5* Pull Rates & Capping Futility

24

Comments

  • justsing
    justsing Posts: 512 Critical Contributor
    jamesh said:
    justsing said:
    No, a reduced pull rate in exchange for an assured minimum rate is not worth it in the long run imo.
    Why do you think it would be worse in the long run?  If it evened out to the same effective pull rate, then it should make no difference in the long run.
    Yeah, my bad. I misinterpreted the math stuff. Apologies to helix!
  • Vins2
    Vins2 Posts: 183 Tile Toppler
    Kolence said:
    All I know is, it's the runs of 30+ pulls without a 5* when I start thinking - what am I doing with this game? why bother with this tinykitty when it can all be in vain because of rng? I realize how the odds work, and can see on my own example that 20-30 runs without a 5* happen all the time. Still, that's how it feels to me. And those are not the trully horrible runs, people have had 50+ or 70+ run dry spells...

    In a similar sense, I can win the ~25 matches in a row to climb to 1200 in a pvp event and I don't bat an eye for the good luck that went my way.  The closest I've gotten to uninstalling this game were the rare times where I'll lose two in a row or within a short time of each other, due to bad cascades. 

    It would just benefit the game's further longevity to reduce the amount of those "feels bad man" type moments to as close to zero as they can.  And that's not just a philosophy for mpq.  It's a thing in modern (board/video/mobile) game design .
  • HoundofShadow
    HoundofShadow Posts: 8,004 Chairperson of the Boards
    If a 5* can be guaranteed within 7 pulls, what's the point of having 5* in HfH?

    By guaranteeing a 5* within 7 pulls, it will affect:

    1) cost of buying opened powers of 5* (720cp)
    2) cost of Latest 5* (500CP) and Classic 5* (250CP)
    3) reduce the value of 5* because they are easier to get than 4*
    4) affect the gameplay for 1*-3* players

    If this is implemented, the number of CP/LT/HT/HP that can be earned in game will be reduced dramaticallt to balance the game as a whole.

    Another reasoning is if 140 CP can guarantee me a 5* cover and 6 4* covers, the cost of no. 1 should be reduced to not more than 420CP and no 2. should be reduced to 120 CP and 80 CP respectively.

    Also, pity timer or streak breaker for the rarest tier implemented in other games don't happen within the display odds. For example, if streak breaker is implemented in MPQ, it's likely to be a guaranteed 5* by the 40th or 50th pull, rather than by the 7th pull before the timer resets.

    Having a streakbreaker don't necessary solve perception problem because like what was mentioned,  players are likely to be emotionally charged when things don't go their way. The devs might probably have new charges laid upon them after implementing streakbreaker (assuming a guaranteed 5* on the 40th pull) such as:

    1) the game was rigged to give them unusable covers or useless frequently. For example, a player has a 0/1/5 Kitty Pryde and he finally drew a 5* which turns out to be Kitty Pryde on the 40th pull; however, it was an unusable cover. He's likely to get  angry and start a topic saying the game is rigged.

    2) if a player finally draw a 5* in Classic on the 40th and it turns out to be Banner or Wasp, he will likely think that the game was rigged to give him useless 5* cover with the most worthless ability.

    I can see  topics like these popping up in other games forums every now and then, even when streakbreak is implemented for the legendary tier.

    End of the day, I think players should simply learn how probabilities actually work. 
  • sambrookjm
    sambrookjm Posts: 2,166 Chairperson of the Boards
    HoundofShadow said:
    If a 5* can be guaranteed within 7 pulls, what's the point of having 5* in HfH?

    By guaranteeing a 5* within 7 pulls, it will affect:

    1) cost of buying opened powers of 5* (720cp)
    2) cost of Latest 5* (500CP) and Classic 5* (250CP)
    3) reduce the value of 5* because they are easier to get than 4*
    4) affect the gameplay for 1*-3* players

    If this is implemented, the number of CP/LT/HT/HP that can be earned in game will be reduced dramaticallt to balance the game as a whole.

    Another reasoning is if 140 CP can guarantee me a 5* cover and 6 4* covers, the cost of no. 1 should be reduced to not more than 420CP and no 2. should be reduced to 120 CP and 80 CP respectively.

    Depends on if you're going for classics or new-characters. HfH for the classic 5-stars will still be WAY cheaper than getting them through tokens thanks to dilution. For the latest ones, it's 500 CP for a specific cover. Let's say it's done so that each 5-star cover is also evenly distributed in the newly non-random distribution pattern:

    First seven tokens - Six 4-stars, then Cover 1 of New 5-star 1
    Next seven tokens - Six 4-stars, then Cover 1 of New 5-star 2
    Next seven tokens - Six 4-stars, then Cover 1 of New 5-star 3

    21 tokens, one cover for each of the new three stars.

    Six 4-stars, then Cover 2 of New 5-star 1
    Six 4-stars, then Cover 2 of New 5-star 2
    Six 4-stars, then Cover 2 of New 5-star 3

    Best case, as always, is 25 CP from a token to get that cover if you happened to be on that particular slot in the rotation.  Worst case is Cover 3 of New 5-star 3, which would take 63 token pulls, at 25 CP, or 1575 CP. To get those three covers from HfH, it would be 1500 CP, so that extra 75 CP essentially gets you six more 5-star covers and 54 4-star covers, and (most likely) three bonus heroes.  That's not to say that people couldn't/wouldn't do both. Whales gonna Whale, and want those champed 5-stars ASAP.

    The 140 CP for seven Classic Legends tokens would guarantee you a Classic 5-star, of which there are 28 of them once Doom goes into LT in a couple of weeks.  This means you would need 28*140, or 3920 CP to get a guaranteed single cover for a particular classic, or close to 12,000 CP for one of each cover for the classic ones.  (This analysis ignores bonus heroes. Pulling 196 tokens would get you seven bonus heroes - six 4-stars and one 5-star.)


    HoundofShadow said:


    Also, pity timer or streak breaker for the rarest tier implemented in other games don't happen within the display odds. For example, if streak breaker is implemented in MPQ, it's likely to be a guaranteed 5* by the 40th or 50th pull, rather than by the 7th pull before the timer resets.

    Having a streakbreaker don't necessary solve perception problem because like what was mentioned,  players are likely to be emotionally charged when things don't go their way. The devs might probably have new charges laid upon them after implementing streakbreaker (assuming a guaranteed 5* on the 40th pull) such as:

    1) the game was rigged to give them unusable covers or useless frequently. For example, a player has a 0/1/5 Kitty Pryde and he finally drew a 5* which turns out to be Kitty Pryde on the 40th pull; however, it was an unusable cover. He's likely to get  angry and start a topic saying the game is rigged.

    2) if a player finally draw a 5* in Classic on the 40th and it turns out to be Banner or Wasp, he will likely think that the game was rigged to give him useless 5* cover with the most worthless ability.

    I can see  topics like these popping up in other games forums every now and then, even when streakbreak is implemented for the legendary tier.

    End of the day, I think players should simply learn how probabilities actually work. 
    I'm not sure that anything would ever solve the perception problem.  People tend to remember things that happen in the extreme cases of probabilities ("Three five stars in a row!!!" vs "80 four stars in a row?!?") and not the middle-of-the-road stuff in between.  My Doom tokens weren't very memorable. I pulled 15, got two five stars (the last of whom was Doom, so I stopped pulling) and one bonus 4-star hero.  That's about as average as you get.  The most memorable thing about them was that they got my Nebula to the 4/4/4 cover distribution, which is the rarest combination for 12 non-saved covers.

    Since the pulls are predetermined, showing you "Your next pull will be Cover X for Character Y" would be great if the character was good, or if you need that cover.  If it shows you a Banner/Wasp...you'll be far less excited.  That would change the whining from "My last cover was ****!" to "My next cover will be ****!"

    I also could not agree more with your last sentence.  Small Sample Sizes really mess up people's perception of probability.
  • Kolence
    Kolence Posts: 969 Critical Contributor
    HoundofShadow said:

    <SNIP>

    End of the day, I think players should simply learn how probabilities actually work. 
    I think many players here on the forum understand well enough how probabilities work. Doesn't mean we have to like the currently chosen way(s) the probabilities are used in the game... :p
  • Richyyy
    Richyyy Posts: 305 Mover and Shaker
    KGB said:
    Why even go with that complex a system that still leads to some players getting more 5* than others if they get a 'lucky' run. You should just go to a fixed system for everyone.

    Lets say you wanted to go with 1/7 odds for a 5*. You just randomize the numbers 1-7 endlessly. So the sequence would look something like: 4531672|1523746| and so on with a '1' representing a 5* and everything else a 4* (you then randomly roll for which 5* or 4* you got).

    The guarantees exactly 1/7 draw rates for everyone for 5* characters and means the *longest* streak you can go without a 5* is 12 draws.

    This also represents a perfectly fair system. I know the idea is that 15% averages out in the long run but I've seen lots of posts by players who track their rates that show they are getting say 13% in over say 2000 draws. Doesn't seem like much but that means they got 260 5* instead of 300. If another player was equally extra lucky and got 17% they'd have 80 extra 5* covers which is a lot of undeserved 'good' or 'bad' luck depending on which side you are on.

    Or just sell them in 'boxes'. If it's supposed to be 15% if you pull long enough, then a box of 20 covers for 500CP/400CP containing a guaranteed 17 4*s and 3 5*s would take sequences out entirely and just compress that percentage. The only random chance you'd be taking is which 4* and 5* covers you'd get.

    You could sell those alongside the current system and let players make their own choice between the guaranteed 3 covers in 20 or taking their chances with the single tokens that could do better or worse.
  • bluewolf
    bluewolf Posts: 5,848 Chairperson of the Boards
    edited January 2019
    What about over the really long term?  What about a consistently low rate?  I started tracking all my pulls in April '16 (before that I had one single 5* cover that I sold, an OML).

    At 500 pulls I had 63 5's, 12.6%
    At 1000 pulls I had 130 5's, 13%
    At 1500 pulls, I had 192 5's, 12.8%
    At 2000 pulls, I had 259 5's, 12.95%
    Recently I hit 2500 pulls, with 331 5* covers, 13.24%
    Currently up to 2530, 333 5's or 13.16%

    I know that this is probably not significantly high in terms of getting enough pulls to sort of expect an evening out, and I know that my results are within the range of possible outcomes blah blah blah.

    THIS IS 3 YEARS OF ME PLAYING THIS GAME.  It's a lot of play and pulls and I'm only a low level spender (buying some bundles and some HP bonus offers, and VIP).  Maybe I'm moderate among all players.  Obviously there are people who get a lot more pulls via spending, but why would I ever think about doing that with my low rate?

    I am about 44 covers lower than a 15% rate would have predicted.

    There is no current alleviation for my shortfall.  I have a number of 5's that are 1-2 covers short that might have been finished if my rate wasn't low.  Feeders have helped a little but it would be nice if there was some mechanism or service.

    My proposal might be that if you have a low rate like me, over a large enough number of pulls (my number seems sufficiently large), there might be some special 5* store you could get tokens to, via a customer service request (I don't expect them to push them out manually since I assume each account would have to be analyzed).

    I also actually don't expect this to every happen since everyone would probably bug CS to see if they could get some free 5's, even if it were 1-2.  I'm not sure I have a solution that makes sense without causing a lot of work and headaches unless something was implemented going forward.  I'm still not sure what that looks like other than some longer range system where you might get to "catch up" if you aren't on target after 100 pulls or whatever that number looks like.

    I suspect the devs worry about implementing system like that, though, that would be nice for people who don't spend massive amounts but would create a spending disincentive for those whales who will otherwise just keep spending to get what they want.  

    Although I fully expect that people who do spend a lot of money would request and receive more 5's if they fell short like I have.

    I shall, I suppose, resume trying not to think about my lousy account which apparently was marked with "low 5* rate".  My champ 4 roster is sufficiently developed that I pull enough to finish all 5's at this point, so yay.
  • helix72
    helix72 Posts: 997 Critical Contributor
    Great discussions, and I really appreciate all the community input.

    For most people, fear of loss is greater than desire for gain. That is why people trade risk all the time, and why there is usually a premium charged for it. An insurance company only expects to pay out $0.95 for each $1 you pay them, but because you don't want the downside risk you gladly pay the premium. Or said another way: imagine each year there is a 9.5% a chance you could have a $1M loss. Your insurance company offers to sell you a policy for $1,000 and in exchange, they take that risk. Most people take that trade all day long.

    As for the other ways to get 5*s, I don't think they're the same because what you'll notice is more control = more cost.

    1) 720 CP lets you pick the exact character and exact color cover you want whenever you want (as long as you already have at least 1) and as many as you want (up to 5, of course)

    2) 250 CP (500 if new) lets you choose for a limited time from a developer-selected specific character and color cover combination exactly one and only one of those covers

    3) the rest are random (you control when you draw, but when you get a 5* and what color it is are random, BH at least gives you some control over character in those 15% x 5% = 0.75% situations)

    For me, I'm looking for some downside protection that isn't too expensive. Obviously I'd love if they got more generous, but for a moment assuming that is off the table, here's where my vote would be that would keep the net effective rate the same:

    Lower the pull rate to 13.2%
    Institute a futility cap at 14 (going 0 for 14 guarantees the 15th pull is a random 5*). Why 14? I like knowing that if I spend 300 CP on classics it guarantees me at least 1 random 5*, maybe more. It just feels elegant. And, with this combination of pull rate and futility cap, you'll get a 5* without hitting the cap 80% of the time.

    Does the chance of getting at least 2 in a row go down? Of course, in this combination, from 19% to 15%. So 4% less, but 20% of my futility gets capped. I like it.

    Next I'm going to look at how options to increase the BH percentage would require decreasing the overall pull rate, because I think I'd also make that trade. If 15% of the time you get a 5* and 5% of those times you're getting a BH, you're roughly getting 0.15 x 1.05 = 0.1575 5*s per pull. But of those, 0.15 / 0.1575 or 95.2% of the 5*s you get are random (granted, pulling from latest it's 1 of 3 but I'm a completionist and want a better way to get classics). I'd prefer having more control over which 5*s I got. Ignoring a futility cap, consider:

    15% pull rate + 5% BH percentage = 15% x (1 + 5%) = 15.75%
    10.5% pull rate + 50% BH percentage = 10.5% x (1+50%) = 15.75%

    Dropping the pull rate to 10.5% though would definitely make a futility cap even more important. LIS, I'll work a bit on the numbers and some scenarios. I doubt we'd get a 50% BH rate, even with a drop in overall pull rates.

  • Banquetto
    Banquetto Posts: 29 Just Dropped In
    edited January 2019
    Richyyy said:
    Or just sell them in 'boxes'. If it's supposed to be 15% if you pull long enough, then a box of 20 covers for 500CP/400CP containing a guaranteed 17 4*s and 3 5*s would take sequences out entirely and just compress that percentage. The only random chance you'd be taking is which 4* and 5* covers you'd get.
    That would be like the way "Magic: The Gathering" cards were sold, yeah? You could buy a single pack and take your chances, or you could buy a full box and have a guaranteed number of the higher rarities in there?
    This idea would also give you the choice between "the grind" and "the gamble". Save up for a box if you want reduce the element of chance. Buy one token at a time if you feel lucky.
  • Kolence
    Kolence Posts: 969 Critical Contributor
    helix72 said:
    [...]

    Next I'm going to look at how options to increase the BH percentage would require decreasing the overall pull rate, because I think I'd also make that trade. If 15% of the time you get a 5* and 5% of those times you're getting a BH, you're roughly getting 0.15 x 1.05 = 0.1575 5*s per pull. But of those, 0.15 / 0.1575 or 95.2% of the 5*s you get are random (granted, pulling from latest it's 1 of 3 but I'm a completionist and want a better way to get classics). I'd prefer having more control over which 5*s I got. Ignoring a futility cap, consider:

    15% pull rate + 5% BH percentage = 15% x (1 + 5%) = 15.75%
    10.5% pull rate + 50% BH percentage = 10.5% x (1+50%) = 15.75%

    Dropping the pull rate to 10.5% though would definitely make a futility cap even more important. LIS, I'll work a bit on the numbers and some scenarios. I doubt we'd get a 50% BH rate, even with a drop in overall pull rates.

    Wow, your example is 7x more BH than as is? That is way too generous for this game...  :D

    Improving BH rate reduces randomness of what characters players get. It gives players more control over that part of it, and as you said, more control may well mean higher cost. The higher the rate of BH, the easier for FTP players to focus on just a few characters and make them much higher level than they ever could currently, without paying (heavily) with real money. So, yeah... 50% is, and probably always will be, way too much.

    The biggest problem with 5* BH for me is, they add another layer of randomness to the already random base pulls. Let's see the expected number of covers, say over a 1000 pulls and within two standard deviations, for your example and the current rates:

    10.5% pull rate, with 50% BH or 5.25%
    expected 105 +/- ~19 and 52 +/- ~14 
    pulls [86 - 124], BH [38 - 66], combined [124-190]

    15% pull rate, with 5% BH or 0.75% 
    expected 150 +/- ~23 and 7.5 +/- 5.45
    pulls [127 - 173], BH [2 - 13], combined [129 - 186]

    Roughly speaking.

    So, with such a change, the resulting range actually increases a little. It's possible to get even fewer or even more covers than it is currently. Of course, in this particular case, I'd be happy with it because the sheer number of BH covers would let me finish a classic character I missed or beef up one of the more used characters I otherwise couldn't. At the cost of missing out on something else perhaps, but the choice would be mine. (And unfortunately, we're still stuck with Banners and Wasps of the game for a long period of time.)

    What I'd love too see is a system where BH could somewhat correct the very bad luck with base pulls, while leaving the good luck as it is (while it lasts... :) ), so the resulting range is a bit narrower. Adjust any part of formulas and odds so the overall rates are acceptable for the devs and "bean counters", just not so drastic a difference between really unlucky pulls and the luckier ones.

    Oh well...
  • AardvarkPepper
    AardvarkPepper Posts: 239 Tile Toppler
    I'm against the very idea of capping, as well as a number of "solutions" proposed in this thread.  Superficial "solutions" cost time and money to implement, and if they don't address the core issues (which caps don't), you end up even worse off than before.

    ==

    MPQ is a loot box game.  Random rewards and reward structure for different event types means players end up with different optimal routes in practice.  (So though 3* Doctor Strange, Iron Man, and Hawkeye are 3* priorities and 4* Rocket and Groot and Medusa are 4* priorities, any *particular* player in the early 2-3* transition may not have a 13-cover Iron Man, or for the early 3*-4* transition Rocket and Groot.)  Without the ability to stick very closely to an "optimal" solution, different players don't feel pressured into following a single simple path, so the game remains interesting in that players adapt to their roster in particular rather than simply following a single "optimal" path that would be an onerous chore - or so the theory goes.

    That said, the proposed "solutions" don't make MPQ any less a loot box game, they simply change a few details around.  And though one could argue these changes would contribute to player perception of "fairness", I say in the end they wouldn't really matter.

    Take the concept of a cap.  A lower bound is defined that affects players that would perform worse than a statistical "average".  This would reduce the - what is it, 500 Latest Legends pulls for a 99% chance at three fully covered 5*s?  So it would seem that this is an improvement, and in some sense it is.

    But what really happened?  A player's already investing hundreds of hours or thousands of dollars to get to the point that they're stocking 500 LL tokens.  If the player only needs 400 or even 300, on paper maybe that seems like a significant difference.  But practically, we're talking about, say, thirty hours a week for a solid year, with perhaps a few hundred dollars invested on top of that.  Players that wanted to make the 5* transition are "hooked" anyways.

    What's the incentive on the part of the developers to change the system?  What's the incentive on the part of the players to change the system?  Instituting any sort of cap still leaves MPQ as a loot box game; some players will still perform worse than statistically normal.

    What I'm getting at is, a vague "cap" solution serves nobody.  The question is, what is the desired rate of payout?  How long should it take a player to make the 5* transition, given different player circumstances and preferences?  Should a player that plays thirty hours a week, that spends $200 USD over the course of a year, be able to make that transition at the end of that year?  What about a player that puts in the same amount of time, but that doesn't have as much control over their schedule for optimal clear times?  What about a player that puts in the time and money investment but doesn't study how to play optimally?

    Then I'd say - consider the rewards systems that are in place, and event structures.  Mind I don't intend to criticize the developers; there are factors I'm leaving off mentioning here (such as costs of servers and trying to spread players out, monetization and licensing costs, development and research costs, &c - suffice it to say I'm leaving a lot out.)  But I think considering the current reward structure, character dilution, &c, it comes down to something like this:

    1)  Players spend a long time in the 3*-4* transition, and even longer in the 4*-5* transition.  But except for the first relatively few key characters in a tier, there is not much perceived benefits to additional character covers.

    Specifically - let's say a new player is in the 2*-3* transition.  Say that player reads some guides, say they get 3* Iron Man, 3* Doctor Strange, 3* Hawkeye.  Good, now they can handle a lot of PvE.  Now let's add in Kamala Khan, Captain America, Hood, Black Widow, Thanos, Deadpool - which each new key addition, players gain unique abilities / combinations to their roster.  But only to a point.  Sure, 3* Magneto and Blade are probably good additions, but what about Punisher?  Sentry?  Vision?  At some point, additional 3*s really don't *mean* much to a 3* player; they're not earning better rewards for those characters (apart from champion rewards), and because some have such niche use they're just not used.

    Then let's say the 3*-4* transition.  4* Rocket and Groot adds speed; 4* Medusa adds durability.  But then again, you add in a few more - Gamora, Vulture, and a few others - then what?  At some point, a player really doesn't care that they have 4* Sandman or 4* Elektra because again - there's no rewards (apart from champion rewards), and because they're so niche they're not used.

    What I'm saying is there's a perceived lack of significance to in-game gains at certain points in MPQ, and I think this perceived lack of progress is frustrating to some players.  I don't say it sums up the reason for asking for caps, but I think it's a factor, and I think it's something that should be addressed.  But note a "cap" does nothing to address this.

    2) Events like Deadpool Daily Quest Crash of the Titans, and SHIELD Training help a player transition - if not so much with specific covers, at least with infrastructure.  But with character dilution, it has become much harder for a player to get any *particular* 3* or 4*.

    Meanwhile, more and more and more and more characters are added to the game, and especially a new player comes to realize - they're paying more for roster slots, they need more and more roster slots, will they ever catch up?

    Again, a "cap" does nothing to combat this perception - or perhaps I should call it "reality", because that's what it is.

    3)  Players that research how to make the 5* transition into three thirteen-cover 5*s realize there's a huge investment.  Players that don't research still realize there's a huge investment involved.

    A "cap" does address this in part.  But really a cap is horribly horribly, well, I don't want to say "so what".  But really, so what?  You pull sixteen Wasp covers and four Okoye covers, is this okay?  No, you still got screwed, you know?  And now, instead of players looking at the core loot box mechanic and considering whether they want so much RNG loot box mechanisms in a game with ever-increasing dilution, now some players THINK they're happy because they're caps.  Oh, maybe those players might come around and realize that caps WEREN'T ENOUGH in a few months.  But in the meantime, what?  You have a good deal of developer time and money going into a proposed "solution" that in the end didn't really solve anything - and once the "solution" is implemented, well, nobody that actually was part of getting the "solution" implemented wants to say "hey, this really wasn't such a great solution after all."

    So what happens is, the REAL issues aren't addressed.  Everyone will clap their hands and say "yay!  what a wonderful victory!"  but what really got solved?  Well that's politics for you.  But instead of superficial solutions that are an easy sell, maybe some hard infrastructure solutions should be considered.

    ==

    Remembering that the developers are working under restrictions and circumstances we don't know about, and have information we don't have access to that may inform their decision making process - I'd propose the following.  Though to be clear I'm not saying they are "solutions"; that would imply they are "better" than what's already in place, and considering all the data I don't have access to, who knows if they're actually better or not?  But anyways -

    1)  More SHIELD training, a lot more.  And some variety of SHIELD training for 3*s as well.  As it is, there really isn't much incentive for a player to collect more and more and more and more and more characters, in terms of rewards.  You earn HP in events, then you have to keep spending HP on new characters, then you end up never using most of those characters, and those characters don't even get you anything much most of the time - and consider how much HP and iso and all it takes to level those investments, it just feels a bit pointless.  On the other hand, if players were regularly using different 4*s and 3*s and earning rewards - it might be a bit different.

    2)  Rotating limited selection of non-boosted characters giving bonus iso and tokens when used.  (So for each event there would be featured characters (super boosted), boosted characters, and "bonus" characters that gave tokens and/or iso).  Rather than players feeling so pressured to use boosted characters (if at all), instead there's a question of payouts, which I think may be more incentivizing.

    3)  More player ability to customize what covers they get.  (This, particularly with dilution of ever more characters being added to the game).  Increasing the odds on gold-starred favorited characters would do nicely.

    Say there are seventy 4*s in the game, and one in twenty tokens awards a favorited 4*. At that rate, it takes about 202 tokens to get thirteen covers on a *particular* 4* - and of course, there's no guarantee one would get 544 or 553.  The more 4*s are released, the higher that number gets, the harder it is for a player to champion a particular 4*, and make progress in the 3*-4* transition.

    4)  Ability to find and select each character (that is, each character of a three character team) by typing in text to search (by hero name and by affiliation), and being able to select from user-defined teams.  As it is now, it's crazy trying to keep track of characters of different tiers and boosts and levels changing as every new cover is added.  Plus the game stating one has to field an "Avenger" - well really now.  Some characters were Avengers in one incarnation but not another, so unless you're really a superfan, who can keep track?
  • HoundofShadow
    HoundofShadow Posts: 8,004 Chairperson of the Boards
    I thought feeders do a pretty good job of guaranteeing 5* covers for most of the five stars.

    Human psychology is a difficult thing to handle.

    Let's say if a streak breaker is introduced to make players happier, I doubt players are going to see a guaranteed 5* within 20 pulls. Working with what we have, the probability of getting a specific 5* in latest is an average of 21 Latest pulls. The streakbreaker can't be more attractive than HfH for Latest 5*, and maybe it can't be more attractive than the 720CP (29 LT or 36 Classics pull). 

    Realistically speaking, streakbreaker on the 7th, 14th, 21st pull is unlikely to happen. It's likely to be twice the price of Latest 5* in HfH, which is (500/25)*2 = 40 LT pulls.

    Will players really be happier if they know that if they didn't pull a 5* on the first 39 pulls, they will be guaranteed a 5* on the 40th pull?
  • AardvarkPepper
    AardvarkPepper Posts: 239 Tile Toppler
    edited January 2019
    justsing said:
    Perception matters. 
    Think on the proposed solution. You're not stating that the game should be more generous with rewards (though that would also increase player perception of progress).  You're not stating that players should have more control over what particular characters they pull (which would help players build their rosters as they wished).

    No, there's a very narrowly defined solution that involves players pulling a guaranteed 5* after a certain number of pulls.

    And does this address dilution, or player ability to progress in game towards having a stronger roster more quickly?

    No.

    Rather, it simply addresses those players that feel they're not getting a "fair" number of 5*s.

    And what of it?  As I noted, there's still a strong RNG element.  A cap offers zero protection against results like opening a couple hundred tokens and getting sixteen Wasp covers, fourteen Loki, and three Okoye (or whatever).  And do you think players would consider such a distribution "fair"?

    I assert they would not.  There would still be fundamental unresolved issues with RNG even after a hypothetical cap were implemented.  Instituting a cap would simply be effort directed towards a stopgap solution that, in the end, wouldn't *really* improve core player satisfaction.

    If you want to say I'm wrong, so be it.  But think on it.  Sixteen Wasp, fourteen Loki, three Okoye.  Is that okay?

    If you say "Yes!"  well - that says something.

    If you say "No!"  well - that also says something.
  • justsing
    justsing Posts: 512 Critical Contributor
    AardvarkPepper said:
    justsing said:
    Perception matters. 
    Think on the proposed solution. You're not stating that the game should be more generous with rewards (though that would also increase player perception of progress).  You're not stating that players should have more control over what particular characters they pull (which would help players build their rosters as they wished).

    No, there's a very narrowly defined solution that involves players pulling a guaranteed 5* after a certain number of pulls.

    And does this address dilution, or player ability to progress in game towards having a stronger roster more quickly?

    No.

    Rather, it simply addresses those players that feel they're not getting a "fair" number of 5*s.

    And what of it?  As I noted, there's still a strong RNG element.  A cap offers zero protection against results like opening a couple hundred tokens and getting sixteen Wasp covers, fourteen Loki, and three Okoye (or whatever).  And do you think players would consider such a distribution "fair"?

    I assert they would not.  There would still be fundamental unresolved issues with RNG even after a hypothetical cap were implemented.  Instituting a cap would simply be effort directed towards a stopgap solution that, in the end, wouldn't *really* improve core player satisfaction.

    If you want to say I'm wrong, so be it.  But think on it.  Sixteen Wasp, fourteen Loki, three Okoye.  Is that okay?

    If you say "Yes!"  well - that says something.

    If you say "No!"  well - that also says something.
    How many 5*s you draw in x number of pulls versus the distribution of those 5* covers are separate RNG issues. No one is saying a streak breaker will magically fix all of the issues related to RNG. At the very least, it will reduce some of the variability in 5* pull rates. 

    Maybe it doesn't go far enough for you, but it's something. RNG is a big part of this game, and the issues associated with it are not going away anytime soon. The boosted odds for the Latest twelve 4*s were a reasonable way to fight dilution, but the devs went and removed that. 
  • AardvarkPepper
    AardvarkPepper Posts: 239 Tile Toppler
    justsing said:

    Maybe it doesn't go far enough for you, but it's something.
    I see I haven't made myself clear.

    Yes, where different players draw the line on RNG is subjective.  Where I draw the line and where you draw the line are different.  Thinking that, one might say "well, if all views are equal, then if one thing is done or another thing is done, or nothing is done, any course of action is equally valid."

    But an argument for mechanics that will boost 1) overall player progress and 2) player ability to select what characters they get is not a purely subjective matter.

    Rather, there are at least two objective matters that make the game different for newer players versus older players that I think the developers may want to consider.

    ==

    First, consider the time scale involved for players to feel they're making significant progress in game.  An older player that has lots of championed 4*s, say, is in a different situation compared to a new player in the 2*-3* transition.  The older player has an infrastructure that lets them earn consistent rewards, so dilution doesn't affect them as much except as it applies to championing older 5*s (if they don't have them already championed) or championing new release 4*s.  The new player, however, is in a significantly different situation because of character dilution.  As I mentioned earlier, players with developing rosters that are transitioning between tiers do best to get more covers on relatively few carefully selected characters - for example, 3* Doctor Strange and 3* Iron Man in the 2-3* transition, or 4* Rocket and Groot and Medusa in the 3*-4* transition.

    And those few first selected characters in a tier make a big difference.  A few carefully selected championed 3*s let a newer player play in SCL 7 so they can earn selected 4* covers (again, not just RANDOM 4*s, but ones they specifically want).  A few carefully selected championed 4*s let a player cut their time investment in the game by 25%-50%, and/or enable a player a viable option of playing at higher SCLs.

    But thanks to character dilution, it takes newer players longer to create those first few selected champion characters in a tier, than it took for older players to do.

    So if this game was ever designed with time scales in mind (and I would imagine it would have been) - then the time scales are now shifted.  Where before perhaps a player was supposed to spend 1 month in the early 2*-3* transition, now perhaps it's 1.5 months.  Where before perhaps a player was supposed to spend 6 months in the early 3*-4* transition, now perhaps it's 11 months - thanks to dilution.

    Then, there's players selling off excess covers for iso - with more dilution, a newer player with limited roster slots ends up selling more covers.

    All of this doesn't affect veterans so much.  A broad roster of tier characters won't change a player's ability to play up or down a tier with the addition of one or two champions of the same tier - considering especially the more important "meta" characters should already have been championed.  And veterans with plenty of roster slots won't need to sell excess covers for iso.

    But it does affect new players.  And again, if the game was designed with time scales for player progress in mind, this is significant because character dilution alters the time needed to reach milestones.

    ==

    Second, consider the "catch up" mentality.  Let's say we do want long-time veterans and spenders to have an advantage.  How much of an advantage do we really want them to have?  So much that newer players look around and realize they're hopelessly behind?   Because that's what we see.

    Imagine you're playing in PvE, and let's say you're at least a bit observant.  You quickly discover you get 4* covers at progression rewards at SCL 7 (and for newly released characters achievable placement rewards as well).  So a new player quickly tries to target SCL 7, at least for select events.  But what do they then discover?  I mean really?  The top ten places, where all the really decent prizes are at SCL 7, are filled with players with championed 5* and deep benches of 4*s.  Further, even going pretty far down the line and checking out the top 50 shows players with highly developed rosters, who benefit from synergetic boosted 4* or better champions.  As a player in the 2*-3* transition, you know you can't beat those times - and you will not be able to either.

    Granted, you could research coordination so you could try to enter a bracket when it just formed.  But players capable of that much thinking, coordination, and planning, will know even better just how much of an uphill grind they're facing.

    Then what of PvP?  If you're a player with a deep bench of championed 4*s and/or 5*s, you can quickly push to earn a 4* based on *score*.  But if you're a player in the 2*-3* transition, unless you very deliberately restricted your level gains, you're going to be facing level 270 boosted synergetic 3* teams - and though I've heard a couple 5* covers won't screw over a new player's PvP MMR any more, you still get this insane mismatch because boosts aren't factored in (much less synergy).  Instead of quickly smashing to a good prize based on score, developing players need to earn based on progression - and though it's a welcome alternative, it still takes much much longer.

    And for both PvE and PvP remember as well that players with developing 2*-3* rosters take longer to finish each match because they don't have developed championed characters in the 4* tier - much less synegetic boosted teams.

    A difference in roster strength between players that dedicated themselves to the game and/or spent, versus players that didn't do so is expected.  But considering how long this game's been going, and the aforementioned issue of newer players progressing even slower to hit those key milestones, the combined fact is - the grind is *objectively* longer for newer players to hit key milestones, when compared to how long it took older players to hit those same milestones.

    ==

    So rather than just saying "well that's how YOU feel" or "it's all subjective" - what I'm saying is, in real and very objective terms, new players have a tougher time of it than veterans did.  And objectively, it's getting worse, as more and more characters are released, and no new mechanics introduced to compensate.

    I'm not saying those two things are the sum total of what I regard as issues with MPQ.  But if someone's going to implement a solution, let it be a solution based on objective measures rather than subjective measures.

    One player says we don't need caps, another says we do need caps, another says we need caps and something else, another says we don't need caps but we do need something else - where does it end?  It can only end when design deliberately incorporates measurable factors in determining player progress.  (As I mentioned, that may have been done once, but there have been no compensatory factors for increasing character dilution introduced into the system for a long time).

    Caps could be part of a redesign, but they would certainly be insufficient of themselves.  That's what I'm saying.
  • helix72
    helix72 Posts: 997 Critical Contributor
    edited January 2019
    There is no perfect system. And when there is no perfect system, people tend to do nothing and argue for no change because fear of messing it up outweighs the possibility of improvement. While I understand the devs can't just give everyone 1 fully champed copy of every character (else why would we play?) I for one do like the idea of limiting frustration and helping speed progression to a degree. When the current meta favors a limited number of characters in each tier, my opinion (and it's just an opinion) is that we should look to reduce deciding who's in and who's out due to luck (my random draws gave me a champed Wasp while yours gave you a champed GladiaThor).

    With that in mind, here's my 5* proposal to move us in a favorable direction (not magically make everyone happy):

    CURRENT
    Pull rate: 15%
    Futility cap: none
    Bonus hero percentage: 5%
    Net effective pull rate: 15.75%

    PROPOSED
    Pull rate: 12.5%
    Futility cap: 15 (16th pull guarantees a 5* if 0 for last 15)
    Bonus hero percentage: 10% (applies to all 5* pulls including Futility Cap)
    Net effective pull rate: ~15.62%

    Explanation:

    Pull rate: A 12.5% pull rate is exactly 1 in 8. Seems elegant. I went with a flat pull rate over a progressive one, because even though I liked the idea of progressive it might be harder to implement and it reduces volatility a lot more. I suspect any palatable scenario needs to leave meaningful volatility in the process.

    Futility cap: You'll never go more than 15 without a 5*, and notice 16 is 2 x 8 which fit nicely with the 1 in 8 pull rate. So you'll average 1 in 8 but at a minimum 1 in 16. Elegant again! You'll have much shorter futility streaks, whereas before you could go 70+ pulls without a single 5*.

    Bonus hero percentage: Bonus hero chance also applies to futility cap guarantee, meaning if your 16th pull is a 5* because the cap kicked in there's still a 10% chance of getting a BH. And with the increase in the BH rate, you go from 4.8% of the 5*s you get being characters you want to ~10.7%, which still leaves quite a bit of chance but will speed up completion of my 2/0/0 GladiaThor while I ignore my 3/1/2 Wasp.

    Net effective pull rate: It's a slight decrease overall, but that should be expected given we're getting more certainty and control.

    So how about it? Would anyone else like to see this change implemented?
  • KGB
    KGB Posts: 3,284 Chairperson of the Boards
    edited January 2019
    helix72 said:


    So how about it? Would anyone else like to see this change implemented?
    I'd vote for it. But I doubt the Dev's would. The simple reason is what you've proposed changes the 15% draw rate to a *minimum* 15% draw rate. In other words there is nothing in your proposal to put a drag on players who are over drawing 5* (ie 17% rate).

    That's why my proposal of using 7 numbers over and over again works better because it gives exactly the same draw rate to everyone. One player noted it was too generous giving a guaranteed 5* within a worst rate of 12 pulls. In that case it could be changed to a 100 numbers over and over with 1-15 being a 5* and the rest 4*. To prevent long droughts the game would break the 100 numbers into groups of 10 and seed each group of 10 with at least 1 5* (and half would have 2 5*). That would make longest drought potential to be 18 while still guaranteeing exactly 15% draw rate for everyone.

    KGB
  • HoundofShadow
    HoundofShadow Posts: 8,004 Chairperson of the Boards
    I think the cap should be raised to a minimum of 20 and not exceed 40. 7, 10 or 16 is too low, realistically speaking. It's not going to happen. Streakbreaker was not meant to make sure the displayed odds happen within the estimated number of pulls. It's meant to let player know that their streak of "bad" luck won't exceed the Xth pull. 


  • AardvarkPepper
    AardvarkPepper Posts: 239 Tile Toppler
    HoundofShadow said:
    I think the cap should be raised to a minimum of 20 and not exceed 40. 7, 10 or 16 is too low, realistically speaking. It's not going to happen. Streakbreaker was not meant to make sure the displayed odds happen within the estimated number of pulls. It's meant to let player know that their streak of "bad" luck won't exceed the Xth pull. 


    Why 20, not 16?

    Why is 4.6% the break point?  Why not 8.5%?  Why not 2.3%?

    Doesn't it come down to metrics?  How do different percentages affect overall player spending differently?  How can profit be maximized?  Or if strict short term profit is not the sole consideration, what about player satisfaction leading to player retention leading to more favorable word of mouth advertising leading to more players leading to more monetization?

    What, to your mind, are the priorities?  Short term profits?  Long term profits?  Player satisfaction?  What are you basing 20, not 16, on?

    I don't mean to put you on the spot, just to advance the discussion.

    KGB said:

    I'd vote for it. But I doubt the Dev's would. The simple reason is what you've proposed changes the 15% draw rate to a *minimum* 15% draw rate. In other words there is nothing in your proposal to put a drag on players who are over drawing 5* (ie 17% rate).

    It's really smart to consider the dev viewpoint.  But I think the devs might support capping - even "generous" capping for purely practical reasons.

    The way I read KGB's argument is, the developers by default don't want to hand out more rare loot, because the more rare loot players have, the less incentivized they are to spend.  Although I could be misinterpreting KGB's posts.

    But the way I think on it, roster slots are the big selling point.  Players get a rare cover, they don't want to sell it for iso, they want to keep it.  But they also don't want to discard what they already have, so they end up wanting to buy roster slots.  So if anything, I'd say the devs may want to INCREASE the 5* drop rate, because it may INCREASE spending.

    More, if players do have the perception that 5* covers are "progress" towards an end game state (and why not, because they sort of are right?), then player perception of progress is satisfied each time they get a 5*.  So they're encouraged to stay with the game rather than leaving &c.

    There's a lot of speculative back and forth I won't get into now, but I think capping wouldn't necessarily mean potential decrease in revenues.  It might.  It would be interesting to look at the metrics on this.

    Though of course the developers won't publish the metrics because the metrics amount to trade secrets is my guess. B)

Categories