**WINTER OF WOE - BONUS OBJECTIVE POINT**
As previously announced, the team will be distributing an additional point toward milestones to anyone who completed the Absorbing Man fight in the first step of the Winter of Woe.
This point will be distributed at a later time as it requires the team to pull and analyze data.
The timeline has not been set, but work has started.
There is currently an issue where some Alliances are are unable to find a match in Alliance Wars, or are receiving Byes without getting the benefits of the Win. We will be adjusting the Season Points of the Alliances that are affected within the coming weeks, and will be working to compensate them for their missed Per War rewards as well.

Additionally, we are working to address an issue where new Members of an Alliance are unable to place Defenders for the next War after joining. We are working to address this, but it will require a future update.

So, I decided to test the RNG on ST's persistent charges 100 times.

1246

Comments

  • DNA3000DNA3000 Posts: 18,558 Guardian
    DNA3000 wrote: »
    Incidentally, there's actually another possibility besides "random, equidistributed" and "random, weighted." And that is "random, constrained." That possibility is exemplified by Void's debuffs. They come up randomly, but they are constrained by a cap of two on each type. So the game picks randomly, but only from the set of constrained possibilities that changes as the game progresses. The first two buffs are random and all equally likely. But if you get two of the same, that possibility becomes excluded.

    That's also reasonably described as random, but with that constraint.

    Really all anyone here is asking for actual visibility into which calculations are used. Which pRNG method is actually implemented.

    What's being discussed is not pRNG method. What's being discussed is weighting. The game doesn't document how the possibilities are weighted. That's where there is general agreement. But that is completely separate from whether any other statement about how the game functions is true or false. False statements to support a generally agreed upon proposition are still false. In fact, they weaken the argument in support of it.
  • mostlyharmlessnmostlyharmlessn Posts: 1,387 ★★★★
    DNA3000 wrote: »
    DNA3000 wrote: »
    DNA3000 wrote: »
    "Random: made, done, happening, or chosen without method or conscious decision."

    If there was a method; i.e. 3-50% 4-30% 5-15% 6-5% then it wouldn't be random.

    False. Craps, for example, presumes random dice rolls but not every number has an equal chance of occurring because the combination of two dies generates a non-equal distribution. Two and Twelve only come up once every thirty six rolls on average while seven comes up six times more often.

    Also important to note that dictionaries are not the arbiter of the meanings of words, by their own mission statement. They document how words are used, they do not mandate how words must be used. The word "random" is used in a lot of contexts; in colloquial speech it might sometimes mean what you quote above, but when it comes to random number generation in computer systems that definition is entirely wrong.

    Craps is a bad analogy here. This is a single 4 sided die being thrown.

    No, it isn't. Hamin very conclusively proves that isn't what's happening, nor is it what the description is saying is happening.

    Having worked with these kinds of implementations in the past, this sort of thing usually gets implemented in one of three ways. The first way is with a partition. You generate a random float from zero to one, and then you state in the ability design that 0 <= r < 0.5 is three, 0.5 <= r < 0.8 is four, 0.8 <= r < 0.95 is five, and 0.95 <= r <= 1 is six. The second way is with multiple random triggers. You design the ability so that there's a 100% chance to get three charges, and then there's a 30% chance to get one more, a 20% chance to get one more, and a 10% chance to get one more (for example: these numbers don't match Hamin's tests). So there would be four possibilities generated randomly: three, four, five, and six depending on how many of those triggers fire. The third way is computationally. You generate a random number and then you map that random range to the target range with a formula. For example, r = random(0,1); c = 3 + RoundUp(2.5 * (r ^4)) (note: also an example: doesn't fit data).

    All of these are common ways for games to map random numbers to output ranges. All of them would be described by most games as "chosen randomly."

    We don't do this in other parts of the game. Either you get a critical hit or you don't, but we don't assume the game flips a coin. We know, or at least most game players know, that critical hits happen on a percentage chance basis, which normally means something similar to some specific chance out of 100 (or if we're playing a PnP game, usually some chance out of twenty).

    Based on the description, it is a 4 sided die.

    No matter how many times people might say this, it simply isn't true. The description is there for all to see:
    At the start of the fight, if Sabretooth has 0 persistent charges, he will randomly receive between 3 and 6 Persistent Charges.

    That could be 1d4+2. That could also be three coin flips, essentially 3d2. It could also be any other method of partitioning, since the method isn't specified.

    To repeat an example that shouldn't have to be repeated, but clearly it must be, in a game of craps the dice randomly generate a number between two and twelve. That is a true statement without qualification (provided you believe single dies are actually random).

    "Random" is not synonymous with "equidistributed." Anyone who thinks otherwise is just plain wrong, and will find themselves being wrong over and over again in every other game they play, and in life in general.

    However in a game of craps you know how many chances you have each number to be rolled.

    2 1 2.78%
    3 2 5.56%
    4 3 8.33%
    5 4 11.11%
    6 5 13.89%
    7 6 16.67%
    8 5 13.89%
    9 4 11.11%
    10 3 8.33%
    11 2 5.56%
    12 1 2.78%

    If it's not 1 in 4 chance then the description should state this.
  • DrZolaDrZola Posts: 8,480 ★★★★★
    Let's be honest here.

    Kabam can't even reliably get variable name to value substitution working right for in-game display. We're still seeing MS_ID_UI all over the place, so if that functionally isn't working correctly then how we can we trust other basic functions of the game are actually working right.

    ^^^This, so much this.

    5q1r2a3r1v6g.png


    Dr. Zola
  • DNA3000DNA3000 Posts: 18,558 Guardian
    Let's be honest here.

    Kabam can't even reliably get variable name to value substitution working right for in-game display. We're still seeing MS_ID_UI all over the place, so if that functionally isn't working correctly then how we can we trust other basic functions of the game are actually working right.

    If you don't trust anything, there's nothing for you to discuss. The OP isn't taking that position. The OP saw that the ability description stated that the charges were random, but didn't specify how likely each possibility was. So the OP tested it, and discovered that within statistical margin for error they could not be distributed evenly. So he concludes the description is incomplete: it says the possibilities are random and his tests show that, but they aren't equally likely and the description should probably state that also.

    That's all entirely reasonable, if your goal is to improve the game. The OP isn't saying the odds *should be* identical and they aren't so the game is bugged. He's saying the odds aren't equal, and that should be properly documented. Others are stating that the description implies or outright states the possibilities are equal, and that's false. It takes nothing away from the OP's work to inform everyone else what those odds actually appear to be.
  • DrZolaDrZola Posts: 8,480 ★★★★★
    DNA3000 wrote: »
    DNA3000 wrote: »
    DNA3000 wrote: »
    "Random: made, done, happening, or chosen without method or conscious decision."

    If there was a method; i.e. 3-50% 4-30% 5-15% 6-5% then it wouldn't be random.

    False. Craps, for example, presumes random dice rolls but not every number has an equal chance of occurring because the combination of two dies generates a non-equal distribution. Two and Twelve only come up once every thirty six rolls on average while seven comes up six times more often.

    Also important to note that dictionaries are not the arbiter of the meanings of words, by their own mission statement. They document how words are used, they do not mandate how words must be used. The word "random" is used in a lot of contexts; in colloquial speech it might sometimes mean what you quote above, but when it comes to random number generation in computer systems that definition is entirely wrong.

    Craps is a bad analogy here. This is a single 4 sided die being thrown.

    No, it isn't. Hamin very conclusively proves that isn't what's happening, nor is it what the description is saying is happening.

    Having worked with these kinds of implementations in the past, this sort of thing usually gets implemented in one of three ways. The first way is with a partition. You generate a random float from zero to one, and then you state in the ability design that 0 <= r < 0.5 is three, 0.5 <= r < 0.8 is four, 0.8 <= r < 0.95 is five, and 0.95 <= r <= 1 is six. The second way is with multiple random triggers. You design the ability so that there's a 100% chance to get three charges, and then there's a 30% chance to get one more, a 20% chance to get one more, and a 10% chance to get one more (for example: these numbers don't match Hamin's tests). So there would be four possibilities generated randomly: three, four, five, and six depending on how many of those triggers fire. The third way is computationally. You generate a random number and then you map that random range to the target range with a formula. For example, r = random(0,1); c = 3 + RoundUp(2.5 * (r ^4)) (note: also an example: doesn't fit data).

    All of these are common ways for games to map random numbers to output ranges. All of them would be described by most games as "chosen randomly."

    We don't do this in other parts of the game. Either you get a critical hit or you don't, but we don't assume the game flips a coin. We know, or at least most game players know, that critical hits happen on a percentage chance basis, which normally means something similar to some specific chance out of 100 (or if we're playing a PnP game, usually some chance out of twenty).

    Based on the description, it is a 4 sided die.

    No matter how many times people might say this, it simply isn't true. The description is there for all to see:
    At the start of the fight, if Sabretooth has 0 persistent charges, he will randomly receive between 3 and 6 Persistent Charges.

    That could be 1d4+2. That could also be three coin flips, essentially 3d2. It could also be any other method of partitioning, since the method isn't specified.

    To repeat an example that shouldn't have to be repeated, but clearly it must be, in a game of craps the dice randomly generate a number between two and twelve. That is a true statement without qualification (provided you believe single dies are actually random).

    "Random" is not synonymous with "equidistributed." Anyone who thinks otherwise is just plain wrong, and will find themselves being wrong over and over again in every other game they play, and in life in general.

    The point, however, is that “random” doesn’t necessarily mean not “equidistributed” either. It means random, and random with an equal chance and random with unequal chances and random with constraints are still subject to probability. Which one is it? Or is it totally something else? The answer, it appears, is play the game and find out and fend for yourself.

    Dr. Zola
  • DNA3000DNA3000 Posts: 18,558 Guardian
    DNA3000 wrote: »
    DNA3000 wrote: »
    DNA3000 wrote: »
    "Random: made, done, happening, or chosen without method or conscious decision."

    If there was a method; i.e. 3-50% 4-30% 5-15% 6-5% then it wouldn't be random.

    False. Craps, for example, presumes random dice rolls but not every number has an equal chance of occurring because the combination of two dies generates a non-equal distribution. Two and Twelve only come up once every thirty six rolls on average while seven comes up six times more often.

    Also important to note that dictionaries are not the arbiter of the meanings of words, by their own mission statement. They document how words are used, they do not mandate how words must be used. The word "random" is used in a lot of contexts; in colloquial speech it might sometimes mean what you quote above, but when it comes to random number generation in computer systems that definition is entirely wrong.

    Craps is a bad analogy here. This is a single 4 sided die being thrown.

    No, it isn't. Hamin very conclusively proves that isn't what's happening, nor is it what the description is saying is happening.

    Having worked with these kinds of implementations in the past, this sort of thing usually gets implemented in one of three ways. The first way is with a partition. You generate a random float from zero to one, and then you state in the ability design that 0 <= r < 0.5 is three, 0.5 <= r < 0.8 is four, 0.8 <= r < 0.95 is five, and 0.95 <= r <= 1 is six. The second way is with multiple random triggers. You design the ability so that there's a 100% chance to get three charges, and then there's a 30% chance to get one more, a 20% chance to get one more, and a 10% chance to get one more (for example: these numbers don't match Hamin's tests). So there would be four possibilities generated randomly: three, four, five, and six depending on how many of those triggers fire. The third way is computationally. You generate a random number and then you map that random range to the target range with a formula. For example, r = random(0,1); c = 3 + RoundUp(2.5 * (r ^4)) (note: also an example: doesn't fit data).

    All of these are common ways for games to map random numbers to output ranges. All of them would be described by most games as "chosen randomly."

    We don't do this in other parts of the game. Either you get a critical hit or you don't, but we don't assume the game flips a coin. We know, or at least most game players know, that critical hits happen on a percentage chance basis, which normally means something similar to some specific chance out of 100 (or if we're playing a PnP game, usually some chance out of twenty).

    Based on the description, it is a 4 sided die.

    No matter how many times people might say this, it simply isn't true. The description is there for all to see:
    At the start of the fight, if Sabretooth has 0 persistent charges, he will randomly receive between 3 and 6 Persistent Charges.

    That could be 1d4+2. That could also be three coin flips, essentially 3d2. It could also be any other method of partitioning, since the method isn't specified.

    To repeat an example that shouldn't have to be repeated, but clearly it must be, in a game of craps the dice randomly generate a number between two and twelve. That is a true statement without qualification (provided you believe single dies are actually random).

    "Random" is not synonymous with "equidistributed." Anyone who thinks otherwise is just plain wrong, and will find themselves being wrong over and over again in every other game they play, and in life in general.

    However in a game of craps you know how many chances you have each number to be rolled.

    2 1 2.78%
    3 2 5.56%
    4 3 8.33%
    5 4 11.11%
    6 5 13.89%
    7 6 16.67%
    8 5 13.89%
    9 4 11.11%
    10 3 8.33%
    11 2 5.56%
    12 1 2.78%

    If it's not 1 in 4 chance then the description should state this.

    That's entirely irrelevant. The statement "the dice generate a random number between two and twelve" is accurate, and thus the statement that Sabretooth randomly gets between 3 and 6 charges is also accurate. If one implies equal odds, so does the other. They are either both right or both wrong. The fact that you have additional information not provided in the statement that allows you to calculate the actual odds of each possibility says nothing about whether either statement is true or false, or implies equal odds or not.

    Most of us agree that if the odds are not equal the description should state that for completeness. But that's different from saying silence implies equal. It does not.
  • CoatHang3rCoatHang3r Posts: 4,965 ★★★★★
    MattMan wrote: »
    I’m sure they can just rewrite the description to state “Likey 3 furies”. That seems to be the accepted practice now
    The irony of mocking description rewrites in a thread asking for description rewrites.
    DNA3000 wrote: »
    DNA3000 wrote: »
    DNA3000 wrote: »
    "Random: made, done, happening, or chosen without method or conscious decision."

    If there was a method; i.e. 3-50% 4-30% 5-15% 6-5% then it wouldn't be random.

    False. Craps, for example, presumes random dice rolls but not every number has an equal chance of occurring because the combination of two dies generates a non-equal distribution. Two and Twelve only come up once every thirty six rolls on average while seven comes up six times more often.

    Also important to note that dictionaries are not the arbiter of the meanings of words, by their own mission statement. They document how words are used, they do not mandate how words must be used. The word "random" is used in a lot of contexts; in colloquial speech it might sometimes mean what you quote above, but when it comes to random number generation in computer systems that definition is entirely wrong.

    Craps is a bad analogy here. This is a single 4 sided die being thrown.

    No, it isn't. Hamin very conclusively proves that isn't what's happening, nor is it what the description is saying is happening.

    Having worked with these kinds of implementations in the past, this sort of thing usually gets implemented in one of three ways. The first way is with a partition. You generate a random float from zero to one, and then you state in the ability design that 0 <= r < 0.5 is three, 0.5 <= r < 0.8 is four, 0.8 <= r < 0.95 is five, and 0.95 <= r <= 1 is six. The second way is with multiple random triggers. You design the ability so that there's a 100% chance to get three charges, and then there's a 30% chance to get one more, a 20% chance to get one more, and a 10% chance to get one more (for example: these numbers don't match Hamin's tests). So there would be four possibilities generated randomly: three, four, five, and six depending on how many of those triggers fire. The third way is computationally. You generate a random number and then you map that random range to the target range with a formula. For example, r = random(0,1); c = 3 + RoundUp(2.5 * (r ^4)) (note: also an example: doesn't fit data).

    All of these are common ways for games to map random numbers to output ranges. All of them would be described by most games as "chosen randomly."

    We don't do this in other parts of the game. Either you get a critical hit or you don't, but we don't assume the game flips a coin. We know, or at least most game players know, that critical hits happen on a percentage chance basis, which normally means something similar to some specific chance out of 100 (or if we're playing a PnP game, usually some chance out of twenty).

    Based on the description, it is a 4 sided die.

    No matter how many times people might say this, it simply isn't true. The description is there for all to see:
    At the start of the fight, if Sabretooth has 0 persistent charges, he will randomly receive between 3 and 6 Persistent Charges.

    That could be 1d4+2. That could also be three coin flips, essentially 3d2. It could also be any other method of partitioning, since the method isn't specified.

    To repeat an example that shouldn't have to be repeated, but clearly it must be, in a game of craps the dice randomly generate a number between two and twelve. That is a true statement without qualification (provided you believe single dies are actually random).

    "Random" is not synonymous with "equidistributed." Anyone who thinks otherwise is just plain wrong, and will find themselves being wrong over and over again in every other game they play, and in life in general.

    However in a game of craps you know how many chances you have each number to be rolled.

    2 1 2.78%
    3 2 5.56%
    4 3 8.33%
    5 4 11.11%
    6 5 13.89%
    7 6 16.67%
    8 5 13.89%
    9 4 11.11%
    10 3 8.33%
    11 2 5.56%
    12 1 2.78%

    If it's not 1 in 4 chance then the description should state this.
    1) Not equatable to Casino’s as MCOC is not under the jurisdiction of or related to Gambling (which is regulated).

    2) In areguendo. But do Casions display those odds on the dice or tables? No, that information is gained outside of the game very much like this thread has done with ST’s odds.
  • BigPoppaCBONEBigPoppaCBONE Posts: 2,262 ★★★★★
    MattMan wrote: »
    I’m sure they can just rewrite the description to state “Likey 3 furies”. That seems to be the accepted practice now

    LOL! That’ll go over like a lead balloon, when the community sees something like, “Randomly 3 to 6, but probably 3”. They should have saved some hassle and just made the options equally likely.
  • mostlyharmlessnmostlyharmlessn Posts: 1,387 ★★★★
    DNA3000 wrote: »
    Let's be honest here.

    Kabam can't even reliably get variable name to value substitution working right for in-game display. We're still seeing MS_ID_UI all over the place, so if that functionally isn't working correctly then how we can we trust other basic functions of the game are actually working right.

    If you don't trust anything, there's nothing for you to discuss. The OP isn't taking that position. The OP saw that the ability description stated that the charges were random, but didn't specify how likely each possibility was. So the OP tested it, and discovered that within statistical margin for error they could not be distributed evenly. So he concludes the description is incomplete: it says the possibilities are random and his tests show that, but they aren't equally likely and the description should probably state that also.

    That's all entirely reasonable, if your goal is to improve the game. The OP isn't saying the odds *should be* identical and they aren't so the game is bugged. He's saying the odds aren't equal, and that should be properly documented. Others are stating that the description implies or outright states the possibilities are equal, and that's false. It takes nothing away from the OP's work to inform everyone else what those odds actually appear to be.

    The updated and corrected description is what we're asking for.
    CoatHang3r wrote: »
    MattMan wrote: »
    I’m sure they can just rewrite the description to state “Likey 3 furies”. That seems to be the accepted practice now
    The irony of mocking description rewrites in a thread asking for description rewrites.
    DNA3000 wrote: »
    DNA3000 wrote: »
    DNA3000 wrote: »
    "Random: made, done, happening, or chosen without method or conscious decision."

    If there was a method; i.e. 3-50% 4-30% 5-15% 6-5% then it wouldn't be random.

    False. Craps, for example, presumes random dice rolls but not every number has an equal chance of occurring because the combination of two dies generates a non-equal distribution. Two and Twelve only come up once every thirty six rolls on average while seven comes up six times more often.

    Also important to note that dictionaries are not the arbiter of the meanings of words, by their own mission statement. They document how words are used, they do not mandate how words must be used. The word "random" is used in a lot of contexts; in colloquial speech it might sometimes mean what you quote above, but when it comes to random number generation in computer systems that definition is entirely wrong.

    Craps is a bad analogy here. This is a single 4 sided die being thrown.

    No, it isn't. Hamin very conclusively proves that isn't what's happening, nor is it what the description is saying is happening.

    Having worked with these kinds of implementations in the past, this sort of thing usually gets implemented in one of three ways. The first way is with a partition. You generate a random float from zero to one, and then you state in the ability design that 0 <= r < 0.5 is three, 0.5 <= r < 0.8 is four, 0.8 <= r < 0.95 is five, and 0.95 <= r <= 1 is six. The second way is with multiple random triggers. You design the ability so that there's a 100% chance to get three charges, and then there's a 30% chance to get one more, a 20% chance to get one more, and a 10% chance to get one more (for example: these numbers don't match Hamin's tests). So there would be four possibilities generated randomly: three, four, five, and six depending on how many of those triggers fire. The third way is computationally. You generate a random number and then you map that random range to the target range with a formula. For example, r = random(0,1); c = 3 + RoundUp(2.5 * (r ^4)) (note: also an example: doesn't fit data).

    All of these are common ways for games to map random numbers to output ranges. All of them would be described by most games as "chosen randomly."

    We don't do this in other parts of the game. Either you get a critical hit or you don't, but we don't assume the game flips a coin. We know, or at least most game players know, that critical hits happen on a percentage chance basis, which normally means something similar to some specific chance out of 100 (or if we're playing a PnP game, usually some chance out of twenty).

    Based on the description, it is a 4 sided die.

    No matter how many times people might say this, it simply isn't true. The description is there for all to see:
    At the start of the fight, if Sabretooth has 0 persistent charges, he will randomly receive between 3 and 6 Persistent Charges.

    That could be 1d4+2. That could also be three coin flips, essentially 3d2. It could also be any other method of partitioning, since the method isn't specified.

    To repeat an example that shouldn't have to be repeated, but clearly it must be, in a game of craps the dice randomly generate a number between two and twelve. That is a true statement without qualification (provided you believe single dies are actually random).

    "Random" is not synonymous with "equidistributed." Anyone who thinks otherwise is just plain wrong, and will find themselves being wrong over and over again in every other game they play, and in life in general.

    However in a game of craps you know how many chances you have each number to be rolled.

    2 1 2.78%
    3 2 5.56%
    4 3 8.33%
    5 4 11.11%
    6 5 13.89%
    7 6 16.67%
    8 5 13.89%
    9 4 11.11%
    10 3 8.33%
    11 2 5.56%
    12 1 2.78%

    If it's not 1 in 4 chance then the description should state this.
    1) Not equatable to Casino’s as MCOC is not under the jurisdiction of or related to Gambling (which is regulated).

    2) In areguendo. But do Casions display those odds on the dice or tables? No, that information is gained outside of the game very much like this thread has done with ST’s odds.

    Casino provide that information freely. That's all any one is asking for.

    I didn't equate the game to a game of craps, someone else did and using that same example there's know odds which with Kabam there is not.

  • V1PER1987V1PER1987 Posts: 3,474 ★★★★★
    CoatHang3r wrote: »
    MattMan wrote: »
    I’m sure they can just rewrite the description to state “Likey 3 furies”. That seems to be the accepted practice now
    The irony of mocking description rewrites in a thread asking for description rewrites.
    DNA3000 wrote: »
    DNA3000 wrote: »
    DNA3000 wrote: »
    "Random: made, done, happening, or chosen without method or conscious decision."

    If there was a method; i.e. 3-50% 4-30% 5-15% 6-5% then it wouldn't be random.

    False. Craps, for example, presumes random dice rolls but not every number has an equal chance of occurring because the combination of two dies generates a non-equal distribution. Two and Twelve only come up once every thirty six rolls on average while seven comes up six times more often.

    Also important to note that dictionaries are not the arbiter of the meanings of words, by their own mission statement. They document how words are used, they do not mandate how words must be used. The word "random" is used in a lot of contexts; in colloquial speech it might sometimes mean what you quote above, but when it comes to random number generation in computer systems that definition is entirely wrong.

    Craps is a bad analogy here. This is a single 4 sided die being thrown.

    No, it isn't. Hamin very conclusively proves that isn't what's happening, nor is it what the description is saying is happening.

    Having worked with these kinds of implementations in the past, this sort of thing usually gets implemented in one of three ways. The first way is with a partition. You generate a random float from zero to one, and then you state in the ability design that 0 <= r < 0.5 is three, 0.5 <= r < 0.8 is four, 0.8 <= r < 0.95 is five, and 0.95 <= r <= 1 is six. The second way is with multiple random triggers. You design the ability so that there's a 100% chance to get three charges, and then there's a 30% chance to get one more, a 20% chance to get one more, and a 10% chance to get one more (for example: these numbers don't match Hamin's tests). So there would be four possibilities generated randomly: three, four, five, and six depending on how many of those triggers fire. The third way is computationally. You generate a random number and then you map that random range to the target range with a formula. For example, r = random(0,1); c = 3 + RoundUp(2.5 * (r ^4)) (note: also an example: doesn't fit data).

    All of these are common ways for games to map random numbers to output ranges. All of them would be described by most games as "chosen randomly."

    We don't do this in other parts of the game. Either you get a critical hit or you don't, but we don't assume the game flips a coin. We know, or at least most game players know, that critical hits happen on a percentage chance basis, which normally means something similar to some specific chance out of 100 (or if we're playing a PnP game, usually some chance out of twenty).

    Based on the description, it is a 4 sided die.

    No matter how many times people might say this, it simply isn't true. The description is there for all to see:
    At the start of the fight, if Sabretooth has 0 persistent charges, he will randomly receive between 3 and 6 Persistent Charges.

    That could be 1d4+2. That could also be three coin flips, essentially 3d2. It could also be any other method of partitioning, since the method isn't specified.

    To repeat an example that shouldn't have to be repeated, but clearly it must be, in a game of craps the dice randomly generate a number between two and twelve. That is a true statement without qualification (provided you believe single dies are actually random).

    "Random" is not synonymous with "equidistributed." Anyone who thinks otherwise is just plain wrong, and will find themselves being wrong over and over again in every other game they play, and in life in general.

    However in a game of craps you know how many chances you have each number to be rolled.

    2 1 2.78%
    3 2 5.56%
    4 3 8.33%
    5 4 11.11%
    6 5 13.89%
    7 6 16.67%
    8 5 13.89%
    9 4 11.11%
    10 3 8.33%
    11 2 5.56%
    12 1 2.78%

    If it's not 1 in 4 chance then the description should state this.
    1) Not equatable to Casino’s as MCOC is not under the jurisdiction of or related to Gambling (which is regulated).

    2) In areguendo. But do Casions display those odds on the dice or tables? No, that information is gained outside of the game very much like this thread has done with ST’s odds.

    For #2 you can’t really equate people determining craps odds vs ST charges because they’re totally different. When playing craps, if you know simple math and the game, then you can calculate the odds before you even get to the table. For ST charges, OP had to actually go through 100 fights to find the distribution of odds. In craps you don’t have to throw the dice 100 times to determine odds and probability.
  • 1_ShuNeu_11_ShuNeu_1 Posts: 375 ★★★
    I was testing ST in ROL the other day and found i got 3 Persistent Charges five times in a row.
  • CoatHang3rCoatHang3r Posts: 4,965 ★★★★★
    DNA3000 wrote: »
    Let's be honest here.

    Kabam can't even reliably get variable name to value substitution working right for in-game display. We're still seeing MS_ID_UI all over the place, so if that functionally isn't working correctly then how we can we trust other basic functions of the game are actually working right.

    If you don't trust anything, there's nothing for you to discuss. The OP isn't taking that position. The OP saw that the ability description stated that the charges were random, but didn't specify how likely each possibility was. So the OP tested it, and discovered that within statistical margin for error they could not be distributed evenly. So he concludes the description is incomplete: it says the possibilities are random and his tests show that, but they aren't equally likely and the description should probably state that also.

    That's all entirely reasonable, if your goal is to improve the game. The OP isn't saying the odds *should be* identical and they aren't so the game is bugged. He's saying the odds aren't equal, and that should be properly documented. Others are stating that the description implies or outright states the possibilities are equal, and that's false. It takes nothing away from the OP's work to inform everyone else what those odds actually appear to be.

    The updated and corrected description is what we're asking for.
    CoatHang3r wrote: »
    MattMan wrote: »
    I’m sure they can just rewrite the description to state “Likey 3 furies”. That seems to be the accepted practice now
    The irony of mocking description rewrites in a thread asking for description rewrites.
    DNA3000 wrote: »
    DNA3000 wrote: »
    DNA3000 wrote: »
    "Random: made, done, happening, or chosen without method or conscious decision."

    If there was a method; i.e. 3-50% 4-30% 5-15% 6-5% then it wouldn't be random.

    False. Craps, for example, presumes random dice rolls but not every number has an equal chance of occurring because the combination of two dies generates a non-equal distribution. Two and Twelve only come up once every thirty six rolls on average while seven comes up six times more often.

    Also important to note that dictionaries are not the arbiter of the meanings of words, by their own mission statement. They document how words are used, they do not mandate how words must be used. The word "random" is used in a lot of contexts; in colloquial speech it might sometimes mean what you quote above, but when it comes to random number generation in computer systems that definition is entirely wrong.

    Craps is a bad analogy here. This is a single 4 sided die being thrown.

    No, it isn't. Hamin very conclusively proves that isn't what's happening, nor is it what the description is saying is happening.

    Having worked with these kinds of implementations in the past, this sort of thing usually gets implemented in one of three ways. The first way is with a partition. You generate a random float from zero to one, and then you state in the ability design that 0 <= r < 0.5 is three, 0.5 <= r < 0.8 is four, 0.8 <= r < 0.95 is five, and 0.95 <= r <= 1 is six. The second way is with multiple random triggers. You design the ability so that there's a 100% chance to get three charges, and then there's a 30% chance to get one more, a 20% chance to get one more, and a 10% chance to get one more (for example: these numbers don't match Hamin's tests). So there would be four possibilities generated randomly: three, four, five, and six depending on how many of those triggers fire. The third way is computationally. You generate a random number and then you map that random range to the target range with a formula. For example, r = random(0,1); c = 3 + RoundUp(2.5 * (r ^4)) (note: also an example: doesn't fit data).

    All of these are common ways for games to map random numbers to output ranges. All of them would be described by most games as "chosen randomly."

    We don't do this in other parts of the game. Either you get a critical hit or you don't, but we don't assume the game flips a coin. We know, or at least most game players know, that critical hits happen on a percentage chance basis, which normally means something similar to some specific chance out of 100 (or if we're playing a PnP game, usually some chance out of twenty).

    Based on the description, it is a 4 sided die.

    No matter how many times people might say this, it simply isn't true. The description is there for all to see:
    At the start of the fight, if Sabretooth has 0 persistent charges, he will randomly receive between 3 and 6 Persistent Charges.

    That could be 1d4+2. That could also be three coin flips, essentially 3d2. It could also be any other method of partitioning, since the method isn't specified.

    To repeat an example that shouldn't have to be repeated, but clearly it must be, in a game of craps the dice randomly generate a number between two and twelve. That is a true statement without qualification (provided you believe single dies are actually random).

    "Random" is not synonymous with "equidistributed." Anyone who thinks otherwise is just plain wrong, and will find themselves being wrong over and over again in every other game they play, and in life in general.

    However in a game of craps you know how many chances you have each number to be rolled.

    2 1 2.78%
    3 2 5.56%
    4 3 8.33%
    5 4 11.11%
    6 5 13.89%
    7 6 16.67%
    8 5 13.89%
    9 4 11.11%
    10 3 8.33%
    11 2 5.56%
    12 1 2.78%

    If it's not 1 in 4 chance then the description should state this.
    1) Not equatable to Casino’s as MCOC is not under the jurisdiction of or related to Gambling (which is regulated).

    2) In areguendo. But do Casions display those odds on the dice or tables? No, that information is gained outside of the game very much like this thread has done with ST’s odds.

    Casino provide that information freely. That's all any one is asking for.

    I didn't equate the game to a game of craps, someone else did and using that same example there's know odds which with Kabam there is not.
    Freely? Ha! They may provide it but if they do there is an extreme likelihood they are duing so due to regulation or under implied duress to bolster their image.

    That aside. The odds have been explained with craps/dice as an example not the descriptions which you have done independant of the orignal argument/explanation. And you’re right this is not a casino and as so equating craps to MCOC is a bad analogy.

    P.S. The way these characters and their nuances encrouge “community” and talk about the game’s mechanics could be part of why these are somewhat obscure; at the least it prolly encouages them to be somewhat obscure. To the people who are interested in ST’s charges they now have a place to discuss it, learn about it and make acquaintances while sharing thier thoughts; changing decriptions would hinder that activity. So I ask, at the end of the day is it actually a negative that ST’s decription does not have an additional paragraph included explaining “randomly 3-6”?

  • DNA3000DNA3000 Posts: 18,558 Guardian
    CoatHang3r wrote: »
    P.S. The way these characters and their nuances encrouge “community” and talk about the game’s mechanics could be part of why these are somewhat obscure; at the least it prolly encouages them to be somewhat obscure. To the people who are interested in ST’s charges they now have a place to discuss it, learn about it and make acquaintances while sharing thier thoughts; changing decriptions would hinder that activity. So I ask, at the end of the day is it actually a negative that ST’s decription does not have an additional paragraph included explaining “randomly 3-6”?

    I would say yes. I've been a part of game communities that had very lively discussions about how the underlying game worked. I've actively participated in some where I literally wrote the book on game mechanics. I don't think you need to deliberately obfuscate mechanics and details to create that environment: in fact in my experience the more the developers disclose, the more lively that kind of discussion becomes. Information breeds the desire for more information, because knowledge is built on other knowledge. When players know things, it creates the opportunity for them to know more things. When players feel they don't know anything very well, it disincentivizes learning anything, or it creates significant hurdles to learning anything.

    I've never seen a game where discussion about how the game worked was absent because everyone thought they knew everything so discussion was pointless. I've always seen a direct correlation between how open and transparent the developers were with game implementation details and how lively the discussion was surrounding those details.

    Take the recent mastery issues that came up recently. I believe if Kabam was more open about game mechanics and in return the game community had a more lively set of "quants" discussing and describing game mechanics for other players, that problem would have been discovered much sooner. Moreover, it would be the players themselves explaining to other players just exactly what happened, and why the implementation no longer matched the description due to the evolving game. And that issue would have been far less of a problem.

    Real, full, accurate, prioritized disclosure tends to get the players on your side. And when they are on your side, they are far more willing to work with you when the game evolves in ways that change the dynamics of the game in unavoidable ways.
  • BigPoppaCBONEBigPoppaCBONE Posts: 2,262 ★★★★★
    P.S. The way these characters and their nuances encrouge “community” and talk about the game’s mechanics could be part of why these are somewhat obscure; at the least it prolly encouages them to be somewhat obscure. To the people who are interested in ST’s charges they now have a place to discuss it, learn about it and make acquaintances while sharing thier thoughts; changing decriptions would hinder that activity. So I ask, at the end of the day is it actually a negative that ST’s decription does not have an additional paragraph included explaining “randomly 3-6”?

    It didn’t take long to see that he almost never got 6 charges. It might have saved an hour and a period of very minor annoyance.
  • DNA3000DNA3000 Posts: 18,558 Guardian
    DrZola wrote: »
    DNA3000 wrote: »
    DNA3000 wrote: »
    DNA3000 wrote: »
    "Random: made, done, happening, or chosen without method or conscious decision."

    If there was a method; i.e. 3-50% 4-30% 5-15% 6-5% then it wouldn't be random.

    False. Craps, for example, presumes random dice rolls but not every number has an equal chance of occurring because the combination of two dies generates a non-equal distribution. Two and Twelve only come up once every thirty six rolls on average while seven comes up six times more often.

    Also important to note that dictionaries are not the arbiter of the meanings of words, by their own mission statement. They document how words are used, they do not mandate how words must be used. The word "random" is used in a lot of contexts; in colloquial speech it might sometimes mean what you quote above, but when it comes to random number generation in computer systems that definition is entirely wrong.

    Craps is a bad analogy here. This is a single 4 sided die being thrown.

    No, it isn't. Hamin very conclusively proves that isn't what's happening, nor is it what the description is saying is happening.

    Having worked with these kinds of implementations in the past, this sort of thing usually gets implemented in one of three ways. The first way is with a partition. You generate a random float from zero to one, and then you state in the ability design that 0 <= r < 0.5 is three, 0.5 <= r < 0.8 is four, 0.8 <= r < 0.95 is five, and 0.95 <= r <= 1 is six. The second way is with multiple random triggers. You design the ability so that there's a 100% chance to get three charges, and then there's a 30% chance to get one more, a 20% chance to get one more, and a 10% chance to get one more (for example: these numbers don't match Hamin's tests). So there would be four possibilities generated randomly: three, four, five, and six depending on how many of those triggers fire. The third way is computationally. You generate a random number and then you map that random range to the target range with a formula. For example, r = random(0,1); c = 3 + RoundUp(2.5 * (r ^4)) (note: also an example: doesn't fit data).

    All of these are common ways for games to map random numbers to output ranges. All of them would be described by most games as "chosen randomly."

    We don't do this in other parts of the game. Either you get a critical hit or you don't, but we don't assume the game flips a coin. We know, or at least most game players know, that critical hits happen on a percentage chance basis, which normally means something similar to some specific chance out of 100 (or if we're playing a PnP game, usually some chance out of twenty).

    Based on the description, it is a 4 sided die.

    No matter how many times people might say this, it simply isn't true. The description is there for all to see:
    At the start of the fight, if Sabretooth has 0 persistent charges, he will randomly receive between 3 and 6 Persistent Charges.

    That could be 1d4+2. That could also be three coin flips, essentially 3d2. It could also be any other method of partitioning, since the method isn't specified.

    To repeat an example that shouldn't have to be repeated, but clearly it must be, in a game of craps the dice randomly generate a number between two and twelve. That is a true statement without qualification (provided you believe single dies are actually random).

    "Random" is not synonymous with "equidistributed." Anyone who thinks otherwise is just plain wrong, and will find themselves being wrong over and over again in every other game they play, and in life in general.

    The point, however, is that “random” doesn’t necessarily mean not “equidistributed” either. It means random, and random with an equal chance and random with unequal chances and random with constraints are still subject to probability. Which one is it? Or is it totally something else? The answer, it appears, is play the game and find out and fend for yourself.

    Dr. Zola

    Are you actually saying that "the point" to the statement "based on the description, it is a 4 sided die" is that we don't actually know if it is equally distributed or not? Because that seems to imply that the point to the statement I was replying to is the logical opposite of what the words state.

    "Based on the description, it is a 4 sided die" is either a true statement or a false statement. It is never besides the point to challenge a false statement when it contains a fundamental point of discussion. If that statement is true, the champion description is erroneous. If that statement is false, the champion description is incomplete. Those are two totally different things.
  • Odd that there is no mod presence here yet, my hat is off to the OP great detective work my friend.
  • HaminHamin Posts: 2,444 ★★★★★
    Odd that there is no mod presence here yet, my hat is off to the OP great detective work my friend.

    Thank you.

    And, they might be discussing how to best address the topic as to properly convey information.
  • DrZolaDrZola Posts: 8,480 ★★★★★
    edited February 2019
    DNA3000 wrote: »
    DrZola wrote: »
    DNA3000 wrote: »
    DNA3000 wrote: »
    DNA3000 wrote: »
    "Random: made, done, happening, or chosen without method or conscious decision."

    If there was a method; i.e. 3-50% 4-30% 5-15% 6-5% then it wouldn't be random.

    False. Craps, for example, presumes random dice rolls but not every number has an equal chance of occurring because the combination of two dies generates a non-equal distribution. Two and Twelve only come up once every thirty six rolls on average while seven comes up six times more often.

    Also important to note that dictionaries are not the arbiter of the meanings of words, by their own mission statement. They document how words are used, they do not mandate how words must be used. The word "random" is used in a lot of contexts; in colloquial speech it might sometimes mean what you quote above, but when it comes to random number generation in computer systems that definition is entirely wrong.

    Craps is a bad analogy here. This is a single 4 sided die being thrown.

    No, it isn't. Hamin very conclusively proves that isn't what's happening, nor is it what the description is saying is happening.

    Having worked with these kinds of implementations in the past, this sort of thing usually gets implemented in one of three ways. The first way is with a partition. You generate a random float from zero to one, and then you state in the ability design that 0 <= r < 0.5 is three, 0.5 <= r < 0.8 is four, 0.8 <= r < 0.95 is five, and 0.95 <= r <= 1 is six. The second way is with multiple random triggers. You design the ability so that there's a 100% chance to get three charges, and then there's a 30% chance to get one more, a 20% chance to get one more, and a 10% chance to get one more (for example: these numbers don't match Hamin's tests). So there would be four possibilities generated randomly: three, four, five, and six depending on how many of those triggers fire. The third way is computationally. You generate a random number and then you map that random range to the target range with a formula. For example, r = random(0,1); c = 3 + RoundUp(2.5 * (r ^4)) (note: also an example: doesn't fit data).

    All of these are common ways for games to map random numbers to output ranges. All of them would be described by most games as "chosen randomly."

    We don't do this in other parts of the game. Either you get a critical hit or you don't, but we don't assume the game flips a coin. We know, or at least most game players know, that critical hits happen on a percentage chance basis, which normally means something similar to some specific chance out of 100 (or if we're playing a PnP game, usually some chance out of twenty).

    Based on the description, it is a 4 sided die.

    No matter how many times people might say this, it simply isn't true. The description is there for all to see:
    At the start of the fight, if Sabretooth has 0 persistent charges, he will randomly receive between 3 and 6 Persistent Charges.

    That could be 1d4+2. That could also be three coin flips, essentially 3d2. It could also be any other method of partitioning, since the method isn't specified.

    To repeat an example that shouldn't have to be repeated, but clearly it must be, in a game of craps the dice randomly generate a number between two and twelve. That is a true statement without qualification (provided you believe single dies are actually random).

    "Random" is not synonymous with "equidistributed." Anyone who thinks otherwise is just plain wrong, and will find themselves being wrong over and over again in every other game they play, and in life in general.

    The point, however, is that “random” doesn’t necessarily mean not “equidistributed” either. It means random, and random with an equal chance and random with unequal chances and random with constraints are still subject to probability. Which one is it? Or is it totally something else? The answer, it appears, is play the game and find out and fend for yourself.

    Dr. Zola

    Are you actually saying that "the point" to the statement "based on the description, it is a 4 sided die" is that we don't actually know if it is equally distributed or not? Because that seems to imply that the point to the statement I was replying to is the logical opposite of what the words state.

    "Based on the description, it is a 4 sided die" is either a true statement or a false statement. It is never besides the point to challenge a false statement when it contains a fundamental point of discussion. If that statement is true, the champion description is erroneous. If that statement is false, the champion description is incomplete. Those are two totally different things.

    I didn’t think I was disagreeing with you.

    I simply think @mostlyharmlessn reads the description as if it implied a 4-sided die, which is a perfectly rational conclusion. The ST description isn’t a technical document; therefore, he concludes it means what the layman would likely conclude: it is a 4-sided die. The game isn’t reserved for statisticians and PhDs only; it carries a 12+ age requirement.

    If I had asked my company at my dinner table tonight (all of whom hold graduate degrees in scientific fields) what they think “random” means in this context, their initial responses would likely be the same as my yardman: something much more like “it is a 4-sided die” than it is random with constraints or it is distributed according to some predetermined percentage or something else.

    They might mention some caveats around their statement and their assumptions, but that would be the essence of their answer. Because that’s usually what an average person would conclude as it relates to a phone game.

    A champion’s description is written for players of the game who, according to the App Store, are supposed to be at least age 12+. Does it make sense to ask those players to understand the nuances of random number theory and probability in order to understand how a champ works? Probably not.

    So who is it incumbent upon to disclose clearly how a champ works?

    Dr. Zola
  • DrZola wrote: »
    I simply think @mostlyharmlessn reads the description as if it implied a 4-sided die, which is a perfectly rational conclusion. The ST description isn’t a technical document; therefore, he concludes it means what the layman would likely conclude: it is a 4-sided die. The game isn’t reserved for statisticians and PhDs only; it carries a 12+ age requirement.

    That's not really a fair characterization: that the "technical" definition is somehow radically different from the "normal" one. The thread has tons of examples where perfectly non-technical people would not jump to that conclusion. Critical hits are random. No one thinks they are 50/50. Craps rolls are random: I have no idea how many people think all the rolls come up evenly but it isn't a reasonable conclusion.

    In fact, it doesn't matter how likely it is that a "normal" person jumps to a conclusion. That's why we normally presume that jumping to a conclusion is bad, no matter how many people do it. What matters is whether that conclusion is a reasonable one when thought about carefully by reasonable people. A reasonable person might jump to the conclusion that the bigger cereal box has more cereal but they can't sue if it doesn't, because no one would agree that was a reasonable conclusion - even if a high percentage of reasonable people initially jump to that conclusion.

    When ever something is described as generating one of a set of options, it is *often* the case, in video games, in gaming in general, in *life* in general that those options aren't equally generated. Life is supposed to teach you that this is not a safe assumption to make. Someone making it isn't usually assuming it because their life experience tells them this is always true, but because they *feel* it *should* be true because of their own imposition of what they demand words mean. Even though the words don't mean that.

    Would a twelve year old believe that since "criticals are random" that they happen 50% of the time? I doubt it. And look at all of the people who post that since Kabam doesn't explicitly state the odds of a specific champion being dropped in a crystal are equal, that means they are hiding the fact they are not equal. Apparently saying something generates one of a number of results without explicitly stating the distribution makes it obvious the distribution is unequal. When it serves one's purposes.

    Here's something I would bet a lot of money on. If I took a hundred average game players with no technical background and across all age groups and I asked them this question: a game says it generates one of four possible options randomly, do you believe the game is explicitly trying to tell you that all four possibilities happen equally likely when it says this, I believe if I give them a reasonable amount of time to think about the question the majority of them would say no.

    The important thing about that question as worded is that it exposes the fact that there are *three* possibilities, not two. Possibility number one: we think the odds are equal. Possibility number two: we think the odds are unequal. Possibility number three: the correct one: we don't know, because we were not told.
  • BigPoppaCBONEBigPoppaCBONE Posts: 2,262 ★★★★★
    DNA3000 wrote: »
    Here's something I would bet a lot of money on. If I took a hundred average game players with no technical background and across all age groups and I asked them this question: a game says it generates one of four possible options randomly, do you believe the game is explicitly trying to tell you that all four possibilities happen equally likely when it says this, I believe if I give them a reasonable amount of time to think about the question the majority of them would say no.
    If that was the exact question you asked, after you explained that “explicitly” had nothing to do with nudity and gave an ELI5 example, they would say no because it wasn’t unequivocally spelled out that the probabilities were equal. If you had asked them to guess the distribution instead, absent other information, I believe the vast majority would guess them to be equal, because why wouldn’t they be equal?
  • CoatHang3rCoatHang3r Posts: 4,965 ★★★★★
    edited February 2019
    If you had asked them to guess the distribution instead, absent other information, I believe the vast majority would guess them to be equal, because why wouldn’t they be equal?
    Is 6 better than 5 than 4 than 3? Intuition tells us the more valuable of the set would be rarer than the next. This is shown almost daily by the claim crystals are rigged with “trash” in spite of people being repeatedly and explicitly told otherwise; and they still choose to believe the more valuable are weighted differently. Not buying it.

    Why would people inherently expect a more valuable outcome to have the same chance as a less valuble outcome unless otherwise stated?

  • HaminHamin Posts: 2,444 ★★★★★
    CoatHang3r wrote: »
    If you had asked them to guess the distribution instead, absent other information, I believe the vast majority would guess them to be equal, because why wouldn’t they be equal?
    Is 6 better than 5 than 4 than 3? Intuition tells us the more valuable of the set would be rarer than the next. This is shown almost daily by the claim crystals are rigged with “trash” in spite of people being repeatedly and explicitly told otherwise.l; and they still choose to believe the more valuable are weighted differently. Not buying it.

    Why would people inherently expect a more valuable outcome to have the same chance as a less valuble outcome unless otherwise stated?

    I agree. And... out of curiosity, I wonder if PM's missions are weighted? I find mission 1 and 3 happen first more often than 2.
  • CoatHang3rCoatHang3r Posts: 4,965 ★★★★★
    edited February 2019
    Hamin wrote: »
    CoatHang3r wrote: »
    If you had asked them to guess the distribution instead, absent other information, I believe the vast majority would guess them to be equal, because why wouldn’t they be equal?
    Is 6 better than 5 than 4 than 3? Intuition tells us the more valuable of the set would be rarer than the next. This is shown almost daily by the claim crystals are rigged with “trash” in spite of people being repeatedly and explicitly told otherwise.l; and they still choose to believe the more valuable are weighted differently. Not buying it.

    Why would people inherently expect a more valuable outcome to have the same chance as a less valuble outcome unless otherwise stated?

    I agree. And... out of curiosity, I wonder if PM's missions are weighted? I find mission 1 and 3 happen first more often than 2.
    I’ve noted nothing like that but haven’t tested it. I can say that from recollection there is no weight for any of her missions in my many experiences. Have you tested it?

    P.S. If it is evenly wieghted you’d expect 2 of the 3 to happen more than the 3rd. So your obervations would support equal distribution.
  • BigPoppaCBONEBigPoppaCBONE Posts: 2,262 ★★★★★
    CoatHang3r wrote: »
    If you had asked them to guess the distribution instead, absent other information, I believe the vast majority would guess them to be equal, because why wouldn’t they be equal?
    Is 6 better than 5 than 4 than 3? Intuition tells us the more valuable of the set would be rarer than the next. This is shown almost daily by the claim crystals are rigged with “trash” in spite of people being repeatedly and explicitly told otherwise; and they still choose to believe the more valuable are weighted differently. Not buying it.

    Why would people inherently expect a more valuable outcome to have the same chance as a less valuble outcome unless otherwise stated?

    In the example I responded to, they weren’t given value. They’re just possible options. I could point to champ crystals as an example where clearly some outcomes have more value than others, yet their probability is given as equal, and then ask the question. Given historical information on how Kabam gets down, they would probably assume equal probabilities because why wouldn’t they be equal, given no information that would lead you to infer otherwise.
  • DrZolaDrZola Posts: 8,480 ★★★★★
    DNA3000 wrote: »
    DrZola wrote: »
    I simply think @mostlyharmlessn reads the description as if it implied a 4-sided die, which is a perfectly rational conclusion. The ST description isn’t a technical document; therefore, he concludes it means what the layman would likely conclude: it is a 4-sided die. The game isn’t reserved for statisticians and PhDs only; it carries a 12+ age requirement.

    That's not really a fair characterization: that the "technical" definition is somehow radically different from the "normal" one. The thread has tons of examples where perfectly non-technical people would not jump to that conclusion. Critical hits are random. No one thinks they are 50/50. Craps rolls are random: I have no idea how many people think all the rolls come up evenly but it isn't a reasonable conclusion.

    In fact, it doesn't matter how likely it is that a "normal" person jumps to a conclusion. That's why we normally presume that jumping to a conclusion is bad, no matter how many people do it. What matters is whether that conclusion is a reasonable one when thought about carefully by reasonable people. A reasonable person might jump to the conclusion that the bigger cereal box has more cereal but they can't sue if it doesn't, because no one would agree that was a reasonable conclusion - even if a high percentage of reasonable people initially jump to that conclusion.

    When ever something is described as generating one of a set of options, it is *often* the case, in video games, in gaming in general, in *life* in general that those options aren't equally generated. Life is supposed to teach you that this is not a safe assumption to make. Someone making it isn't usually assuming it because their life experience tells them this is always true, but because they *feel* it *should* be true because of their own imposition of what they demand words mean. Even though the words don't mean that.

    Would a twelve year old believe that since "criticals are random" that they happen 50% of the time? I doubt it. And look at all of the people who post that since Kabam doesn't explicitly state the odds of a specific champion being dropped in a crystal are equal, that means they are hiding the fact they are not equal. Apparently saying something generates one of a number of results without explicitly stating the distribution makes it obvious the distribution is unequal. When it serves one's purposes.

    Here's something I would bet a lot of money on. If I took a hundred average game players with no technical background and across all age groups and I asked them this question: a game says it generates one of four possible options randomly, do you believe the game is explicitly trying to tell you that all four possibilities happen equally likely when it says this, I believe if I give them a reasonable amount of time to think about the question the majority of them would say no.

    The important thing about that question as worded is that it exposes the fact that there are *three* possibilities, not two. Possibility number one: we think the odds are equal. Possibility number two: we think the odds are unequal. Possibility number three: the correct one: we don't know, because we were not told.

    A few things:

    1) Stop with the craps example. We know what the odds are because we can see the dice and calculate the odds. We usually have no idea what the “dice” are in this game.
    2) No one said anything about “jumping” to a conclusion. You applied that label to people who reached the determination that “between 3 and 6 charges” meant an equal shot at each.
    3) Critical hits depend on a critical rate stat in game. It used to be a % but is now an explicit flat value for each champ. It’s perfectly reasonable to assume that a higher flat value, like a higher % previously, means a greater chance to get a critical hit. That’s an actual value listed in game—not sure why you think this is a logical analogy to a description that is utterly devoid of any numerical value.
    4) If you ask gamers and feed them a question worded in a certain way, they might agree that random is susceptible to multiple interpretations. They might even do that if you didn’t word it in a certain way. But I’m willing to bet if you told them you have 4 different names in a hat and you are going to pull one of them out at random, most of them don’t think you’ve actually got 45 Johns, 30 Bills, 15 Toms and 10 Freds in that hat.

    Dr. Zola
  • CoatHang3rCoatHang3r Posts: 4,965 ★★★★★
    CoatHang3r wrote: »
    If you had asked them to guess the distribution instead, absent other information, I believe the vast majority would guess them to be equal, because why wouldn’t they be equal?
    Is 6 better than 5 than 4 than 3? Intuition tells us the more valuable of the set would be rarer than the next. This is shown almost daily by the claim crystals are rigged with “trash” in spite of people being repeatedly and explicitly told otherwise; and they still choose to believe the more valuable are weighted differently. Not buying it.

    Why would people inherently expect a more valuable outcome to have the same chance as a less valuble outcome unless otherwise stated?

    In the example I responded to, they weren’t given value. They’re just possible options. I could point to champ crystals as an example where clearly some outcomes have more value than others, yet their probability is given as equal, and then ask the question. Given historical information on how Kabam gets down, they would probably assume equal probabilities because why wouldn’t they be equal, given no information that would lead you to infer otherwise.
    The question was building on STs 3-6 random chance (3<6) as being equivalent to a 4 sided die. All we need to do for your argument to work is to take it out of context like you did. So yeah, grats I guess?

    People quickly learn from observational experience that rarer prizes are less likely through the game; See Crystals of the following types, arena, AQ, gold, phc, ultimate, quest, and solo. Dunno where you are gathering your historical information from as this directly contradicts the assertion people would assume distribution of the most favorable outcomes are equal to the least based on “the way Kabam gets down.”
  • BigPoppaCBONEBigPoppaCBONE Posts: 2,262 ★★★★★
    CoatHang3r wrote: »
    CoatHang3r wrote: »
    If you had asked them to guess the distribution instead, absent other information, I believe the vast majority would guess them to be equal, because why wouldn’t they be equal?
    Is 6 better than 5 than 4 than 3? Intuition tells us the more valuable of the set would be rarer than the next. This is shown almost daily by the claim crystals are rigged with “trash” in spite of people being repeatedly and explicitly told otherwise; and they still choose to believe the more valuable are weighted differently. Not buying it.

    Why would people inherently expect a more valuable outcome to have the same chance as a less valuble outcome unless otherwise stated?

    In the example I responded to, they weren’t given value. They’re just possible options. I could point to champ crystals as an example where clearly some outcomes have more value than others, yet their probability is given as equal, and then ask the question. Given historical information on how Kabam gets down, they would probably assume equal probabilities because why wouldn’t they be equal, given no information that would lead you to infer otherwise.
    The question was building on STs 3-6 random chance (3<6) as being equivalent to a 4 sided die. All we need to do for your argument to work is to take it out of context like you did. So yeah, grats I guess?

    People quickly learn from observational experience that rarer prizes are less likely through the game; See Crystals of the following types, arena, AQ, gold, phc, ultimate, quest, and solo. Dunno where you are gathering your historical information from as this directly contradicts the assertion people would assume distribution of the most favorable outcomes are equal to the least based on “the way Kabam gets down.”

    I was responding to and building on the scenario posed in the post I quoted and not trying to sum up the thread with an analogy. Of course the more desirable outcomes are almost always given as less frequent and many people think they are less frequent even when given as equal. I know if there was a case where I’d get a trip to Hawaii if I pulled a Magneto, Colossus, or Vision in my next 3 PHCs, I’d tell my wife to get her bags ready given my experience.

    Really, they should have just given the odds and saved some hassle. Using him a handful of times would tell you that you’ll hardly ever get 6 charges. I would advise them to just save themselves future headaches and be annoyingly clear about things. They can’t even be cute like in Yondu’s abilities because people can’t handle it. Be clear and concise. It’s not worth it to chose to do otherwise unless there’s a very compelling reason to not spell things out clearly and if the odds you’ve chosen look terrible in print, that should tell you something.
  • Ultra8529Ultra8529 Posts: 526 ★★★
    Is there any reason Kabam should not amend the description to state the weighted probabilities of each number from 3 to 6? @DNA3000
  • BigPoppaCBONEBigPoppaCBONE Posts: 2,262 ★★★★★
    They could have said something like, “Sabretooth uses persistent charges to maintain his furies for the fights to come, generating at least 3, and as many as 6! He exhausts these charges 1 at a time to carryover his furies to the next fight and generates a new set when he reaches 0” if they didn’t want to print the unimpressive actual numbers while implying that 6 was highly desirable and less likely than the default 3.
  • iRetr0iRetr0 Posts: 1,252 ★★★★
    We do really need clarification here, would just like to say hats off to the OP for this!
Sign In or Register to comment.