(Now 0 for 73) - Devs, is the system working as you intend?

Mercurywolf

Raffoon wrote:

carrion pigeons wrote:

Question: If 1600 people play your game, and each of them pulls 70 legendary tokens, what are the odds that one or more of them gets no 5*s?

Answer: just a bit better than 50%. The average case is that it will happen to one person. If 20000 people do it, ~12 of them will open no legendaries.

Question: If someone opens 70 legendary tokens, and doesn't pull a 5*, what are the odds of them complaining about it on the forums?

Answer: 100%, obviously.

So by statistics, I'm either grossly overestimating the number of people who have opened 70 legendary tokens, or else we're missing a whole bunch more threads about people who opened no 5*s after 70+ tokens. Either way, looks to me like we're doing better than average.

I think you've got it about right. The percentage complaining about it on the forum is probably a bit less than 100%, though. Also, there have been a fair number of threads complaining about similar odds. Maybe not 12 of them.

This opens up a new question though.

Question: Why do the devs insist on implementing the system this way, knowing that it will almost certainly completely screw a certain number of people?

I don't know the answer, but here are a few that might be possible.
Answer: They don't care about the player experience.
Answer: They do care, but feel it's much more important to keep making new characters, instead of fixing broken systems.
Answer: They're lazy and don't feel like implementing the changes needed to introduce a streakbreaker to the system.
Answer: They're bad at coding and can't implement a streakbreaker.
Answer: They'd like to implement a different system but don't have the time.
Answer: They'd like to implement a different system but can't think of one.
Answer: They don't actually realize it's a problem because they don't read the forums
Answer: They're rolling in money from buy clubs and are willing to sacrifice the experience of a certain number of people in order to keep doing that.

I don't know which of those would be the answer, but I sure wish I knew. Almost as annoying as going 70 pulls without a legendary is the complete lack of communication on the subject.

Another possible answer for why they chose this is because it is better for the long-term health of the game. Think about what would happen if they changed the system.

If they changed it from the current system, they'd have to determine how to make 5* available. If you make the system based solely on progression, then you have to make progression harder than it currently is. If you make it competition based, competition is going to get harder. Which means people who already have 5* are going to have an even bigger advantage than they already do.

If they switch it to giving people a boost after a number of non-5* pulls, then suddenly, 5*s become much more available, which means the Devs must focus entirely on the 5* tier as people will suddenly be able to max and champ these characters much faster. This would also force them to develop a 6* tier even sooner. Which would then put them right back into the situation of how to distribute these characters. Finally, it would exacerbate the already crippling ISO storage.

mpqr7

Mercurywolf wrote:

Finally, it would exacerbate the already crippling ISO storage.

That's not true, though. The ISO shortage is to level up the endless group of 4*s.

Leveling up a 5* isn't a problem because each level actually matters.

I spend 380K ISO to champion a 4* for an LT and some prizes... and to make my scaling more difficult (??).

I spend 45K to level up a 5* with a single additional cover, and suddenly I'm winning way more matches much more easily.

carrion_pigeons

OJSP wrote:

carrion pigeons wrote:

*I don't actually know how many people play MPQ, this number seems like a reasonable guess to me. If someone has a better number, I'll spend another 30 seconds to improve my calculation.

Not trying to make the discussion to get more heated, but it's also one of the subject I'm most interested in: there were 20161 alliance that finished last season (thanks Moon Roach). Taking into account partially filled alliances, I'm theorising, the number is closer to 350000 if we're being optimistic.

I think that's probably going to be my only worthwhile contribution on this matter.

I imagine some very large proportion of that group has never earned a legendary token, but let's assume the active population of the game really is that big. In that case, it would take a person 150 consecutive failed pulls at a legendary token to be statistically convincing to the population of players as a whole that something is "going on".

I would quite happily switch sides after 123, though. I really doubt that there are *that* many more than 20k players of the game who actively farm legendary tokens enough to realistically ever get to the 70 mark, let alone the 123 mark.

jobob

I would guess that the percentage of players who have made at least 100 LT draws is much better represented on these forums than the player base as a whole.

jobob

Also... It surprises me that people will talk about how implementing a draw system is "not a simple" thing to do... And we see code changes have unintended consequences in the game all the time... Why is it so hard to believe that there could be some weird influence on the output that hasn't been tracked down yet?

I never said the system was broken or rigged at all. I just don't know.

smkspy

Guess I'm one of the lucky ones, I'm always getting five stars. From maybe the last twenty tokens I've pulled one phoenix, 2 spidermans, a great goblin, and two oml.

thing is, other than jean, I don't want any five stars other than Jean and ss yet I have to use LT tokens to progress all my 4 stars.

carrion_pigeons

jobob wrote:

Also... It surprises me that people will talk about how implementing a draw system is "not a simple" thing to do... And we see code changes have unintended consequences in the game all the time... Why is it so hard to believe that there could be some weird influence on the output that hasn't been tracked down yet?

I never said the system was broken or rigged at all. I just don't know.

The thing is that selecting an object from a list in a way that has an identical distribution to real randomness is a solved problem, that any programmer can just go and copy off the API for whatever programming language they feel like using. No programmer needs to figure out how to do this if they don't want to.

I've been trying to come up with ways in which random number generators pull inputs, that could be exploited to generate non-random results. There are things I can imagine, but none of them seem likely. Especially not after several years of being the core distribution mechanic for a semi-popular game.

Raffoon

Mercurywolf wrote:

If they switch it to giving people a boost after a number of non-5* pulls, then suddenly, 5*s become much more available, which means the Devs must focus entirely on the 5* tier as people will suddenly be able to max and champ these characters much faster. This would also force them to develop a 6* tier even sooner. Which would then put them right back into the situation of how to distribute these characters. Finally, it would exacerbate the already crippling ISO storage.

If they introduced a streakbreaker and had to decrease the base odds slightly to account for that, it would still be preferable. The overall number of 5*s being pulled would be the same, but it would eliminate the extreme cases where people can play for months hitting every opportunity and still not find 5*s.

The request isn't to hand out more 5*s. The request is to hand out the same number of 5*s in a manner that is more reliable/consistent.

Mercurywolf

Raffoon wrote:

Mercurywolf wrote:

If they switch it to giving people a boost after a number of non-5* pulls, then suddenly, 5*s become much more available, which means the Devs must focus entirely on the 5* tier as people will suddenly be able to max and champ these characters much faster. This would also force them to develop a 6* tier even sooner. Which would then put them right back into the situation of how to distribute these characters. Finally, it would exacerbate the already crippling ISO storage.

If they introduced a streakbreaker and had to decrease the base odds slightly to account for that, it would still be preferable. The overall number of 5*s being pulled would be the same, but it would eliminate the extreme cases where people can play for months hitting every opportunity and still not find 5*s.

The request isn't to hand out more 5*s. The request is to hand out the same number of 5*s in a manner that is more reliable/consistent.

So what you're asking for is some people to get less so that you can get more? This would have to happen to preserve the "same number of 5*". I get that you're unhappy with how your pulls have worked out, but don't you think this would upset people who have actually gotten good pulls, which puts the devs right back in a position that people are unhappy with. And a streak-breaker would have to reduce the initial % by a fair amount for it to equalize the amount, otherwise we'll be in a situation where there's a huge glut of 5* available.

Having got 70 legendary tokens since their release, I have only gotten three 5* (I have 1 cover of SS, BSSM, and OML), my pull-rate is no-where near the 10%. That being said, I also understand this is how RNG has worked for me, and that someone else has likely gotten twice the 10% effective rate. It sucks, but it's fair overall.

I know I'm advocating for the devil here, but I don't see a real need for this particular system to be changed. Does the game need balance tweaks? For certain, but giving people more 5* isn't the way to go.

Raffoon

Mercurywolf wrote:

So what you're asking for is some people to get less so that you can get more? This would have to happen to preserve the "same number of 5*". I get that you're unhappy with how your pulls have worked out, but don't you think this would upset people who have actually gotten good pulls, which puts the devs right back in a position that people are unhappy with. And a streak-breaker would have to reduce the initial % by a fair amount for it to equalize the amount, otherwise we'll be in a situation where there's a huge glut of 5* available.

This doesn't affect pulls that have already happened, so it shouldn't affect anyone that already got good pulls any differently. Those people are just as likely to have their next 70 pulls be 0% 5*s as any other person.

Yes, I'm asking for a system where it's less likely someone will have a 25% pull rate, in order to avoid cases where people get screwed and have a 2% pull rate.

Those poor, poor lucky people. Now they'll pull at 20% now instead of 25%, so that the unlucky people can get a 6% pull rate instead of 2%. What a tragedy.

Mercurywolf wrote:

Having got 70 legendary tokens since their release, I have only gotten three 5* (I have 1 cover of SS, BSSM, and OML), my pull-rate is no-where near the 10%. That being said, I also understand this is how RNG has worked for me, and that someone else has likely gotten twice the 10% effective rate. It sucks, but it's fair overall.

If you have a 4% pull rate and someone else has a 20% pull rate just because of chance, no that's not fair. In fact, that's the opposite of fair. The fact that when you look at the entire playerbase there's an average 10% pull doesn't make the system fair.

Mercurywolf wrote:

I know I'm advocating for the devil here, but I don't see a real need for this particular system to be changed. Does the game need balance tweaks? For certain, but giving people more 5* isn't the way to go.

Again, this isn't asking for more 5*s to be put into the system. It's asking that the 5*s be distributed in a way that keeps the exact same number going out, but reduces the occurrence of very extreme situations.

Chief270

I just hit 0 for the last 70 mark yesterday as well.

0-71 now.
0-75 now 4/20/16
0-78 now 4/28/16
0-88 now 5/02/16
0-91 now 5/08/16

Polares

Raffoon wrote:

Mercurywolf wrote:

I know I'm advocating for the devil here, but I don't see a real need for this particular system to be changed. Does the game need balance tweaks? For certain, but giving people more 5* isn't the way to go.

Again, this isn't asking for more 5*s to be put into the system. It's asking that the 5*s be distributed in a way that keeps the exact same number going out, but reduces the occurrence of very extreme situations.

And not just that, which covers you pull and in which order is also very important in this particular case. I have one teammate in my alliance who has pulled 14 - 5s, but he has been lucky and he has pulled 7 from the same one, Phoenix, and in particular he has pulled 1/5/1, which is perfect because she has all the colors and has 5 from the best power. So he has been quite lucky in his pulls (it would be even better if it was OML, but Phoenix is quite good with 5 purple), he can start to use his Phoenix whenever he wanted.

I have been more or less lucky and I have a 10 cover OML, but also I have a 10 cover SS (then 6 cover Phoenix, 1 Spidey and 1 GG), so my numbers in general are quite good, I am above 10% (around 13%), but then if instead of 10SS-6Phoenix it was 6SS-10Phoenix, I would be cruisng through the game right now.

So, my point is, it is very important not only the number of 5s you pull, also WHICH 5s you pull. This is something that already was an issue with 4s and 3s, but much smaller because they are easy to get AND we can win them as progression and placement rewards. Not so much for 5s. Everything is just plain and simple LUCK, the most unfair thing in the world.

As you said RANDOMNESS makes the distribution of 5s BALANCED inside the game, but very much UNFAIR.

billsrule7

Chief270 wrote:

I just hit 0 for the last 70 mark yesterday as well.

So sorry to hear this Chief....I feel your pain.....

Cthulhu

Hey All,

We are definitely looking into this and while we don't have any answers just yet, we are reviewing and will have a plan very very soon!

Cthulhu

Polares

Cthulhu wrote:

Hey All,

We are definitely looking into this and while we don't have any answers just yet, we are reviewing and will have a plan very very soon!

Cthulhu

Hey just add a pity counter and it will mostly fix the problem (if you open 14 tokens without getting a 5, on the 15 pull you will always pull a 5, or 20 if you think 15 is too generous).

I had a streak of 51 tokens without pulling a 5 and it is really the worst you can feel in this game. I almost quit.

PS: This would improve things for the future a lot, but for some players it may be too late.

Chief270

Cthulhu wrote:

Hey All,

We are definitely looking into this and while we don't have any answers just yet, we are reviewing and will have a plan very very soon!

Cthulhu

Thank you. Was legit weighing my options after the civil war event.

Pongie

I hope an annoucement is made before the end of the next run of civil war. Need to decide if it's worthwhile to pull my hoard and hope for oml or wait for whatever chance will come?

billsrule7

Just an update for anyone who might care. My 0-for streak was finally snapped at 0-76.

Began a new streak of 0-13, pulled a 5*, and now have another streak of 0-19.

Overall since I began tracking in late January, I am 2 for 110. Something has to change, please Devs!!!!!

OneLastGambit

This discussion reminds me of that scene in casino when de niro complains to the chef that the blueberry muffins don't all have an equal number of blueberries in them

DarkWalkerWP

Currently 0 for 68. My last 5* draw was the first day goblin was available.