AI advantage

24

Comments

  • jambattles
    jambattles Posts: 18 Just Dropped In
    edited September 2019
    IIAlonditeII said:
    jambattles said:
    After reading some responses about confirmation bias, I decided to keep up with some stats over my next 100 games. After keeping with the stats, there is zero question of AI advantage. It's a joke to consider it not a cheating pos algorithm. I get the disadvantages of the AI but to claim the AI isn't given an advantage is a lie.


    In the games I played and tracked (100 straight games):

    In PVE events:
    Human: strike, attack, protect, and count down tiles: when randomly placed, these tiles had a 91% chance of being placed for immediate match. Although, the tiles may or may not be matched was not considered only if they could be matched with 1 move. The chance of being destroyed through a match 4 increased the probability to 94%. 

    AI: strike, attack, protect and countdown tiles: when randomly placed, these tiles had a 7% chance of being placed for immediate match. 

    In PVP events: 
    Human: matching 3 or more colors: I did not track this.
    AI: matching 3 or more colors: the AI matched or accumulated more than 3 AP in 84% of turns. I was actually shocked it was this high. 

    I wish I had tracked color availability and how it adjusted through the match. Seems that when I need a color the color isn't available but when the AI needs it, it drops immediately. 
    These numbers are too wild, haha. You're going to need to evidence these claims, because they are far too extreme to not be fabricated. I'll need to see the total number of Strike, Attack and Protect tiles placed, and the number of them that were placed such that they could be immediately matched. And so on, for all of the following "statistics" you provided. 
    If you don't or do believe these numbers is your problem, don't make it mine. If you doubt them, do it for yourself. Come up with your own numbers and post them then I will call you a liar. Fair enough?
  • Sm0keyJ0e
    Sm0keyJ0e Posts: 730 Critical Contributor
    jambattles said:
    IIAlonditeII said:
    jambattles said:
    After reading some responses about confirmation bias, I decided to keep up with some stats over my next 100 games. After keeping with the stats, there is zero question of AI advantage. It's a joke to consider it not a cheating pos algorithm. I get the disadvantages of the AI but to claim the AI isn't given an advantage is a lie.


    In the games I played and tracked (100 straight games):

    In PVE events:
    Human: strike, attack, protect, and count down tiles: when randomly placed, these tiles had a 91% chance of being placed for immediate match. Although, the tiles may or may not be matched was not considered only if they could be matched with 1 move. The chance of being destroyed through a match 4 increased the probability to 94%. 

    AI: strike, attack, protect and countdown tiles: when randomly placed, these tiles had a 7% chance of being placed for immediate match. 

    In PVP events: 
    Human: matching 3 or more colors: I did not track this.
    AI: matching 3 or more colors: the AI matched or accumulated more than 3 AP in 84% of turns. I was actually shocked it was this high. 

    I wish I had tracked color availability and how it adjusted through the match. Seems that when I need a color the color isn't available but when the AI needs it, it drops immediately. 
    These numbers are too wild, haha. You're going to need to evidence these claims, because they are far too extreme to not be fabricated. I'll need to see the total number of Strike, Attack and Protect tiles placed, and the number of them that were placed such that they could be immediately matched. And so on, for all of the following "statistics" you provided. 
    If you don't or do believe these numbers is your problem, don't make it mine. If you doubt them, do it for yourself. Come up with your own numbers and post them then I will call you a liar. Fair enough?

    I've done my own unscientific observations and, in fact, the AI was just as likely to place a special tile that could be matched as I was. So, case closed. No cheating AI at play here. We can close this thread.
  • ThaRoadWarrior
    ThaRoadWarrior Posts: 9,387 Chairperson of the Boards
    edited September 2019
    I did have a match in Trial by Combat today where Main Event Hulk busted in on a kitty/rocket team and the animation played, but no tiles were destroyed at all as if it never happened. Like no effect on the board at all, not like i missed them all. That turned out to be fairly advantageous to the AI...
  • jambattles
    jambattles Posts: 18 Just Dropped In
    Sm0keyJ0e said:
    jambattles said:
    IIAlonditeII said:
    jambattles said:
    After reading some responses about confirmation bias, I decided to keep up with some stats over my next 100 games. After keeping with the stats, there is zero question of AI advantage. It's a joke to consider it not a cheating pos algorithm. I get the disadvantages of the AI but to claim the AI isn't given an advantage is a lie.


    In the games I played and tracked (100 straight games):

    In PVE events:
    Human: strike, attack, protect, and count down tiles: when randomly placed, these tiles had a 91% chance of being placed for immediate match. Although, the tiles may or may not be matched was not considered only if they could be matched with 1 move. The chance of being destroyed through a match 4 increased the probability to 94%. 

    AI: strike, attack, protect and countdown tiles: when randomly placed, these tiles had a 7% chance of being placed for immediate match. 

    In PVP events: 
    Human: matching 3 or more colors: I did not track this.
    AI: matching 3 or more colors: the AI matched or accumulated more than 3 AP in 84% of turns. I was actually shocked it was this high. 

    I wish I had tracked color availability and how it adjusted through the match. Seems that when I need a color the color isn't available but when the AI needs it, it drops immediately. 
    These numbers are too wild, haha. You're going to need to evidence these claims, because they are far too extreme to not be fabricated. I'll need to see the total number of Strike, Attack and Protect tiles placed, and the number of them that were placed such that they could be immediately matched. And so on, for all of the following "statistics" you provided. 
    If you don't or do believe these numbers is your problem, don't make it mine. If you doubt them, do it for yourself. Come up with your own numbers and post them then I will call you a liar. Fair enough?

    I've done my own unscientific observations and, in fact, the AI was just as likely to place a special tile that could be matched as I was. So, case closed. No cheating AI at play here. We can close this thread.
    An unscientific observation? What does that even mean? Did you make it up just so you could respond to my thread or was it to impress someone? Did you disagree with my empirical data because it goes against how you feel? I hope you get the person's approval you are seeking by posting on this thread. Maybe they will pat you the head and say good job.
  • Sm0keyJ0e
    Sm0keyJ0e Posts: 730 Critical Contributor
    edited September 2019
    jambattles said:
    Sm0keyJ0e said:
    jambattles said:
    IIAlonditeII said:
    jambattles said:
    After reading some responses about confirmation bias, I decided to keep up with some stats over my next 100 games. After keeping with the stats, there is zero question of AI advantage. It's a joke to consider it not a cheating pos algorithm. I get the disadvantages of the AI but to claim the AI isn't given an advantage is a lie.


    In the games I played and tracked (100 straight games):

    In PVE events:
    Human: strike, attack, protect, and count down tiles: when randomly placed, these tiles had a 91% chance of being placed for immediate match. Although, the tiles may or may not be matched was not considered only if they could be matched with 1 move. The chance of being destroyed through a match 4 increased the probability to 94%. 

    AI: strike, attack, protect and countdown tiles: when randomly placed, these tiles had a 7% chance of being placed for immediate match. 

    In PVP events: 
    Human: matching 3 or more colors: I did not track this.
    AI: matching 3 or more colors: the AI matched or accumulated more than 3 AP in 84% of turns. I was actually shocked it was this high. 

    I wish I had tracked color availability and how it adjusted through the match. Seems that when I need a color the color isn't available but when the AI needs it, it drops immediately. 
    These numbers are too wild, haha. You're going to need to evidence these claims, because they are far too extreme to not be fabricated. I'll need to see the total number of Strike, Attack and Protect tiles placed, and the number of them that were placed such that they could be immediately matched. And so on, for all of the following "statistics" you provided. 
    If you don't or do believe these numbers is your problem, don't make it mine. If you doubt them, do it for yourself. Come up with your own numbers and post them then I will call you a liar. Fair enough?

    I've done my own unscientific observations and, in fact, the AI was just as likely to place a special tile that could be matched as I was. So, case closed. No cheating AI at play here. We can close this thread.
    An unscientific observation? What does that even mean? Did you make it up just so you could respond to my thread or was it to impress someone? Did you disagree with my empirical data because it goes against how you feel? I hope you get the person's approval you are seeking by posting on this thread. Maybe they will pat you the head and say good job.

     I'm sorry you're having such a hard time matching 3. But the AI is not cheating and there is no "AI advantage". In fact there is a distinct AI disadvantage. The following are facts:
    1. You get to go first
    2. The AI won't take L or T match 5's
    3. The AI doesn't prioritize special tiles or beneficial neutral tiles likes webbed or charged tiles
    4. The AI can't predict cascades or set the board up in its favor
    5. The AI only chases their colors, not necessarily active colors that could be used to fire powers
    6. The AI can only fire one power per turn, even if there is enough AP to double-fire a power
    I could go on...

    So coming on here and posting baseless stats to help support your claim, and then deriding another poster for (validly) asking for more proof... I don't know what to tell you other than Candy Crush is still available for download.
  • jp1
    jp1 Posts: 1,081 Chairperson of the Boards
    ThaRoadWarrior said:
    I'd like to see somebody track the probability that matches involving an opposing Teen Jean present you with a turn-1 match 5 opportunity.
    This feels so real. Sometimes two of them.
  • ThaRoadWarrior
    ThaRoadWarrior Posts: 9,387 Chairperson of the Boards
    Sm0keyJ0e said:


     I'm sorry you're having such a hard time matching 3. But the AI is not cheating and there is no "AI advantage". In fact there is a distinct AI disadvantage. The following are facts:
    1. You get to go first
    2. The AI won't take L or T match 5's
    3. The AI doesn't prioritize special tiles or beneficial neutral tiles likes webbed or charged tiles
    4. The AI can't predict cascades or set the board up in its favor
    5. The AI only chases their colors, not necessarily active colors that could be used to fire powers
    6. The AI can only fire one power per turn, even if there is enough AP to double-fire a power
    I could go on...

    So coming on here and posting baseless stats to help support your claim, and then deriding another poster for (validly) asking for more proof... I don't know what to tell you other than Candy Crush is still available for download.
    A couple of these aren't strictly accurate.

    2. The Ai WILL take these by accident if the board is set up appropriately. In the scenario where there is an l-shaped match 5 set up, and the missing piece will create a match 4, there is a 50% chance it will slide it into the match 5 position. They don't avoid match 5s, they just don't know they exist. This happens to me at least a few times/week, and you can engineer the scenario with tile swappers if you want to test it yourself. 

    6. The ai will only fire the same power once per turn, i've had them fire different powers they have enough ap to fire on their turn. You see this often with 5* hawkeye who has AP coming out of his ears, and he will fire blue and red on the same turn. Gladiathor similarly will do it. 
  • jambattles
    jambattles Posts: 18 Just Dropped In
    Sm0keyJ0e said:
    jambattles said:
    Sm0keyJ0e said:
    jambattles said:
    IIAlonditeII said:
    jambattles said:
    After reading some responses about confirmation bias, I decided to keep up with some stats over my next 100 games. After keeping with the stats, there is zero question of AI advantage. It's a joke to consider it not a cheating pos algorithm. I get the disadvantages of the AI but to claim the AI isn't given an advantage is a lie.


    In the games I played and tracked (100 straight games):

    In PVE events:
    Human: strike, attack, protect, and count down tiles: when randomly placed, these tiles had a 91% chance of being placed for immediate match. Although, the tiles may or may not be matched was not considered only if they could be matched with 1 move. The chance of being destroyed through a match 4 increased the probability to 94%. 

    AI: strike, attack, protect and countdown tiles: when randomly placed, these tiles had a 7% chance of being placed for immediate match. 

    In PVP events: 
    Human: matching 3 or more colors: I did not track this.
    AI: matching 3 or more colors: the AI matched or accumulated more than 3 AP in 84% of turns. I was actually shocked it was this high. 

    I wish I had tracked color availability and how it adjusted through the match. Seems that when I need a color the color isn't available but when the AI needs it, it drops immediately. 
    These numbers are too wild, haha. You're going to need to evidence these claims, because they are far too extreme to not be fabricated. I'll need to see the total number of Strike, Attack and Protect tiles placed, and the number of them that were placed such that they could be immediately matched. And so on, for all of the following "statistics" you provided. 
    If you don't or do believe these numbers is your problem, don't make it mine. If you doubt them, do it for yourself. Come up with your own numbers and post them then I will call you a liar. Fair enough?

    I've done my own unscientific observations and, in fact, the AI was just as likely to place a special tile that could be matched as I was. So, case closed. No cheating AI at play here. We can close this thread.
    An unscientific observation? What does that even mean? Did you make it up just so you could respond to my thread or was it to impress someone? Did you disagree with my empirical data because it goes against how you feel? I hope you get the person's approval you are seeking by posting on this thread. Maybe they will pat you the head and say good job.

     I'm sorry you're having such a hard time matching 3. But the AI is not cheating and there is no "AI advantage". In fact there is a distinct AI disadvantage. The following are facts:
    1. You get to go first
    2. The AI won't take L or T match 5's
    3. The AI doesn't prioritize special tiles or beneficial neutral tiles likes webbed or charged tiles
    4. The AI can't predict cascades or set the board up in its favor
    5. The AI only chases their colors, not necessarily active colors that could be used to fire powers
    6. The AI can only fire one power per turn, even if there is enough AP to double-fire a power
    I could go on...

    So coming on here and posting baseless stats to help support your claim, and then deriding another poster for (validly) asking for more proof... I don't know what to tell you other than Candy Crush is still available for download.
    I acknowledged the AI disadvantage but to claim the AI doesn't have the advantages I showed is a lie. Over the next 100 games you play in PVE and PVP, keep track of the data. It's not hard. It is time consuming. It's difficult to track with lower tier games because the match is over so quickly (sometimes the AI only gets 1 or 2 moves). When you doubt the validity of data, showing more proof won't help you to accept the data as fact. The only way for you to overcome your disbelief is to do the work yourself. Take climate change for an example. There are people who deny the earth is getting warmer. Yet, everytime empirical data is released proving the earth is getting hotter, people demand different data. It just the way it is. Once you require more data than what is presented, it is better for you to conduct the experiment. So, when a person asks for more proof, they deserve all the deriding they get. 
  • BriMan2222
    BriMan2222 Posts: 1,231 Chairperson of the Boards
    edited September 2019
    ThaRoadWarrior said:
    Sm0keyJ0e said:


     I'm sorry you're having such a hard time matching 3. But the AI is not cheating and there is no "AI advantage". In fact there is a distinct AI disadvantage. The following are facts:
    1. You get to go first
    2. The AI won't take L or T match 5's
    3. The AI doesn't prioritize special tiles or beneficial neutral tiles likes webbed or charged tiles
    4. The AI can't predict cascades or set the board up in its favor
    5. The AI only chases their colors, not necessarily active colors that could be used to fire powers
    6. The AI can only fire one power per turn, even if there is enough AP to double-fire a power
    I could go on...

    So coming on here and posting baseless stats to help support your claim, and then deriding another poster for (validly) asking for more proof... I don't know what to tell you other than Candy Crush is still available for download.
    A couple of these aren't strictly accurate.

    2. The Ai WILL take these by accident if the board is set up appropriately. In the scenario where there is an l-shaped match 5 set up, and the missing piece will create a match 4, there is a 50% chance it will slide it into the match 5 position. They don't avoid match 5s, they just don't know they exist. This happens to me at least a few times/week, and you can engineer the scenario with tile swappers if you want to test it yourself. 

    6. The ai will only fire the same power once per turn, i've had them fire different powers they have enough ap to fire on their turn. You see this often with 5* hawkeye who has AP coming out of his ears, and he will fire blue and red on the same turn. Gladiathor similarly will do it. 
    I can't speak for Road Warrior, but on 6 I don't think he meant that the AI won't fire 2 powers in one turn, I think he mean that they won't fire the exact same power more than once.

    For instance if I have 30 blue ap with hawkeye I will fire that power over and over on the same turn until I use it all while the AI will fire that power once and save the other 25 ap until its next turn.
  • ThaRoadWarrior
    ThaRoadWarrior Posts: 9,387 Chairperson of the Boards
    edited September 2019
    It's a syntax distinction worth clarifying in a thread where somebody is salty about AI advantage.

    In the spirit of clarity, this is the scenario I was describing where the AI will "accidentally" make an l-shaped match 5:

    It can either move the blue one into place and diffuse the match 5, or it can slot either of the two end caps into place to complete the L. If it were actively avoiding match 5s, it would do the blue move 100% of the time, but since it's only looking for match 4 opportunities, there is a 50% chance it will do one of red the match 5 moves.

    Again, I think it's valuable to note that the AI isn't actively avoiding match 5 opportunities, it just doesn't know about them if they're non-linear. So they'll happen if this scenario exists.
  • Zeofar
    Zeofar Posts: 45 Just Dropped In
    edited September 2019
    jambattles said:
    Sm0keyJ0e said:
    I've done my own unscientific observations and, in fact, the AI was just as likely to place a special tile that could be matched as I was. So, case closed. No cheating AI at play here. We can close this thread.
    An unscientific observation? What does that even mean? Did you make it up just so you could respond to my thread or was it to impress someone? Did you disagree with my empirical data because it goes against how you feel? I hope you get the person's approval you are seeking by posting on this thread. Maybe they will pat you the head and say good job.
    You've been pushing out invented numbers since the opening post here. There is no "empirical data" if nobody else gets to look at it. Either release video of your tracked sessions or take screenshots of everything and put it in a google doc spreadsheet with your derived statistics if you want claims this extreme to be taken seriously.
  • ThaRoadWarrior
    ThaRoadWarrior Posts: 9,387 Chairperson of the Boards
    I'd actually be pretty impressed to see a match where Welcome to the Team managed to put out 7 strikes with none of them being matchable at all.
  • jambattles
    jambattles Posts: 18 Just Dropped In
    ThaRoadWarrior said:
    I'd actually be pretty impressed to see a match where Welcome to the Team managed to put out 7 strikes with none of them being matchable at all.
    A couple of precautions had to be taken. Like I couldn't use 3* Storm. I tried various teams but it was mainly iron fist 3*, Luke Cage 3*, Scarlett Witch 3*, and Draken. I tried to use 4* Rocket and Groot as well as Hawkeye Ronin but felt like they were skewing the data and they were hard to keep up with, lol. 
  • ThaRoadWarrior
    ThaRoadWarrior Posts: 9,387 Chairperson of the Boards
    edited September 2019
    Ok, so that's what people are reacting to with "unscientific." you need to be very explicit about what you're testing, and control all the variables save 1 every time. follow the scientific method, as it were, to be scientific. 

    It's an interesting distinction you make though. Some countdowns that are randomly placed are great when they end up in a corner. like you want them to be hidden from the board and safe until they burn down. Photonic Barrage is like this. Other countdowns you want out in the middle, like 2* magneto's countdowns or Scarlet Witch so they can affect the maximum number of surrounding tiles. You may or may not want Medusa's Entanglement tile to be matchable depending on how you're playing her. If you're using Okoye to force multiply, you probably want it to live a long time. But if you're up against a Kitty team, you probably need to match it a few times to populate the board with attack tiles. So you're going to "feel" situationally like things are or are not going the way you want them because the randomness isn't going to follow your plan situationally.

    If you find your tiles are getting matched away, and you're worried about them, you can leave available match-4 opportunities for the opponent which they will take regardless of color desirability. That can give you a chance to disrupt some of the bad positioning on your tiles. You don't always have that option, but if you look for it sometimes you can get away with it.
  • jambattles
    jambattles Posts: 18 Just Dropped In
    ThaRoadWarrior said:
    Ok, so that's what people are reacting to with "unscientific." you need to be very explicit about what you're testing, and control all the variables save 1 every time. follow the scientific method, as it were, to be scientific. 

    It's an interesting distinction you make though. Some countdowns that are randomly placed are great when they end up in a corner. like you want them to be hidden from the board and safe until they burn down. Photonic Barrage is like this. Other countdowns you want out in the middle, like 2* magneto's countdowns or Scarlet Witch so they can affect the maximum number of surrounding tiles. You may or may not want Medusa's Entanglement tile to be matchable depending on how you're playing her. If you're using Okoye to force multiply, you probably want it to live a long time. But if you're up against a Kitty team, you probably need to match it a few times to populate the board with attack tiles. So you're going to "feel" situationally like things are or are not going the way you want them because the randomness isn't going to follow your plan situationally.

    If you find your tiles are getting matched away, and you're worried about them, you can leave available match-4 opportunities for the opponent which they will take regardless of color desirability. That can give you a chance to disrupt some of the bad positioning on your tiles. You don't always have that option, but if you look for it sometimes you can get away with it.
    I'm not sure why they started with the "unscientific" discussion. I figured I would post the findings. Discuss the outcomes. Discuss how I controlled the variables. Overall, what was my method. Figured we could find ways to improve the experiment (such as using a certain team or character). Then we could move into the second iteration. Instead, I was called a liar and my data was unscientific. There are trolls in all forums.

    with all that said, I want you know. I appreciate your feedback.  
  • Zeofar
    Zeofar Posts: 45 Just Dropped In
    jambattles said:
    I'm not sure why they started with the "unscientific" discussion. I figured I would post the findings. Discuss the outcomes. Discuss how I controlled the variables. Overall, what was my method. Figured we could find ways to improve the experiment (such as using a certain team or character). Then we could move into the second iteration. Instead, I was called a liar and my data was unscientific. There are trolls in all forums.

    with all that said, I want you know. I appreciate your feedback.  
    jambattles said:
    I acknowledged the AI disadvantage but to claim the AI doesn't have the advantages I showed is a lie. Over the next 100 games you play in PVE and PVP, keep track of the data. It's not hard. It is time consuming. It's difficult to track with lower tier games because the match is over so quickly (sometimes the AI only gets 1 or 2 moves). When you doubt the validity of data, showing more proof won't help you to accept the data as fact. The only way for you to overcome your disbelief is to do the work yourself. Take climate change for an example. There are people who deny the earth is getting warmer. Yet, everytime empirical data is released proving the earth is getting hotter, people demand different data. It just the way it is. Once you require more data than what is presented, it is better for you to conduct the experiment. So, when a person asks for more proof, they deserve all the deriding they get. 
    "There are trolls in all forums" is right. This isn't like climate change denial, this is like a flat-earther relying on broken psuedoscience to "prove" something totally beyond belief and saying "How about I don't" when someone asks them to explain themselves in the most basic way. Comparing your two paragraphs of "X% of the time" and an assurance that you didn't make it up to climate change science which is backed up by hundreds of independent studies in multiple fields, going back decades (with published data sets and meticulously detailed methods and analysis) is so incredibly dishonest that I really can't see how calling you a liar would be unfair.
    It would be possible for someone to waste their time doing their own experiment to demonstrate that your results are bunk... but why would they? Personal experience already tells me that what you claim isn't remotely true. No-one is going to believe you anyway, especially since you won't even back up your results with actual data (which, if it exists, must certainly be readily available since you only posted those results today). Not to mention the that tinykitty asymmetry principle goes exponential in this case since 1) there's no reason at this point to believe you actually conducted any experiment and 2) even if you did, conducting an experiment that, unlike yours, could be confirmed to have any validity would be much more difficult than what you've put on offer in this thread.
    You're not being treated unfairly. Give us the data if you want to be taken seriously, and don't try to act like you should be protected by an aura of science if you don't know what data actually is.
  • jambattles
    jambattles Posts: 18 Just Dropped In
    BriMan2222 said:
    jambattles said:
    Zeofar said:
    jambattles said:
    I'm not sure why they started with the "unscientific" discussion. I figured I would post the findings. Discuss the outcomes. Discuss how I controlled the variables. Overall, what was my method. Figured we could find ways to improve the experiment (such as using a certain team or character). Then we could move into the second iteration. Instead, I was called a liar and my data was unscientific. There are trolls in all forums.

    with all that said, I want you know. I appreciate your feedback.  
    jambattles said:
    I acknowledged the AI disadvantage but to claim the AI doesn't have the advantages I showed is a lie. Over the next 100 games you play in PVE and PVP, keep track of the data. It's not hard. It is time consuming. It's difficult to track with lower tier games because the match is over so quickly (sometimes the AI only gets 1 or 2 moves). When you doubt the validity of data, showing more proof won't help you to accept the data as fact. The only way for you to overcome your disbelief is to do the work yourself. Take climate change for an example. There are people who deny the earth is getting warmer. Yet, everytime empirical data is released proving the earth is getting hotter, people demand different data. It just the way it is. Once you require more data than what is presented, it is better for you to conduct the experiment. So, when a person asks for more proof, they deserve all the deriding they get. 
    "There are trolls in all forums" is right. This isn't like climate change denial, this is like a flat-earther relying on broken psuedoscience to "prove" something totally beyond belief and saying "How about I don't" when someone asks them to explain themselves in the most basic way. Comparing your two paragraphs of "X% of the time" and an assurance that you didn't make it up to climate change science which is backed up by hundreds of independent studies in multiple fields, going back decades (with published data sets and meticulously detailed methods and analysis) is so incredibly dishonest that I really can't see how calling you a liar would be unfair.
    It would be possible for someone to waste their time doing their own experiment to demonstrate that your results are bunk... but why would they? Personal experience already tells me that what you claim isn't remotely true. No-one is going to believe you anyway, especially since you won't even back up your results with actual data (which, if it exists, must certainly be readily available since you only posted those results today). Not to mention the that tinykitty asymmetry principle goes exponential in this case since 1) there's no reason at this point to believe you actually conducted any experiment and 2) even if you did, conducting an experiment that, unlike yours, could be confirmed to have any validity would be much more difficult than what you've put on offer in this thread.
    You're not being treated unfairly. Give us the data if you want to be taken seriously, and don't try to act like you should be protected by an aura of science if you don't know what data actually is. 
    Wow! Did you mean to sound so stupid? Don't get me wrong. You write better than I do. You presentation is decent, above average. But your logic is stupid, just so stupid.

    Let me get this straight. Using climate change as an example of how an incredibility researched topic is still denied by people isn't a good example that proof doesn't change people's opinions? Did I compare my test with the test of all the independent studies? No, I didn't. I used a well researched topic to show that proof doesn't matter to people. What matters to them is their opinion. Like you said, your personal experience already tells you that my claim isn't remotely true, which means data means nothing to you. I have presented you with an opportunity to use the scientific knowledge you want people to think you possess. Yet, you just talk. 

    Like I said earlier, believe me or don't that's your problem. Don't make it my problem by asking for proof that you would probably dismiss anyway. Instead, do your own research. See if what I say is valid. The PVP is not hard. Did the AI's move result in more than 3AP. If it is true, 1 in that 3+ column. If it is not true, 1 in the 3 column. 

    Your entire 2nd paragraph is the tinykitty asymmetry principle. I mean its overflowing. Are you a neckbeard? You seem like a neckbeard. 
    Dude, you basically said "I have research that shows the AI has an insanely huge advantage but I'm not actually showing any evidence of this" to which Zeofar replied stating that you need some evidence for such a large claim.

    To which you basically replied "omg are you calling me a liar?! Go do your own research then.  I totally have tons of evidence for my outlandish claims but I don't feel like showing it and if you don't believe me then too bad"

    He has tried to explain to you what is required for real scientific research to which you have responded with calling him stupid and a troll and basically acting like a child having a temper tantrum.

    Some one is definitely coming across as a not very smart troll but it's not Zeofar.
    If do or don’t believe me, that’s your problem. Don’t make it mine. How hard is that to understand? If you don’t believe me and you’re not willing to do the very simple experiment, why are you on this post? Do you like getting with others that share your opinion and bullying people? What kind of magical data do you think exists for a video game forum? It’s literally basic math. How many “yes”. How many “no”. Add “yes” and “no” then divide “no” by the sum of “yes and no”. If you weren’t able to figure out the formula, there it is. Try the experiment. I can only assume you work for Marvel Puzzlequest and you hope to discredit my claim so you don’t lose your job. 
  • jp1
    jp1 Posts: 1,081 Chairperson of the Boards
    edited September 2019
    What the Talos are you going on about? All that effort not to post data you have readily available? Too far calling people stupid, especially considering your presentation.

    Just saying.

    Anyway, keep banging that drum. I bet any moment the MPQ dev team will jump in and finally admit their wrongdoings amidst such irrefutable evidence. 
  • BriMan2222
    BriMan2222 Posts: 1,231 Chairperson of the Boards
    jambattles said:
    BriMan2222 said:
    jambattles said:
    Zeofar said:
    jambattles said:
    I'm not sure why they started with the "unscientific" discussion. I figured I would post the findings. Discuss the outcomes. Discuss how I controlled the variables. Overall, what was my method. Figured we could find ways to improve the experiment (such as using a certain team or character). Then we could move into the second iteration. Instead, I was called a liar and my data was unscientific. There are trolls in all forums.

    with all that said, I want you know. I appreciate your feedback.  
    jambattles said:
    I acknowledged the AI disadvantage but to claim the AI doesn't have the advantages I showed is a lie. Over the next 100 games you play in PVE and PVP, keep track of the data. It's not hard. It is time consuming. It's difficult to track with lower tier games because the match is over so quickly (sometimes the AI only gets 1 or 2 moves). When you doubt the validity of data, showing more proof won't help you to accept the data as fact. The only way for you to overcome your disbelief is to do the work yourself. Take climate change for an example. There are people who deny the earth is getting warmer. Yet, everytime empirical data is released proving the earth is getting hotter, people demand different data. It just the way it is. Once you require more data than what is presented, it is better for you to conduct the experiment. So, when a person asks for more proof, they deserve all the deriding they get. 
    "There are trolls in all forums" is right. This isn't like climate change denial, this is like a flat-earther relying on broken psuedoscience to "prove" something totally beyond belief and saying "How about I don't" when someone asks them to explain themselves in the most basic way. Comparing your two paragraphs of "X% of the time" and an assurance that you didn't make it up to climate change science which is backed up by hundreds of independent studies in multiple fields, going back decades (with published data sets and meticulously detailed methods and analysis) is so incredibly dishonest that I really can't see how calling you a liar would be unfair.
    It would be possible for someone to waste their time doing their own experiment to demonstrate that your results are bunk... but why would they? Personal experience already tells me that what you claim isn't remotely true. No-one is going to believe you anyway, especially since you won't even back up your results with actual data (which, if it exists, must certainly be readily available since you only posted those results today). Not to mention the that tinykitty asymmetry principle goes exponential in this case since 1) there's no reason at this point to believe you actually conducted any experiment and 2) even if you did, conducting an experiment that, unlike yours, could be confirmed to have any validity would be much more difficult than what you've put on offer in this thread.
    You're not being treated unfairly. Give us the data if you want to be taken seriously, and don't try to act like you should be protected by an aura of science if you don't know what data actually is. 
    Wow! Did you mean to sound so stupid? Don't get me wrong. You write better than I do. You presentation is decent, above average. But your logic is stupid, just so stupid.

    Let me get this straight. Using climate change as an example of how an incredibility researched topic is still denied by people isn't a good example that proof doesn't change people's opinions? Did I compare my test with the test of all the independent studies? No, I didn't. I used a well researched topic to show that proof doesn't matter to people. What matters to them is their opinion. Like you said, your personal experience already tells you that my claim isn't remotely true, which means data means nothing to you. I have presented you with an opportunity to use the scientific knowledge you want people to think you possess. Yet, you just talk. 

    Like I said earlier, believe me or don't that's your problem. Don't make it my problem by asking for proof that you would probably dismiss anyway. Instead, do your own research. See if what I say is valid. The PVP is not hard. Did the AI's move result in more than 3AP. If it is true, 1 in that 3+ column. If it is not true, 1 in the 3 column. 

    Your entire 2nd paragraph is the tinykitty asymmetry principle. I mean its overflowing. Are you a neckbeard? You seem like a neckbeard. 
    Dude, you basically said "I have research that shows the AI has an insanely huge advantage but I'm not actually showing any evidence of this" to which Zeofar replied stating that you need some evidence for such a large claim.

    To which you basically replied "omg are you calling me a liar?! Go do your own research then.  I totally have tons of evidence for my outlandish claims but I don't feel like showing it and if you don't believe me then too bad"

    He has tried to explain to you what is required for real scientific research to which you have responded with calling him stupid and a troll and basically acting like a child having a temper tantrum.

    Some one is definitely coming across as a not very smart troll but it's not Zeofar.
    If do or don’t believe me, that’s your problem. Don’t make it mine. How hard is that to understand? If you don’t believe me and you’re not willing to do the very simple experiment, why are you on this post? Do you like getting with others that share your opinion and bullying people? What kind of magical data do you think exists for a video game forum? It’s literally basic math. How many “yes”. How many “no”. Add “yes” and “no” then divide “no” by the sum of “yes and no”. If you weren’t able to figure out the formula, there it is. Try the experiment. I can only assume you work for Marvel Puzzlequest and you hope to discredit my claim so you don’t lose your job. 
    Is english your first language? I can only assume it must not be because this is barely english. 

    "What you’ve just said is one of the most insanely idiotic things I have ever heard. At no point in your rambling, incoherent response were you even close to anything that could be considered a rational thought. Everyone in this room is now dumber for having listened to it. I award you no points, and may God have mercy on your soul.”
This discussion has been closed.

Categories