The problem is the 3rd streak where the prob is much higher.
Definitely something in their algorithm to piss ppl more because DB doesn’t want you to open the crate (event at 30+ streak ppl reporting blue trainers anyway)
in general ppl know something fishy going on
if they could just have option to bypass that stupid animation and see the result, I think most ppl will feel 90% better, just like the shuttle failure
I just push the button. Keep the sound off (so you don’t hear all the crits) and look away for three seconds. Surprisingly, it makes losing less painful.
The problem is the 3rd streak where the prob is much higher.
Definitely something in their algorithm to piss ppl more because DB doesn’t want you to open the crate (event at 30+ streak ppl reporting blue trainers anyway)
in general ppl know something fishy going on
if they could just have option to bypass that stupid animation and see the result, I think most ppl will feel 90% better, just like the shuttle failure
I just push the button. Keep the sound off (so you don’t hear all the crits) and look away for three seconds. Surprisingly, it makes losing less painful.
Sometimes I think something like this is happening in DB's headquarters.
Gang, every single one of us has lost a gauntlet match-up that we never ever ever should have lost, but we did anyway. As long as your opponent's 3 (or 6) max rolls + crits can beat your 3 (or 6) min rolls w/ no crits, then you can lose. The tail ends of the gaussian curve are not fake - low percentage outcomes can and do happen.
There also seems to be an issue with clustering in the RNG so that you tend to get series of good/bad rolls (which is part of the problem with the improbable since it feeds into you/the opponent rolling all high or all low or rolling 6 crits @25% or no crits @65%). While as they are randomly generated you should see this some of the time (less than 1% of the time for the 2 crit examples I gave) again it happens too much - the improbable is the norm in this game.
Whether or not it's rigged in the favour of the opponent, definitely, if for no other reason than they do not suffer from tiredness meaning that the defender is always at there maximum potential values.
Beyond that though I have no idea, I'd really need to spend a lot of time doing symmetric fights and see what way they come out. I do suspect the odds to crit are better for the opponent but that's just a sneaking suspicion.
*Basically plot the ((score rolled-min)/(max-min)) to see what they rolled. Where min = the minimum they could have rolled and max = the maximum they could have rolled.
The problem is gauntlet is built around streaks, but going for streaks means luck will eventually go against you. If you only play matches where you have a 90% chance to win, then you only have a 20% chance of getting a 15 streak. Eventually one of those 10% losses will show up and break your streak.
The other problem is you pick your matchups. If you played a match where you were only a 10% favored to win, you would win 1/10 and be excited about it. But since you never pick that matchup, when you overperform it means winning by a larger amount while when you underperform it means losing the match. You only remember the loss, because that's the only unexpected outcome.
Nope, I remember the odd (as in they shouldn't have happened) wins I get too, usually they happen when I'm effectively trying to refresh the opponents without paying merits. Heck had one earlier today.
I only tend to worry about streaks for the first 24hrs, after that I focus on points as while I do often manage to get streaks in the later 24hrs it does get tougher and I'm rather hoping to get another Caretaker (or 4 ).
I believe I lose 75% of my "mirror" matches aka my character against the exact same character. That drives me insane as I should be pretty close to 50-50, even better when their character is not maxed. I also seem to lose a lot of matches where the 5% crit person crits twice and my 25% crit person does not crit. The only reason why I do not swear at it is because I am around my family or co-workers when doing it.
I'm considering to join the game and for few days collecting intel on the forum. I've seen many, like really many, negative comments on rng here and gauntlet itself particularly. Is it really that bad? Is this mirror matches disparity "natural" thing or bug?
I'm considering to join the game and for few days collecting intel on the forum. I've seen many, like really many, negative comments on rng here and gauntlet itself particularly. Is it really that bad? Is this mirror matches disparity "natural" thing or bug?
Lots of people anecdotally claiming to lose a significant amount. Very little actual statistical e evidence.
I'm considering to join the game and for few days collecting intel on the forum. I've seen many, like really many, negative comments on rng here and gauntlet itself particularly. Is it really that bad? Is this mirror matches disparity "natural" thing or bug?
Lots of people anecdotally claiming to lose a significant amount. Very little actual statistical e evidence.
Jury is still out.
There is a saying that you can find a seed of truth in every fairy tale though. I hope it's not a redwood seed in this case.
It's true I've never collected data on mirror matches. I certainly can but after a year of shuttle data showing it's broken, it fell on deaf ears. Until a smart cookie figured out a chart that hit home to me.
I'm considering to join the game and for few days collecting intel on the forum. I've seen many, like really many, negative comments on rng here and gauntlet itself particularly. Is it really that bad? Is this mirror matches disparity "natural" thing or bug?
Many negative comments, true.
But there are truth behind them... if it is not that frustrating, you will see negative comments, not (quoting you) like really many negative comments.
Once you play, you will notice them unless DB just screws certain player
So my advice... play other games if you have choice.
I have cut down my playing time to 1/5 than before. And i am happier now and ready to fully retire.
But if you do want to play, don’t ever read the forums.
Or don’t spend too much money in the game (unless you have too much money, ofc) so you don’t have to be those ppl that cannot afford to quit the game because they have spent too much money
I think I posted this in another thread, but use your frustration and commit chaos. Yes, it goes against your natural survival instinct, but I am telling you, it works!
I have now changed my thinking in the gauntlet to "if the opponent has a higher stat, they should lose". And more often than not, this is correct, especially when facing mirror opponents.
Try it for yourself, and tell me I'm wrong.
EDIT: I should mention this generally only works in the left two opponent slots. The other 4 seem to be correct for win/lose stats.
As long as DB’s data on RNG works out in the aggregate, why would they worry about individual results? In that situation, RNG is less like a steady line, and more like the ups and downs of a roller coaster. Some people will have “good” luck while others get the “red-shirt treatment”. DB also does not do any sort of weighted rolls (in an attempt to increase validity and reliability of outcomes). It is what it is, and they have no incentive to change... Knowing that, play as many rounds as you can, don’t waste merits or dilithium, and hopefully you can obtain some descent rewards without investing in the system.
“A committee is a cul-de-sac, down which good ideas are lured and quietly strangled.” —Mark TwainMEMBER: [BoB] Barrel of Bloodwine... We are recruiting and putting the “curv” in scurvy! Best Event Finish: #3 Honor Debt: Inconceivable...Honor Bank Account: Slowly building...
I know a couple caretakers will come my way eventually, I'm not too worried about it. I don't even get mad at the bad beats anymore because I can predict them almost 100% of the time. I'll catch myself saying "alright he's going to go 3/3 at 25% and I'll go 0/6 at 25% and lose by a dozen points" and it'll happen. It's too bad I was paying for streak saves for the first couple months. Once I realized I could usually get a streak into the 20s after losing my first 30-40 streak I stopped bothering.
The very fact that we can predict when bad luck goes against would suggest that the system is rigged.
I have never got one of the legendary characters and the same is true of other senior players on my fleet but yet other players are talking about picking up the legend on more than 10 occasions. This suggests that the RNG is not as random as we are lead to believe
Streak at 38 just now. We both have a 5% crit chance, defenders range is 200-500 mine is 350-900, I already knew they would crit twice and high roll all 3 times. Not even surprised anymore.
So you failed on a multiple of 3? That never happens!
I know a couple caretakers will come my way eventually, I'm not too worried about it. I don't even get mad at the bad beats anymore because I can predict them almost 100% of the time. I'll catch myself saying "alright he's going to go 3/3 at 25% and I'll go 0/6 at 25% and lose by a dozen points" and it'll happen. It's too bad I was paying for streak saves for the first couple months. Once I realized I could usually get a streak into the 20s after losing my first 30-40 streak I stopped bothering.
The very fact that we can predict when bad luck goes against would suggest that the system is rigged.
I have never got one of the legendary characters and the same is true of other senior players on my fleet but yet other players are talking about picking up the legend on more than 10 occasions. This suggests that the RNG is not as random as we are lead to believe
Random is not the same as eveny distributed. They are like...polar opposite things, in fact.
I know a couple caretakers will come my way eventually, I'm not too worried about it. I don't even get mad at the bad beats anymore because I can predict them almost 100% of the time. I'll catch myself saying "alright he's going to go 3/3 at 25% and I'll go 0/6 at 25% and lose by a dozen points" and it'll happen. It's too bad I was paying for streak saves for the first couple months. Once I realized I could usually get a streak into the 20s after losing my first 30-40 streak I stopped bothering.
The very fact that we can predict when bad luck goes against would suggest that the system is rigged.
I have never got one of the legendary characters and the same is true of other senior players on my fleet but yet other players are talking about picking up the legend on more than 10 occasions. This suggests that the RNG is not as random as we are lead to believe
Random is not the same as eveny distributed. They are like...polar opposite things, in fact.
A poorly generated random number generator can result in clustering. The fact that people are reporting that they are getting multiple copies of a legendary in quick succession (including someone who managed to get two copies in a single set of end of gauntlet rewards) suggests that clustering could be taking place.
The chance of having any one particular set of numbers coming up on a lottery is exactly the same as any other set of numbers. However if you had the exact same set of numbers three times a week you would begin to suspect the random element.
I know a couple caretakers will come my way eventually, I'm not too worried about it. I don't even get mad at the bad beats anymore because I can predict them almost 100% of the time. I'll catch myself saying "alright he's going to go 3/3 at 25% and I'll go 0/6 at 25% and lose by a dozen points" and it'll happen. It's too bad I was paying for streak saves for the first couple months. Once I realized I could usually get a streak into the 20s after losing my first 30-40 streak I stopped bothering.
The very fact that we can predict when bad luck goes against would suggest that the system is rigged.
I have never got one of the legendary characters and the same is true of other senior players on my fleet but yet other players are talking about picking up the legend on more than 10 occasions. This suggests that the RNG is not as random as we are lead to believe
Random is not the same as eveny distributed. They are like...polar opposite things, in fact.
A poorly generated random number generator can result in clustering. The fact that people are reporting that they are getting multiple copies of a legendary in quick succession (including someone who managed to get two copies in a single set of end of gauntlet rewards) suggests that clustering could be taking place.
The chance of having any one particular set of numbers coming up on a lottery is exactly the same as any other set of numbers. However if you had the exact same set of numbers three times a week you would begin to suspect the random element.
This is completely false. True RNG always end up experiencing some amount of clustering.
I know a couple caretakers will come my way eventually, I'm not too worried about it. I don't even get mad at the bad beats anymore because I can predict them almost 100% of the time. I'll catch myself saying "alright he's going to go 3/3 at 25% and I'll go 0/6 at 25% and lose by a dozen points" and it'll happen. It's too bad I was paying for streak saves for the first couple months. Once I realized I could usually get a streak into the 20s after losing my first 30-40 streak I stopped bothering.
The very fact that we can predict when bad luck goes against would suggest that the system is rigged.
I have never got one of the legendary characters and the same is true of other senior players on my fleet but yet other players are talking about picking up the legend on more than 10 occasions. This suggests that the RNG is not as random as we are lead to believe
Random is not the same as eveny distributed. They are like...polar opposite things, in fact.
A poorly generated random number generator can result in clustering. The fact that people are reporting that they are getting multiple copies of a legendary in quick succession (including someone who managed to get two copies in a single set of end of gauntlet rewards) suggests that clustering could be taking place.
The chance of having any one particular set of numbers coming up on a lottery is exactly the same as any other set of numbers. However if you had the exact same set of numbers three times a week you would begin to suspect the random element.
This is completely false. True RNG always end up experiencing some amount of clustering.
I know a couple caretakers will come my way eventually, I'm not too worried about it. I don't even get mad at the bad beats anymore because I can predict them almost 100% of the time. I'll catch myself saying "alright he's going to go 3/3 at 25% and I'll go 0/6 at 25% and lose by a dozen points" and it'll happen. It's too bad I was paying for streak saves for the first couple months. Once I realized I could usually get a streak into the 20s after losing my first 30-40 streak I stopped bothering.
The very fact that we can predict when bad luck goes against would suggest that the system is rigged.
I have never got one of the legendary characters and the same is true of other senior players on my fleet but yet other players are talking about picking up the legend on more than 10 occasions. This suggests that the RNG is not as random as we are lead to believe
Random is not the same as eveny distributed. They are like...polar opposite things, in fact.
A poorly generated random number generator can result in clustering. The fact that people are reporting that they are getting multiple copies of a legendary in quick succession (including someone who managed to get two copies in a single set of end of gauntlet rewards) suggests that clustering could be taking place.
The chance of having any one particular set of numbers coming up on a lottery is exactly the same as any other set of numbers. However if you had the exact same set of numbers three times a week you would begin to suspect the random element.
This is completely false. True RNG always end up experiencing some amount of clustering.
I know a couple caretakers will come my way eventually, I'm not too worried about it. I don't even get mad at the bad beats anymore because I can predict them almost 100% of the time. I'll catch myself saying "alright he's going to go 3/3 at 25% and I'll go 0/6 at 25% and lose by a dozen points" and it'll happen. It's too bad I was paying for streak saves for the first couple months. Once I realized I could usually get a streak into the 20s after losing my first 30-40 streak I stopped bothering.
The very fact that we can predict when bad luck goes against would suggest that the system is rigged.
I have never got one of the legendary characters and the same is true of other senior players on my fleet but yet other players are talking about picking up the legend on more than 10 occasions. This suggests that the RNG is not as random as we are lead to believe
Random is not the same as eveny distributed. They are like...polar opposite things, in fact.
A poorly generated random number generator can result in clustering. The fact that people are reporting that they are getting multiple copies of a legendary in quick succession (including someone who managed to get two copies in a single set of end of gauntlet rewards) suggests that clustering could be taking place.
The chance of having any one particular set of numbers coming up on a lottery is exactly the same as any other set of numbers. However if you had the exact same set of numbers three times a week you would begin to suspect the random element.
This is completely false. True RNG always end up experiencing some amount of clustering.
Perhaps, but that's not the argument that was made.
That’s because some people doesn’t know they are different or they were misinformed. Even though RNG has issue, the culprit is the algorithm used in thus game
When I was studying random number generation and probability for my university computer science course we were told that clustering was the result of either a bad seed or a bad algorithm.
Some clustering can happen naturally but should even itself up over time (if you roll a true die you might get roll a 6 multiple times in a row if you roll a million times but the final split should be fairly equal)
A badly designed algorithm can create a RNG which clusters heavily a well written one would give a result which is statistically pretty even.
As long as DB’s data on RNG works out in the aggregate, why would they worry about individual results? In that situation, RNG is less like a steady line, and more like the ups and downs of a roller coaster. Some people will have “good” luck while others get the “red-shirt treatment”. DB also does not do any sort of weighted rolls (in an attempt to increase validity and reliability of outcomes). It is what it is, and they have no incentive to change... Knowing that, play as many rounds as you can, don’t waste merits or dilithium, and hopefully you can obtain some descent rewards without investing in the system.
If that's all it was, that would be fine. But it took a year of players insisting shuttles were broken and DB not caring/claiming their logs showed they were fine before someone produced enough evidence on a specific bug for Shan to file a bug report(that still has had no followup).
I'll also remind you that DB claimed to have updated the premium portal, players claimed the update didn't work, DB insisted that it did. Players spent a significant amount of real money to produce statistically significant samples proving an issue. Shan still reported that the code had been tested live in production and was working. 30 minutes later the Product Manager finally stepped in and admitted the portals were broken all along.
It's not that DB isn't perfect, it's that DB refuses to admit their mistakes and hides behind RNG to make it difficult to prove otherwise.
Comments
I just push the button. Keep the sound off (so you don’t hear all the crits) and look away for three seconds. Surprisingly, it makes losing less painful.
Sometimes I think something like this is happening in DB's headquarters.
I thought that was blackjack since the house wins on a draw allowing them to play more cautiously and still win?
Except that if you were to plot the outcomes for Gauntlet* I'm pretty certain it would not be a gaussian, as what should be improbable is too common. If I had to guess I'd expect the outcome to be more like a symmetric version of the graph here:
https://www.researchgate.net/figure/An-example-of-fitting-the-secondary-peak-using-a-double-Gaussian-profile-The-fringes-are_fig5_224455930
There also seems to be an issue with clustering in the RNG so that you tend to get series of good/bad rolls (which is part of the problem with the improbable since it feeds into you/the opponent rolling all high or all low or rolling 6 crits @25% or no crits @65%). While as they are randomly generated you should see this some of the time (less than 1% of the time for the 2 crit examples I gave) again it happens too much - the improbable is the norm in this game.
Whether or not it's rigged in the favour of the opponent, definitely, if for no other reason than they do not suffer from tiredness meaning that the defender is always at there maximum potential values.
Beyond that though I have no idea, I'd really need to spend a lot of time doing symmetric fights and see what way they come out. I do suspect the odds to crit are better for the opponent but that's just a sneaking suspicion.
*Basically plot the ((score rolled-min)/(max-min)) to see what they rolled. Where min = the minimum they could have rolled and max = the maximum they could have rolled.
The other problem is you pick your matchups. If you played a match where you were only a 10% favored to win, you would win 1/10 and be excited about it. But since you never pick that matchup, when you overperform it means winning by a larger amount while when you underperform it means losing the match. You only remember the loss, because that's the only unexpected outcome.
I only tend to worry about streaks for the first 24hrs, after that I focus on points as while I do often manage to get streaks in the later 24hrs it does get tougher and I'm rather hoping to get another Caretaker (or 4
John Scalzi's book is excellent, by the way.
EDIT: and looking the other direction does help, if you aren't trying for a streak.
This is why i hate gauntlet completely.
Hopefully this hasn't been a long term bug. But I've batted 25% in mirror matches also.
Lots of people anecdotally claiming to lose a significant amount. Very little actual statistical e evidence.
Jury is still out.
There is a saying that you can find a seed of truth in every fairy tale though. I hope it's not a redwood seed in this case.
Many negative comments, true.
But there are truth behind them... if it is not that frustrating, you will see negative comments, not (quoting you) like really many negative comments.
Once you play, you will notice them unless DB just screws certain player
So my advice... play other games if you have choice.
I have cut down my playing time to 1/5 than before. And i am happier now and ready to fully retire.
But if you do want to play, don’t ever read the forums.
Or don’t spend too much money in the game (unless you have too much money, ofc) so you don’t have to be those ppl that cannot afford to quit the game because they have spent too much money
I have now changed my thinking in the gauntlet to "if the opponent has a higher stat, they should lose". And more often than not, this is correct, especially when facing mirror opponents.
Try it for yourself, and tell me I'm wrong.
EDIT: I should mention this generally only works in the left two opponent slots. The other 4 seem to be correct for win/lose stats.
The very fact that we can predict when bad luck goes against would suggest that the system is rigged.
I have never got one of the legendary characters and the same is true of other senior players on my fleet but yet other players are talking about picking up the legend on more than 10 occasions. This suggests that the RNG is not as random as we are lead to believe
So you failed on a multiple of 3? That never happens!
Random is not the same as eveny distributed. They are like...polar opposite things, in fact.
Most common
Surak
Mirror Picard
Mirror Geordi
Admiral Riker
Mirror Troi
rotation if they have crits
Nurse Faith Garland
Mirror Phlox
Elen Landry
Weeding T'Pol
I will never get Caretaker, Guinan or Locutus...
A poorly generated random number generator can result in clustering. The fact that people are reporting that they are getting multiple copies of a legendary in quick succession (including someone who managed to get two copies in a single set of end of gauntlet rewards) suggests that clustering could be taking place.
The chance of having any one particular set of numbers coming up on a lottery is exactly the same as any other set of numbers. However if you had the exact same set of numbers three times a week you would begin to suspect the random element.
This is completely false. True RNG always end up experiencing some amount of clustering.
https://en.m.wikipedia.org/wiki/Clustering_illusion
There is RNG and there is algorithm.
The issue is with the algorithm, not the RNG.
Perhaps, but that's not the argument that was made.
That’s because some people doesn’t know they are different or they were misinformed. Even though RNG has issue, the culprit is the algorithm used in thus game
When I was studying random number generation and probability for my university computer science course we were told that clustering was the result of either a bad seed or a bad algorithm.
Some clustering can happen naturally but should even itself up over time (if you roll a true die you might get roll a 6 multiple times in a row if you roll a million times but the final split should be fairly equal)
A badly designed algorithm can create a RNG which clusters heavily a well written one would give a result which is statistically pretty even.
If that's all it was, that would be fine. But it took a year of players insisting shuttles were broken and DB not caring/claiming their logs showed they were fine before someone produced enough evidence on a specific bug for Shan to file a bug report(that still has had no followup).
I'll also remind you that DB claimed to have updated the premium portal, players claimed the update didn't work, DB insisted that it did. Players spent a significant amount of real money to produce statistically significant samples proving an issue. Shan still reported that the code had been tested live in production and was working. 30 minutes later the Product Manager finally stepped in and admitted the portals were broken all along.
It's not that DB isn't perfect, it's that DB refuses to admit their mistakes and hides behind RNG to make it difficult to prove otherwise.
Except Kortar. Kortar is my nemesis and whoops my butt any time I face him