that isn't even backed up by statistics on what the actual result should be. Easy enough to test the claim though, so:

 

Match 1: Three basics

Match 2: Three basics

Match 3: Three basics

Match 4: Four basics

Match 5: Two basics

Match 6: One basic

Match 7: Three basics

Match 8: One basic

Match 9: Three basics

Match 10: One basic

Match 11: One basic

Match 12: Four basics

Match 13: Three basics

Match 14: Four basics

Match 15: Two basics

 

At this point, I'm going to skip to a mulligan...

 

Match 30: Revising the deck to include 25 non-basics...

 

Match 40: Three basics.

 

I haven't had 1 mulligan yet. So what are the odds supposed to be?

Link to comment
Share on other sites

Yeah sure, make a deck with 20 pokemon (all basics) --> You will definitely have a 4 mulligan in a row start after say 6-7 matches :) [its funny enough that its either a start, 1 mulligan or 4 mulligans always]

 

Or lets say how sometimes people get 6 heads in a row and then suddenly 6 tails in a row (The malamar EX special challenge a year and a half ago)

 

If you don't count this as proof fine enough, but I apologise I don't have enough time to provide statistics of over 100 games.

 

All this happening in real life has a very minute chance, but it happens too often on PTCGO !

Best. Either me or my opponent.

Mulligan x4+... finally set two basics onto field. Like what now you get two basics, when 4 or more times (I've gone past 7) you get your singleton supporter constantly.

Link to comment
Share on other sites

So working out the odds...

If 20 of 60 cards are basics, and we don't care about the others, the first card drawn has a .33 chance (out of 1) of being a basic. If it isn't, the odds get a little better for each consecutive draw of the initial 7, with them averaging about .35.

That means that for each of the 7 cards, the chance of it not being a basic if none of the others are basic is .65.

 

0.65^7=0.049

 

That looks to me like 5 in 100, or a 5% chance of a mulligan. Where is this math wrong?

 

Note I previously attempted to get a mulligan with this deck and stopped after not getting one 40 times in a row.

Link to comment
Share on other sites

So working out the odds...

If 20 of 60 cards are basics, and we don't care about the others, the first card drawn has a .33 chance (out of 1) of being a basic. If it isn't, the odds get a little better for each consecutive draw of the initial 7, with them averaging about .35.

That means that for each of the 7 cards, the chance of it not being a basic if none of the others are basic is .65.

 

0.65^7=0.049

 

That looks to me like 5 in 100, or a 5% chance of a mulligan. Where is this math wrong?

 

Note I previously attempted to get a mulligan with this deck and stopped after not getting one 40 times in a row.

Maybe I should apologise for saying I can prove, and I would have still stand by my word, if I had time. Surely maybe on weekend I will give you my stats and then we can ponder over it. Sorry for speaking out of my turn :(

Link to comment
Share on other sites

Maybe I should apologise for saying I can prove, and I would have still stand by my word, if I had time. Surely maybe on weekend I will give you my stats and then we can ponder over it. Sorry for speaking out of my turn :(

Well, I'm not going to say there definitely isn't a problem, I'd just like to see real evidence of it.

What we see now is limited to people complaining when they have what could well be be bad luck, and not saying anything at all when they don't have that bad luck.

 

Also, my test was done against trainers, so that I could start and quit a bunch of games without getting hit by the system they put in place to detect genting. I don't have a 20-basic deck that would stand a chance of winning real versus matches, and I won't be going there if that doesn't change.

 

Also, if I don't round aggressively and average, that mulligan number should be 4.83%. It does drop quickly for multiple consecutive mulligans with that deck, but I've never played a deck that has that favorable a basic draw chance. One of my go-to decks has only 7 basics in it.

Link to comment
Share on other sites

If anyone is familiar/fluent with Unity programming we could settle this (in theory); I'd like to write a program that will simply read and parse the client-side data from the PTCGO client and filter the results into a user-friendly database (like csv).

 

Then raw data could be collected and compared to the theoretical probability of drawing a certain card where for example:

 

[staple Trainer] 4/60 = 15% (Nominal) ∆ ?/60 = ??% (Actual).

 

Unfortunately, I have neither Unity Plus/Pro or the time to write such a program. It may also be possible that such a program would violate the ToS of PTCGO (even if it is only reading output sent to an individual client).

Link to comment
Share on other sites

Or lets say how sometimes people get 6 heads in a row and then suddenly 6 tails in a row (The malamar EX special challenge a year and a half ago)

 

If you don't count this as proof fine enough, but I apologise I don't have enough time to provide statistics of over 100 games.

 

All this happening in real life has a very minute chance, but it happens too often on PTCGO !

I can say from experience that randomness is always perceived differently than what it actually is, mainly because people expect random values to be normalized. By this I mean that if I say something has a 1 in 3 chance of happening, people will expect it to happen every 3rd time or so, but that isn't truly random.

 

Something that usually surprises people is that getting 6 heads in a row has EXACTLY the same probability of happening as getting, for example, h, t, t, h, t, h but people will usually not notice a sequence like this one because it is varied but if you do the math, the probabilities are the same..

 

The fact that each coin flip is independent of the previous one means that no matter how many times you hit heads before, in true randomness the probability of hitting heads on the next throw is still the same as if you had thrown all

 

Another factor is that people will often run into an ocurrence that has a low probability of happening, then they will run into another occurrence that also has low probability of happening and think that something is wrong, without realizing that in most games with randomness there are a large number of different things that have low probabilities of happening and that there is a relatively high probability of any one of them happening and that it is likely that one of them will happen every so often.

 

These simple examples help to illustrate how different what we expect from randomness is from what true randomness is but the also apply to more complex cases like drawing multiple cards from a deck.

 

BTW, in real life clumping (when randomness generates similar values continuously instead of varying through the whole range) happens less in real life because we tend to pile the decks to prevent it from happening. Essentially, randomness in real life is less random than one from an RNG.

Link to comment
Share on other sites

I can say from experience that randomness is always perceived differently than what it actually is, mainly because people expect random values to be normalized. By this I mean that if I say something has a 1 in 3 chance of happening, people will expect it to happen every 3rd time or so, but that isn't truly random.

 

Something that usually surprises people is that getting 6 heads in a row has EXACTLY the same probability of happening as getting, for example, h, t, t, h, t, h but people will usually not notice a sequence like this one because it is varied but if you do the math, the probabilities are the same..

 

The fact that each coin flip is independent of the previous one means that no matter how many times you hit heads before, in true randomness the probability of hitting heads on the next throw is still the same as if you had thrown all

 

Another factor is that people will often run into an ocurrence that has a low probability of happening, then they will run into another occurrence that also has low probability of happening and think that something is wrong, without realizing that in most games with randomness there are a large number of different things that have low probabilities of happening and that there is a relatively high probability of any one of them happening and that it is likely that one of them will happen every so often.

 

These simple examples help to illustrate how different what we expect from randomness is from what true randomness is but the also apply to more complex cases like drawing multiple cards from a deck.

 

BTW, in real life clumping (when randomness generates similar values continuously instead of varying through the whole range) happens less in real life because we tend to pile the decks to prevent it from happening. Essentially, randomness in real life is less random than one from an RNG.

You are mixing the fact of getting an outcome of a single flip with getting an outcome with the probability of getting similar outcomes over and over again.

 

For example, in a single flip the probability to get heads is 0.5, so is for the next flip.

 

But when you say, probability of getting 2 head simultaneously, the probability is 0.25 and is not at all 0.5

The easiest and simplest way you can understand this is by maybe drawing a sample space for all the outcomes of the 2 flips of a coin.

 

However, for each singular flip, probability of getting a coin is 0.5, but when you ask about getting 2 in a row, probability changes to 0.25 and doesn't remain 0.5.

 

P.S --> Sample space for 2 flips --> {HH, HT, TH, TT}

 

I hope that clears the confusions.

 

Another thing it is quite right that many things that have smaller probabilities and one of them happens quite often which is as it should be, PTCGO, still does give a surprisingly different outcomes that don't happen in real world. And of course we know it can't be like real world, but at least it shouldn't be so absurd either that you get streaks of small probability doing one thing and then another streak of small probability doing the opposite thing making it exactly fair !

Link to comment
Share on other sites

Another thing it is quite right that many things that have smaller probabilities and one of them happens quite often which is as it should be, PTCGO, still does give a surprisingly different outcomes that don't happen in real world. And of course we know it can't be like real world, but at least it shouldn't be so absurd either that you get streaks of small probability doing one thing and then another streak of small probability doing the opposite thing making it exactly fair !

While there's something to that, you have to keep in mind how often it occurs.  If it's rare enough, then the statistics predict it.

Link to comment
Share on other sites

You are mixing the fact of getting an outcome of a single flip with getting an outcome with the probability of getting similar outcomes over and over again.

 

For example, in a single flip the probability to get heads is 0.5, so is for the next flip.

 

But when you say, probability of getting 2 head simultaneously, the probability is 0.25 and is not at all 0.5

The easiest and simplest way you can understand this is by maybe drawing a sample space for all the outcomes of the 2 flips of a coin.

 

However, for each singular flip, probability of getting a coin is 0.5, but when you ask about getting 2 in a row, probability changes to 0.25 and doesn't remain 0.5.

 

P.S --> Sample space for 2 flips --> {HH, HT, TH, TT}

 

I hope that clears the confusions.

 

Another thing it is quite right that many things that have smaller probabilities and one of them happens quite often which is as it should be, PTCGO, still does give a surprisingly different outcomes that don't happen in real world. And of course we know it can't be like real world, but at least it shouldn't be so absurd either that you get streaks of small probability doing one thing and then another streak of small probability doing the opposite thing making it exactly fair !

But what he said is correct. Getting a specific sequence has the same odds whether it be HHHHHH or HTTHTH.

 

I'd like to see someone actually examine the results of card draws in the game rather than just claim that PTCGO's results are not in line with physical cards over and over.

Link to comment
Share on other sites

But what he said is correct. Getting a specific sequence has the same odds whether it be HHHHHH or HTTHTH.

 

I'd like to see someone actually examine the results of card draws in the game rather than just claim that PTCGO's results are not in line with physical cards over and over.

Agreed.  All specific sequences have the same probability.  And unless you're running a deck based around cards like Hammers or Malamar-EX, the order is very important.

Link to comment
Share on other sites

But what he said is correct. Getting a specific sequence has the same odds whether it be HHHHHH or HTTHTH.

 

I'd like to see someone actually examine the results of card draws in the game rather than just claim that PTCGO's results are not in line with physical cards over and over.

The chance of getting 3 heads and 3 tails (if we disregard the sequence in which they occur like TH and HT is same for me if we consider a bracket of only 2 flips) is far more than getting 6 heads :/

 

It will be a sample space of 2^6 :/ Don't make me do that.

 

However, since you used the word specific, I am assuming you are taking in the account that HTTHTH is not same as HHHTTT ? In that case yes, each sequence is same, but if overall account is taken into consideration 6 heads is far more rare to come.

 

I am however, not saying RNG is not fair or something it just doesn't replicate anything nearly in life (which is quite difficult to do :/ )

 

As for the real statistics, I will surely do so and let you all the raw data, but.. but.. oh, I don't have enough time :( Sorry for disappointing again :/

 

Agreed.  All specific sequences have the same probability.  And unless you're running a deck based around cards like Hammers or Malamar-EX, the order is very important.

From Malamar-EX around 1 and half year ago, I don't know if you know about it --> we were given a special challenge to flip 100 heads (special challenge).

 

I do think the intention behind it was to see if flips are going correctly or not. The results were quite obvious, the RNG was rather exactly fair without much variation. Like what I had done was, use a deck with Malamar-EX and 59 dark energies :P and with the help of Vs mode (chat was on, asked for opponent to help me)

 

Result back then even went to like 10 flips in a row occurring with same outcome and then followed by opposite outcome streak, making it 50-50. It can be confirmed by any old players who know about it.

 

Now if I am annoying everyone with my not-so-filled-with-statistics-post, then sorry :P

Link to comment
Share on other sites

The chance of getting 3 heads and 3 tails (if we disregard the sequence in which they occur like TH and HT is same for me if we consider a bracket of only 2 flips) is far more than getting 6 heads :/

 

 

That's the thing. In practice you can't disregard the sequence and order because it's part of the equation. That's why any given sequence has exactly the same probability of happening. Indeed if you just take the odds of getting a certain number of of a particular result there will be different probabilities but that is not what I'm referring to.

 

I'll give a better example. If you get H H H H H H, then get H H H H H H, that has the same probability as getting, in this exact order, T T H T H H and then again getting T T H T H H. However, the all heads pattern is much easier to detect than the second pattern, because the second pattern has variations. So even though randomness generates things that would be considered anomalies like that all the time, we only notice the ones that our brains can detect as a noticeable pattern.

 

Due to our own biases, we also tend to notice more when these patterns have a negative outcome, so it is natural that in a truly random environment, whenever there is a interesting pattern that has a negative outcome, we notice it, and if a few of these patterns occur in sequence, which is surprisingly probable, then our perception seems to go agains what we expect form how we understand probabilities work.

 

Basically, the human mind has a different expectation from randomness than what randomness truly results in. That's why every single game with any amount of randomness in it always has people in their forums asking if there is some kind of problem with the RNG. 

Link to comment
Share on other sites

 

I am however, not saying RNG is not fair or something it just doesn't replicate anything nearly in life (which is quite difficult to do :/ )

 

 

The problem is that in real life we don't use anything close to true randomness (which an easily implemented RNG does approach) because we tend to normalize the results by doing things like pile shuffling the different types of cards so that we have a more normalized  result with less clumping.

 

It has been debated that online games could benefit from having a normalized RNG instead of a fully random generator, but I've yet to see an online CCG that actually implements that.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...