I would actually estimate that a thorough probabilistic analysis would show that the chance of you winning in such a scenario is above 50% if you were to appropriately determine the factors leading to a 6m GP player dropping to that level.
I win = i met an opponent with higher GP AND he doesn't attack me.
How do you count that probability?
In your math, you have 25% of meeting a much stronger opponent and winning. You have 75% of the exact opposite, which is not losing but not meeting a much stronger opponent and winning. You can very well meet someone with similar gp in those 75%.
Except that there isn't necessarily a 50% chance of either happening.
Example: I roll a die and there are two outcomes. Either it lands on six or it doesn't land on six , but as you can appreciate the chance of it landing on a 6 is not 50%.
So, when i roll a die, it will either land on six or it wont and i agree, the chance of landing on 6 is not 50%.
Because there are other possibilities, they are 1, 2, 3, 4 and 5.
So, the chance of landing on six = 1/6
And the chance of not landing on six = 5/6
And the chance of winning the lottery is 1/(how many lotteries have been sold) (i guess).
It is not 50% (either i win or not) because there are other possibilities, they are lottery No 000,000,000,001 until idk, lottery number 100.000.000.000. Between these numbers, some 99,888,777,666 lottery are sold...so the chance of winning my 1 lottery is 1/99,888,777,666.
So, what are other possibilities that would be presented when i click that button other than "higher GP" and "same/lower GP"?
And what other possibilities would be presented other than "the higher GP attacks" and "he doesn't"?
Either way...the chance of lower GP will win against higher GP is above 50% is just counter logic.
But your argument is flawed by the assumption that there is a 50/50 chance that they play based on the premise that they can play or not. The factors affecting win/loss in GAC are very complex and so winning/losing or simple participation are not 50% options, that's what you seem to be ignoring or not understanding.
If the chance that they play, against a player with your GP given that they've dropped that far (and that's the important bit), is only 10% then you have a 90% chance of winning against such a player. Not all outcomes are equally likely, and that's what you seem to be missing.
I'm not sure I can find another way of explaining this.
Account started June 2020. 100% FTP. 8.2m GP. JMK, JML, SLKR, and SEE. Exe and Levi. Ally code 117-269-921. Swgoh.gg
Well I think the most important question right now is as follows
What are we calculating the probability of? Are you winning a GAC match or a GAC match against an opponent with a statistically significantly higher GP than you?
The use of statistically significant is crucial here as a person with 1 more GP than you probably isn't going to make that much of a difference.
Except that there isn't necessarily a 50% chance of either happening.
Example: I roll a die and there are two outcomes. Either it lands on six or it doesn't land on six , but as you can appreciate the chance of it landing on a 6 is not 50%.
So, when i roll a die, it will either land on six or it wont and i agree, the chance of landing on 6 is not 50%.
Because there are other possibilities, they are 1, 2, 3, 4 and 5.
So, the chance of landing on six = 1/6
And the chance of not landing on six = 5/6
And the chance of winning the lottery is 1/(how many lotteries have been sold) (i guess).
It is not 50% (either i win or not) because there are other possibilities, they are lottery No 000,000,000,001 until idk, lottery number 100.000.000.000. Between these numbers, some 99,888,777,666 lottery are sold...so the chance of winning my 1 lottery is 1/99,888,777,666.
So, what are other possibilities that would be presented when i click that button other than "higher GP" and "same/lower GP"?
And what other possibilities would be presented other than "the higher GP attacks" and "he doesn't"?
Either way...the chance of lower GP will win against higher GP is above 50% is just counter logic.
When I face someone with lower GP than me, I assume they've got good mods, datacrons, and strategy. It's often a tougher matchup than someone with higher GP that's falling because of lack of interest or poor strategy.
That's why GP isn't the match making metric anymore. It's 'skill rating' which is based on participation and success/failure in this game mode.
Now that's not to say there are some outlier matchups that are extremely out of balance, but that's not the topic of this thread anyway. It's about how to manage ties. It's been states several times... We have a method of determining winners in the case of a tie. It's the person who has invested more in their roster. Seems logical to give them a tiny advantage for their efforts over time in the very rare case of a tie.
So much chatter about something that so rarely happens when there are legitimate issues the community could be discussing, it's just silly.
You and I can't answer that, because we don't have the datas.
To prove that some1 intentionally lowered his skill rating is another matter.
The point of my posts is to bring up that it is indeed unfair to do so to other player, according to my amateur mathematic.
But the astrophysicist has said that lower GP had above 50% chance of defeating higher GP.
So, i could be wrong and I already said I am sorry.
No data can directly show someone's intention. In theory, a heuristic model could be trained from a large amount of "ground truth" data, but the data set would need to include two categories: 1. a large number of players' in-game behaviors. This CG has. 2. those players' intentions while they demonstrated the in-game behaviors. This CG cannot possibly have. So you can't really sweep this under the "well we can't but CG can" rug.
And the lower GP does have 50% chance of defeating the higher GP. That's what the SR is there for. It is essentially a sum value of a player's history. What it really represents is that either of two players with the same SR has 50% chance of winning, judging from their history data. If one of them has such a high GP that they would certainly win if they play, then the history data must be that they only played 50% of the time.
By the way, that, i.e. 50% math expectation of winning, is actually how "fair chance" is defined in game theory.
If the chance that they play, against a player with your GP given that they've dropped that far (and that's the important bit), is only 10% then you have a 90% chance of winning against such a player. Not all outcomes are equally likely, and that's what you seem to be missing.
I'm not sure I can find another way of explaining this.
Ok, I understand.
I had 90% of winning against that 6M GP, given that he might have dropped that far from Aurodium III to Carb III.
Why didn't i feel that?
As soon as i saw his deployed squads, i knew I could do nothing.
Just for fun, i had 1 squad with 2 relic-ed characters, so why not...and I won only that battle.
BTW, i am out.
Next time, i will feel grateful when i see higher GP, knowing that my winning chance is 90%.
But they aren’t gaining an unfair advantage. Arguably they are entitled to have a 4m GP buffer over you by virtue of their investment in the game (be it time or money) comparative to yours. The fact that they have lost out on resources by not playing GAC to the potential of their roster is a choice they are choosing to make.
Also, he’s not necessarily ruining your rewards. How do you know you would have beaten another player with, say 1.5m GP with the same skill rating that could have been in their place?
But they aren’t gaining an unfair advantage. Arguably they are entitled to have a 4m GP buffer over you by virtue of their investment in the game (be it time or money) comparative to yours. The fact that they have lost out on resources by not playing GAC to the potential of their roster is a choice they are choosing to make.
Also, he’s not necessarily ruining your rewards. How do you know you would have beaten another player with, say 1.5m GP with the same skill rating that could have been in their place?
If the chance that they play, against a player with your GP given that they've dropped that far (and that's the important bit), is only 10% then you have a 90% chance of winning against such a player. Not all outcomes are equally likely, and that's what you seem to be missing.
I'm not sure I can find another way of explaining this.
Ok, I understand.
I had 90% of winning against that 6M GP, given that he might have dropped that far from Aurodium III to Carb III.
Why didn't i feel that?
As soon as i saw his deployed squads, i knew I could do nothing.
Just for fun, i had 1 squad with 2 relic-ed characters, so why not...and I won only that battle.
BTW, i am out.
Next time, i will feel grateful when i see higher GP, knowing that my winning chance is 90%.
If you’re punching up then it’s higher than 50% that you will see an opponent with more GP. Conversely, if you are losing a bunch, it’s higher than 50% that you will see opponents with less GP. Instead of trying to calculate the probability of your chance to win by looking at GP, just remember that no matter what you think the percentage is it’s zero if you don’t even try.
You can’t control who you are matched up against. You can’t control whether or not they “show up” in most cases. So try your best and move to the next round. As long as you keep growing you should eventually move past the people who aren’t trying.
Give a man a fish and you feed him for a day. Teach a man to fish and you feed him for life.
I can appreciate the frustration but many sports and games have "unfair" mechanics/rules like this that reward people who are already ahead. Where winners go first or get to keep possession of the ball ("make it take it"), serve first, get first move next game, etc. Are these systems all fair? No, probably not. But they're the rules we all play by and there's not much we can do about them. If it helps you maintain your sanity think of it not as a tie, but a 1 point loss (since you needed 1 extra point to avoid that situation and failed to achieve that).
Also, as someone who plays 6 accounts I can tell you ties are indeed very rare. I think I've had one, maybe two in the last few YEARS.
By unfair advantage, i am talking about their manipulation of skill rating. They intentionally lower their skill rating so that they can achieve whatever goal they have.
Than ofcourse, an player who has taken an hiatus from GAC has to be identified in ban investigation.
They have the datas whether a player intentionally lower their skill rating or not.
Tha fact is, by doing that, they put other in disadvantage (75% losing probability)
How do you prove someone intentionally lowered their skill rating? I mean, it's easy to say "... has to be identified in ban investigation", but what do you think CG should/can/will do when somebody says "well my brother borrowed my phone so I couldn't play last GAC", ask them to send a picture of their brother holding up today's paper and their phone?
+1 for the "my brother" reference from another equally comical thread that we see about as often as these.
Replies
In your math, you have 25% of meeting a much stronger opponent and winning. You have 75% of the exact opposite, which is not losing but not meeting a much stronger opponent and winning. You can very well meet someone with similar gp in those 75%.
But your argument is flawed by the assumption that there is a 50/50 chance that they play based on the premise that they can play or not. The factors affecting win/loss in GAC are very complex and so winning/losing or simple participation are not 50% options, that's what you seem to be ignoring or not understanding.
If the chance that they play, against a player with your GP given that they've dropped that far (and that's the important bit), is only 10% then you have a 90% chance of winning against such a player. Not all outcomes are equally likely, and that's what you seem to be missing.
I'm not sure I can find another way of explaining this.
What are we calculating the probability of? Are you winning a GAC match or a GAC match against an opponent with a statistically significantly higher GP than you?
The use of statistically significant is crucial here as a person with 1 more GP than you probably isn't going to make that much of a difference.
When I face someone with lower GP than me, I assume they've got good mods, datacrons, and strategy. It's often a tougher matchup than someone with higher GP that's falling because of lack of interest or poor strategy.
That's why GP isn't the match making metric anymore. It's 'skill rating' which is based on participation and success/failure in this game mode.
Now that's not to say there are some outlier matchups that are extremely out of balance, but that's not the topic of this thread anyway. It's about how to manage ties. It's been states several times... We have a method of determining winners in the case of a tie. It's the person who has invested more in their roster. Seems logical to give them a tiny advantage for their efforts over time in the very rare case of a tie.
So much chatter about something that so rarely happens when there are legitimate issues the community could be discussing, it's just silly.
No data can directly show someone's intention. In theory, a heuristic model could be trained from a large amount of "ground truth" data, but the data set would need to include two categories: 1. a large number of players' in-game behaviors. This CG has. 2. those players' intentions while they demonstrated the in-game behaviors. This CG cannot possibly have. So you can't really sweep this under the "well we can't but CG can" rug.
And the lower GP does have 50% chance of defeating the higher GP. That's what the SR is there for. It is essentially a sum value of a player's history. What it really represents is that either of two players with the same SR has 50% chance of winning, judging from their history data. If one of them has such a high GP that they would certainly win if they play, then the history data must be that they only played 50% of the time.
By the way, that, i.e. 50% math expectation of winning, is actually how "fair chance" is defined in game theory.
Ok, I understand.
I had 90% of winning against that 6M GP, given that he might have dropped that far from Aurodium III to Carb III.
Why didn't i feel that?
As soon as i saw his deployed squads, i knew I could do nothing.
Just for fun, i had 1 squad with 2 relic-ed characters, so why not...and I won only that battle.
BTW, i am out.
Next time, i will feel grateful when i see higher GP, knowing that my winning chance is 90%.
This part of your post was delicious.
I'm not saying it is that high, just that it could be that case. Assuming it is 50/50 isn't necessarily true either.
U like crusty issues?
If you’re punching up then it’s higher than 50% that you will see an opponent with more GP. Conversely, if you are losing a bunch, it’s higher than 50% that you will see opponents with less GP. Instead of trying to calculate the probability of your chance to win by looking at GP, just remember that no matter what you think the percentage is it’s zero if you don’t even try.
You can’t control who you are matched up against. You can’t control whether or not they “show up” in most cases. So try your best and move to the next round. As long as you keep growing you should eventually move past the people who aren’t trying.
Also, as someone who plays 6 accounts I can tell you ties are indeed very rare. I think I've had one, maybe two in the last few YEARS.
+1 for the "my brother" reference from another equally comical thread that we see about as often as these.