Building a Better PVE Experience

JamieMadrox

It's no secret that PVE events, especially those introducing new characters, are incredibly broken. Long event times, short node refreshes, poor end times for subs and events. It's angered even the best of player, burned out many more, and made every PVE a grind at best and an angering and demoralizing experience at the worst.

Enter The Gauntlet. This is the PVE the vocal player base has been asking for. No placement rewards for individuals or alliances. No node refreshes. A PVE where all rewards are progression based, all nodes only provide points once, and players can play on their own schedule without having to worry about losing out or letting their alliance down. The response to this event was overly positive and nearly everyone had a lot of fun with it. So why can't all PVE's be like The Gauntlet? The answer is they can.

All that needs to be done is some minor tweaks to the current PVEs and we could do away with everything that makes PVE soul crushing grinds. Here's the short list of the three changes needed to make all PVEs better:

Progression stye reward structure: No individual rewards; all rewards come from nodes and progression
End grinding: The Gauntlet style nodes, but with no random 20 ISO rewards.
End player exclusion: Provide a base level, single cover, essential character for the essential nodes that require them.

So how do these translate to the events?

Progression Style Reward Structure
Let's just toss the current reward structure for individual players and move to a mainly progression based reward structure like what we had in The Gauntlet. Make the final progression reward the cover, and make the rest of the rewards a combination of the individual and progression rewards that we get now. So for the current PVE (Dark Avengers: Heroic) it would look something like this:

140,000: (placement)
130,000: 500 (placement)
120,000: (placement)
100,000: 1000 (placement)
90,000: (progression)
80,000: 500 (progression)
70,000: Event (progression)
60,000: 1000 (progression)
50,000: :ares (placement)
40,000: 500 (placement)
30,000: (progression)
20,000: 500 (progression)
15,000: (progression)
12,500: 250 (progression)
10,000: Event (progression)
8,000: 250 (progression)
7,000: (progression)
6,000: 250 (progression)
5,000: Standard (placement)
4,000: 25 (placement)
3,500: Stockpile (all) (progression)
3,000: (progression)
2,500: Standard (placement)
2,000: 250 (placement)
1,600: Power Boost (all) (progression)
1,200: 100 (progression)
800: Standard (placement)
400: 250 (placement)
200: Standard (progression)

The only rewards missing from here are , , 500 , and 75 . These four rewards can serve as the four rewards for the final node of the event. The node would be really hard to beat meaning the top players will still be able to get the covers and the players still developing their rosters/skills will be scaled out.

For alliance rewards there are two easy options.

• Make them event completion based: If two alliances have the same number of nodes completed, then fall back to number of rewards earned. If there is still a tie, fall back to first alliance to reach that point. This is by far the best option as the reward structure wouldn't need to be changed.
• Create alliance progression rewards: This would be easy to do as well, but the lack of variety in the alliance reward structure would make this boring. This could be made more exciting by taking some of the rewards from the individual progression and adding them here, but that may not sit well with all players.

End Grinding

This is simple. Just like The Gauntlet make each node worth a fixed number of points the first time it is defeated. After that, the node gives no further points, but will give up to 3 more rewards. Also removing the random 20 reward for successive completions of a node would be a big step towards less grinding on the players' part. This also means no rubber-banding which also means no need to grind constantly to win only to have someone come out of nowhere in the last hours to win it all. Everyone has a fair chance to make the points needed to get the progression rewards.

End Player Exclusion
This is another simple one. For essential nodes that require a featured character, provide everyone the option of using a single cover version of that character like what is done in feature character PVP events. This will make all players feel included and still give the players with better developed rosters an advantage.

Scaling can be used to control how far players can make it through the event thus limiting the top rewards to the players with the most developed rosters. This is nothing different from how events work now and is a fair way to reward everyone appropriately based on their "skill level". Overall I think a move to PVE events that use this format would be highly beneficial for the players and developers. It would kill a lot of the complaints that players have now about PVE and increase engagement in the events. Revenue would still be generated by players with less developed rosters purchasing consumables and covers to get an edge and progress farther, much like they do now, and with increased engagement it could even generate more revenue than current PVE events.

over_clocked

There needs to be something for the newbies (more 'equality') but hell yes, no refreshes and no 20 ISO non-rewards!

Unknown

Gauntlet does highlight the scaling based on roster doesn't work well when there isn't community scaling to bail you out later (by making everything impossible for everyone so that it's fair) but it's probably a better event type than anything else. That said having everything with +80 levels on it compared to a max 166 roster for having a max level X Force is quite discouraging too.

Unknown

Phantron wrote:

Gauntlet does highlight the scaling based on roster doesn't work well when there isn't community scaling to bail you out later (by making everything impossible for everyone so that it's fair) but it's probably a better event type than anything else. That said having everything with +80 levels on it compared to a max 166 roster for having a max level X Force is quite discouraging too.

I'll be the one to say it, and in sorry to have to be that guy, but either your part of the solution or part of the problem.

Feel free, to offer suggestions on how to solve this problem. You got a great grasp of the have and it's mechanics, so you could probably offer great insights on what further tweaks could make the pve work better for everyone. In other words, I expect more from you than just poking a few holes in everyone else's theories. You should be able to offer suggestions on how to fix the holes you find. If you can't, im nit sure who can.

NorthernPolarity

Your idea seems pretty cool. except for this one little snag:

Jamie Madrox wrote:

Scaling can be used to control how far players can make it through the event thus limiting the top rewards to the players with the most developed rosters. This is nothing different from how events work now and is a fair way to reward everyone appropriately based on their "skill level".

Getting scaling right to this level of precision seems very, very difficult, and getting it wrong is disasterous. For example, as Phantron said, the people with max level X-Forces were at a disadvantage in the Gauntlet because that added + 80 levels to all of the nodes. This didn't matter as much in that PvE since the final rewards didn't matter to a guy with max rosters, but balance as you said is tuned to the degree of the rate at which they're handing out covers normally (top 1% get 3 new character covers, top 2% get 2 new character covers, so on and so forth), then that means that these nodes are going to be extremely hard. If the nodes are extremely hard and the optimal way to play the event is to deliberately underlevel your 4*s to combat scaling, then a large portion of the veterans like Phantron and I are going to be pissed when we can't the final progression awards because we leveled X-Force.

The cool thing about gauntlet was that they didn't care if they got scaling sort of wrong. They could balance the nodes to be challenging yet doable, and even if an extra 5% of people beat everything, they didn't care because they weren't giving out old character covers. Giving out 5% more Doc Ock covers, however, is something that they never want to happen, so then you get into this ridiculous balancing act where you're trying to make sure that 1% of people get 4 Doc Ock covers, 2% of people get 3 Doc Ock covers, so on and so forth. Honestly, this seems like infinitely more work and time than just having gauntlet events with relatively fewer rewards and standard PvE events, since placement awards guarantee Demiurge that only a certain amount of players get the covers.

If somehow Demiurge comes up with a mythical scaling algorithm that doesn't take any work to create/maintain, then this idea would be amazing. As is, I'm highly skeptical that they can pull off something like this correctly, let alone even try to do so. To me, this idea seems like something that a Demiurge employee probably thought up of at some point, told their boss, and immediately got shut down because fitting it to their current reward structure is incredibly difficult. There isn't a good solution for this, and interia dictates that it doesn't really make sense to overhaul your existing system and try to solve a very very hard problem when you could just play it safe/easy with alternating both event structures, and get probably around the same amount of revenue anyways. The only way I could see them even try to do this is if Gauntlet got so much more user engagement / revenue that it would be worth it to switch everything over to that structure, but that seems unlikely.

GrumpySmurf1002

Jamie Madrox wrote:

Scaling can be used to control how far players can make it through the event thus limiting the top rewards to the players with the most developed rosters. This is nothing different from how events work now and is a fair way to reward everyone appropriately based on their "skill level".

I'm not sure I entirely agree with this premise, but it might just be the word 'developed.' I don't have any max 3*s and I can and have easily competed for top rewards in PvE if I can put the time in. At the same time, I consider my roster 'developed,' in that I have all 2*s maxed and have a decent amount of 3*s that are 10+ covered.

Anyway, point is: I wouldn't want scaling to be altered such that now you need 166s to get the top prizes, because then it just becomes exactly like PvP, where the top prizes are completely unattainable for a 1-2* roster. I don't think that's what you mean, but it's ambiguous enough that I wanted to chime in.

Nellyson

You and I are on the same page Madrox. This rubberbanding and constant grinding for new characters is just ****. I like getting the new characters. And recently, excluding , they've done a damn good job at making the new characters usable and interesting with powers that actually matter. with his fastball is just awesome! Who doesn't want to throw Thor into the air, have him land with pretty much any one of his powers and then have enough AP to have him activate another power! I freaking love it! But I totally agree that the PVE rewards need to be redone. Where is the PVE element to getting the new character? They did a good job with the Hunt story. I was genuinely interested in where they were going with the story and being able to get Hulk onto the good guys side. That was awesome! Why can't we have stories like that every month or two months to introduce a new character? Hell, I think even Ares as a was introduced in a cool fashion. They could keep doing these **** Heroics for all I care, as long as they also get that one story in.

As for your suggestion on rewards, there are still people that really like the PVP element of rankings for rewards. Who are these people?? Well, I think they hate themselves. Hahaha, but honestly, they could keep the same format. Just have a node that unlocks after you meet a criteria at which point you are awarded a cover. Why not?

Unknown

stephen43084 wrote:

Phantron wrote:

Gauntlet does highlight the scaling based on roster doesn't work well when there isn't community scaling to bail you out later (by making everything impossible for everyone so that it's fair) but it's probably a better event type than anything else. That said having everything with +80 levels on it compared to a max 166 roster for having a max level X Force is quite discouraging too.

I'll be the one to say it, and in sorry to have to be that guy, but either your part of the solution or part of the problem.

Feel free, to offer suggestions on how to solve this problem. You got a great grasp of the have and it's mechanics, so you could probably offer great insights on what further tweaks could make the pve work better for everyone. In other words, I expect more from you than just poking a few holes in everyone else's theories. You should be able to offer suggestions on how to fix the holes you find. If you can't, im nit sure who can.

Scaling is something completely beyond my or any player's control. That is it just doesn't look like we can convince them to change the scaling based on roster.

I did offer a suggestion before, that instead of scaling content upwards to roster strength, we should have static content and just scale the weaker roster up to the content. That is we can have all encounter at a fixed level that's challenging for a 3*, say level 300, and then the max 2*s will be boosted to level 140 and max 1*s will be boosted to level 125 which should give you comparable base character strength to 3*s. The advantage of this system is that if it turns out some 1*/2* are way too good, at least people with the max roster can simply use those instead.

mjh

In addition to what was said they need to address the tile match damage for above cap level characters. There should be a cap on tile damage, only the attack levels should increase. 1000+ damage for a basic match4 is just stupid.

_RiO_

NorthernPolarity wrote:

Your idea seems pretty cool. except for this one little snag:

Jamie Madrox wrote:

Scaling can be used to control how far players can make it through the event thus limiting the top rewards to the players with the most developed rosters. This is nothing different from how events work now and is a fair way to reward everyone appropriately based on their "skill level".

Getting scaling right to this level of precision seems very, very difficult, and getting it wrong is disasterous. For example, as Phantron said, the people with max level X-Forces were at a disadvantage in the Gauntlet because that added + 80 levels to all of the nodes.

Getting scaling right to this level of precision is a matter of using a weighed average of the levels of the roster members you are actively playing in the event, with a sanity check tied to the average of your entire roster to ensure that you are not 'cheating' by using all your one stars to pull levels down and then switching to 4*s to smash through nodes. (Actually; you could balance that quite well by lowering enemy levels less quickly than you would gain them, for instance.)

Let's start out with a roster average. Let's say for a roster that has 3 3-stars at 166, 1 4-star at 270 and 8 2-stars at 94.
That's (( 3 × 166 ) + ( 1 × 270 ) + ( 8 × 94 )) / ( 3 + 1 + 8 ) = 1520 / 12 ≈ 126
Your opponents' levels will never drop lower than this average.

Then let's start playing with the 4-star and 2 of your 3-stars. Each match played alters the weight factor for characters, giving played characters a positive bias.
After 2 matches:
(( 2 × 166 × 3 ) + ( 1 × 270 × 3 ) + ( 1 × 166 ) + ( 8 × 94 )) / (( 3 × 3 ) + 1 + 8 ) = 2724 / 18 ≈ 151

After 3 matches:
(( 2 × 166 × 4 ) + ( 1 × 270 × 4 )+ ( 1 × 166 ) + ( 8 × 94 )) / (( 3 × 4 ) + 1 + 8 ) = 3326 / 21 ≈ 158

After 9 matches:
(( 2 × 166 × 10 ) + ( 1 × 270 × 10 )+ ( 1 × 166 ) + ( 8 × 94 )) / (( 3 × 10 ) + 1 + 8 ) = 6938 / 39 ≈ 178

Notice how this doesn't enter a completely runaway state even with the 270 in there?
Ofcourse, this would still climb faster with another 270 in there. (And it would start out at a higher initial average.)


Let's look at the same progression for someone with 2 3-stars at 120 and 8 2-stars at 94.
That player's roster starts out with ( 2 × 120 + 8 × 94 ) / ( 2 + 8 ) = 992 / 10 = 62.

And let's assume this player plays with both 3-stars and one 2-star.
After 2 matches:
(( 2 × 120 × 3 ) + ( 1 × 94 × 3 ) + ( 7 × 94 )) / (( 3 × 3 ) + 7 ) = 1660 / 16 ≈ 104

After 3 matches:
(( 2 × 120 × 4 ) + ( 1 × 94 × 4 ) + ( 7 × 94 )) / (( 3 × 4 ) + 7 ) = 1994 / 19 ≈ 105

After 9 matches:
(( 2 × 120 × 10 ) + ( 1 × 94 × 10 ) + ( 7 × 94 )) / (( 3 × 10 ) + 7 ) = 3998 / 37 ≈ 108

Notice how that doesn't climb quite as fast to the limits of the roster? Notice how it also balances closer to the maximum roster level?
That's mostly because in the other case, the level 270 4-star was an extreme outlier and its influence was marginalized. That is; until the weighed factor started boosting it when that 4-star was being played.

Now one of the 3-stars gets knocked out in the last match and the player is forced to rely on two 2-stars instead. And after 2 more matches:
(( 1 × 120 × 10 ) + ( 1 × 120 × 12 ) + ( 1 × 94 × 12 ) + ( 1 × 94 × 2 ) + ( 6 × 94 )) / (( 1 × 10 ) + ( 2 × 12 ) + ( 1 × 2 ) + 7 ) = 4520 / 43 ≈ 105
The level will be coming down again to compensate...


Ofcourse, the weight factors and the speed at which they change are up for tuning and you could incorporate historical information for the initial weights into this as well.
Really, it's all a matter of starting out with the proper mathematical model in the first place and then tuning weight factors from there. I get the distinct feeling that this is something Demiurge got wrong and that they end up having to tune constants or variables in more unstable parts of the formula; factors that have wildly unpredictable impact at the extremes.

emaker27

NorthernPolarity wrote:

Your idea seems pretty cool. except for this one little snag:

Jamie Madrox wrote:

Scaling can be used to control how far players can make it through the event thus limiting the top rewards to the players with the most developed rosters. This is nothing different from how events work now and is a fair way to reward everyone appropriately based on their "skill level".

Getting scaling right to this level of precision seems very, very difficult, and getting it wrong is disasterous. <snip>.

I always found the "it could be difficult" excuse very lame. And there were plenty of issues with the current PvE reward structure months ago. Sometimes very few if any got the max progression reward, and sometimes too many. And then they finally figured out how to balance it correctly.

I would love to see the changes the OP outlines, but ultimately D3 has access to statistics that we don't. And I'm sure they've used those stats to deem the current structure as financially sound. And for a company to ditch that for an unknown new system would be extremely rare.

NorthernPolarity

_RiO_ wrote:

[lot of good theorycrafting on scaling for a single player]

I'm not saying that it's incredibly difficult to make scaling reasonable for a particular player, I'm saying that it's very difficult to tune scaling to a point where you're giving out similar amounts of covers as you are now.
That is, look at the placement award ratios right now for the current PvE. D3 is guaranteed that:
2 / 1000 players will get 1 fury cover
10 / 1000 players will get 3 doc ocks
20 / 1000 players will get 2 doc ocks

so on and so forth. If we change to a pure progression model the way that OP says, this means we need to change SCALING in such a way that those ratios (2 / 1000 gets 1 fury cover, 10 / 1000 gets 3 doc ocks, so on and so forth) are accomplished. Do you guys see how much harder this is to implement? You need to tune the scaling and levels so that those ratios (or similar amounts) are achieved, and that's insanely hard because its completely non-deterministic: you have no idea how well/poorly your players will do in the event, so theres no way to make sure that you aren't suddenly giving out fury covers to 20% of the population.

As long as d3 cares about how many new/important covers they're giving out (and that seems pretty important given their current business model), this problem is always going to exist, and since it's incredibly hard to solve, why would they bother when they can just leave it as is. Maybe if d3 suddenly feels like Oprah and doesn't mind handing out 3 doc ock covers to 30% of their playerbase then this could become feasible, but until that day, it's not happening.

NorthernPolarity

emaker27 wrote:

NorthernPolarity wrote:

Your idea seems pretty cool. except for this one little snag:

Jamie Madrox wrote:

Scaling can be used to control how far players can make it through the event thus limiting the top rewards to the players with the most developed rosters. This is nothing different from how events work now and is a fair way to reward everyone appropriately based on their "skill level".

Getting scaling right to this level of precision seems very, very difficult, and getting it wrong is disasterous. <snip>.

I always found the "it could be difficult" excuse very lame. And there were plenty of issues with the current PvE reward structure months ago. Sometimes very few if any got the max progression reward, and sometimes too many. And then they finally figured out how to balance it correctly.

I would love to see the changes the OP outlines, but ultimately D3 has access to statistics that we don't. And I'm sure they've used those stats to deem the current structure as financially sound. And for a company to ditch that for an unknown new system would be extremely rare.

Well, the progression awards were changed so that now anyone who starts on the first sub is basically guaranteed all the progression awards. The current problem (tuning the scaling such that 0.2% of the population gets the progression award of Fury, 1% of the population gets the progression award of 3 Doc Ocks) is insane compared to that. It's actually kinda funny, because I'm sure that was their goal of progression awards in the first place, and tuning it ended up being so difficult (as you noticed with "issues with the previous PvE reward structure") that they gave up tuning it in that way, and just reduced all the rewards so that they didn't care if a lot of people got all of them (notice how in previous PvEs with issues, you could get multiple 3* covers off of progression awards, whereas now you can only get 1 max). This should be evidence enough that the problem is hard enough that d3 will probably never get it right should they choose this path.

Unknown

Let's say we got scaling solved by magic because otherwise a progression based PvE would never work because someone's going to have it way easier/harder just for having characters leveled a certain way, I can imagine a system like this. You'd have 3 sub brackets called Dr. Octopus blue/black/green. Completing all the nodes in each sub gets you a cover of that color, and it'd take some practice but you'll eventually be enough to make them hard enough to get the expected distribution. If the ability to choose which cover you get is not desireable then simply make the brackets unlocked in sequence instead.

For the 0.2% getting the 4* you can either say all the guys completing all 3 brackets gets a random 3* token with 1/5 chance being a 4*. Since there are usually 10 guys getting 3 3*s this preserves the same ratio. If we want to make it more skill based we can unlock some boss nodes in the main bracket and we can just group people into brackets of 1000 people by the same current mechanism, and the first two person to clear this gets the 4*s. But for this to make sense the boss nodes would probably need to have boosts disabled and Whales should not work on it. The boss nodes themselves should give a decent amount of iso (say 1000-2000) so that even if the top 2 is already taken in your bracket there might still be reason to try it.

_RiO_

NorthernPolarity wrote:

Well, the progression awards were changed so that now anyone who starts on the first sub is basically guaranteed all the progression awards.

So explain again why after consistently clearing all my nodes optimally since the very start of the event, for every free moment of my available time, I am still stuck at a sub 60,000 score and will never, ever be able to reach the requisite 120,000?

NorthernPolarity wrote:

If we change to a pure progression model the way that OP says, this means we need to change SCALING in such a way that those ratios (2 / 1000 gets 1 fury cover, 10 / 1000 gets 3 doc ocks, so on and so forth) are accomplished. Do you guys see how much harder this is to implement?

I only see that the conclusion you are alluding to is tantamount to stating the current business model can only be achieved by compromising gameplay to a level where it is intentionally left broken for a large amount of the playerbase. Think of that what you will.

Also, I'd like to add that with every increase in 3* population, the artificial induced scarcity through the distribution numbers Demiurge originally published breaks down further and further. The odds of earning the specific covers you need to bring a 3* up to level for use drop further with each additional character and with a fixed number of opportunities to earn character covers, that translates directly into a devaluation of all possible 3* covers, simply because it is going to continue to take longer and longer to complete a 3* and you have no idea which 3* it is going to be. In other words; the numbers that were originally published mean ****. (And probably for a great deal already meant **** when they were first published.)

NorthernPolarity

Phantron wrote:

Let's say we got scaling solved by magic because otherwise a progression based PvE would never work because someone's going to have it way easier/harder just for having characters leveled a certain way, I can imagine a system like this. You'd have 3 sub brackets called Dr. Octopus blue/black/green. Completing all the nodes in each sub gets you a cover of that color, and it'd take some practice but you'll eventually be enough to make them hard enough to get the expected distribution. If the ability to choose which cover you get is not desireable then simply make the brackets unlocked in sequence instead.

For the 0.2% getting the 4* you can either say all the guys completing all 3 brackets gets a random 3* token with 1/5 chance being a 4*. Since there are usually 10 guys getting 3 3*s this preserves the same ratio. If we want to make it more skill based we can unlock some boss nodes in the main bracket and we can just group people into brackets of 1000 people by the same current mechanism, and the first two person to clear this gets the 4*s. But for this to make sense the boss nodes would probably need to have boosts disabled and Whales should not work on it. The boss nodes themselves should give a decent amount of iso (say 1000-2000) so that even if the top 2 is already taken in your bracket there might still be reason to try it.

I would need to look at the statistics of the average player roster to see if getting the expected distribution is possible. Like, if even 3% of the playerbase has say a decked out roster, then this means that you would have to tune the scaling such that 1/3rd of those players gets 3 doc ock covers while 2/3rds of those players don't, and I don't see how thats possible given that players with a similar roster will probably complete roughly the same amount of subs, so this means that you would either see those max players all complete or fail any given sub. I guess you could make the events long enough such that 2/3rds of the max playerbase simply gives up because there are too many nodes or the nodes are too time intensive, but that seems like a feat in of itself.

Unknown

NorthernPolarity wrote:

I would need to look at the statistics of the average player roster to see if getting the expected distribution is possible. Like, if even 3% of the playerbase has say a decked out roster, then this means that you would have to tune the scaling such that 1/3rd of those players gets 3 doc ock covers while 2/3rds of those players don't, and I don't see how thats possible given that players with a similar roster will probably complete roughly the same amount of subs, so this means that you would either see those max players all complete or fail any given sub. I guess you could make the events long enough such that 2/3rds of the max playerbase simply gives up because there are too many nodes or the nodes are too time intensive, but that seems like a feat in of itself.

It's not nearly as hard as it looks. Imagine if the finale bracket was duplicated twice at the same difficulty that doesn't scale up any further and we didn't get the one day extension. I'm pretty sure most people would not have close to finishing all 3 finale brackets. Assuming scaling is fair in some way this means you only have to do this right once, because if you have a fair system of scaling (like scaling roster instead of scaling content) and you had the result you're expecting to get after one event, then you can be pretty sure that structure will work for all future events too. This is also why I think you should be able to pick the order you get to do those subs because otherwise someone who can consistently finish 2 but not 3 brackets will never get the third cover color, and while you can argue this already happens anyway, there's some illusion of hope for a 2 cover finisher to sneak into 3, but beating an extra bracket compared to normal would likely be out of the question.

NorthernPolarity

_RiO_ wrote:

NorthernPolarity wrote:

Well, the progression awards were changed so that now anyone who starts on the first sub is basically guaranteed all the progression awards.

So explain again why after consistently clearing all my nodes optimally since the very start of the event, for every free moment of my available time, I am still stuck at a sub 60,000 score and will never, ever be able to reach the requisite 120,000?

Because you're stuck in an unfavorable time zone, I would imagine. I guess I should have changed my original statement to "anyone who starts on the first sub and is able to play semi-consistently and grind the end of the sub".

_RiO_ wrote:

NorthernPolarity wrote:

If we change to a pure progression model the way that OP says, this means we need to change SCALING in such a way that those ratios (2 / 1000 gets 1 fury cover, 10 / 1000 gets 3 doc ocks, so on and so forth) are accomplished. Do you guys see how much harder this is to implement?

I only see that the conclusion you are alluding to is tantamount to stating the current business model can only be achieved by compromising gameplay to a level where it is intentionally left broken for a large amount of the playerbase. Think of that what you will.

Also, I'd like to add that with every increase in 3* population, the artificial induced scarcity through the distribution numbers Demiurge originally published breaks down further and further. The odds of earning the specific covers you need to bring a 3* up to level for use drop further with each additional character and with a fixed number of opportunities to earn character covers, that translates directly into a devaluation of all possible 3* covers. The numbers that were originally published mean tinykitty.

I'd like to see the statistics for how many paying EU players there actually are. Either there isn't enough for Demiurge to cater to you guys (in which case the way they're treating you makes perfect financial sense right now even though its ****), or there is a good amount and they're just completely incompetent.

Look, all I'm doing is accessing the likeliness of them actually implementing something like this. I'm assuming that the way things currently are exist for some sort of financial reason (since otherwise why would a company NOT want to make their customers happy), and justifying based off of those assumed financial reasons whether or not it would make sense to implement this. I mean, don't you think that some random dev at Demiurge thought, "Hey, it'd be nice if EU players could have a fair shot at things" or "Hey, it'd be cool if PvE could be all progression based so we can hand out covers solely based off of player skill and not off of stupid things like when they joined their bracket". The devs aren't evil bastards that are like "Yeah, lets add rubberbanding to completely screw over the playerbase!", they're probably gamers just like us passionate about what they're doing, so it's not like they didn't think of these ideas: there must be something thats preventing them from implementing them, and that something is probably opportunity cost in development effort or directly contradicting what they think will be best for the company's financial future. It's not like I WANT the system to be the way it is, I'm just saying that this change is unlikely to happen because of what I stated earlier. I'd love for them to say "screw all the logic behind what we did in the past year. Let's focus on player retention and change our business model to focus less on the new character release train and more on long-term stuff like new game modes and better events" but like most things, inertia (and probably financial feasibility) is probably going to prevent that happening, and I'm giving why I think that's the case. I hope I'm proven wrong since I think a system like this would be really cool, but logic dictates otherwise.

Unknown

The current (heroic DA) isn't really that time zone sensitive because it's so grindy that it's a lot more important whether you get your 15 cycles' worth of nodes in as opposed to doing them in the most optimal pattern. Yes missing the end of sub still sucks (I miss them both too) but if I was actually grinding like past events it wouldn't have mattered much because everyone gets very little from rubberband, even on the last cycle. Now I chose to not grind much for heroic DA and my place reflect that (started out #100 after first sub, went up to about 20 yesterday but probably fell back below 100 again since I had a block of 16 or so hours without playing).

NorthernPolarity

Phantron wrote:

NorthernPolarity wrote:

I would need to look at the statistics of the average player roster to see if getting the expected distribution is possible. Like, if even 3% of the playerbase has say a decked out roster, then this means that you would have to tune the scaling such that 1/3rd of those players gets 3 doc ock covers while 2/3rds of those players don't, and I don't see how thats possible given that players with a similar roster will probably complete roughly the same amount of subs, so this means that you would either see those max players all complete or fail any given sub. I guess you could make the events long enough such that 2/3rds of the max playerbase simply gives up because there are too many nodes or the nodes are too time intensive, but that seems like a feat in of itself.

It's not nearly as hard as it looks. Imagine if the finale bracket was duplicated twice at the same difficulty that doesn't scale up any further and we didn't get the one day extension. I'm pretty sure most people would not have close to finishing all 3 finale brackets. Assuming scaling is fair in some way this means you only have to do this right once, because if you have a fair system of scaling (like scaling roster instead of scaling content) and you had the result you're expecting to get after one event, then you can be pretty sure that structure will work for all future events too. This is also why I think you should be able to pick the order you get to do those subs because otherwise someone who can consistently finish 2 but not 3 brackets will never get the third cover color, and while you can argue this already happens anyway, there's some illusion of hope for a 2 cover finisher to sneak into 3, but beating an extra bracket compared to normal would likely be out of the question.

I remember being in ~the 700th place range for the first 2 brackets in the gauntlet and ~300 in the finale gauntlet having almost all of the nodes done, which means that at least for my finale bracket, probably ~100-200 people would have ended up getting the doc ock covers. If every bracket was as hard as the finale bracket, I would still think that more than 2% of the population would be able to finish all 3 brackets just based off of my one data point. Also if there is a prize that veterans actually want, I would imagine that most of us would try harder to get it, leading to more people getting the cover and farther away from the desired distributions.

I guess what they could do as a proof of concept is just give one, highly wanted cover (say, Doc Ock blue) in a gauntlet rerun as the cover for the final bracket, and tune the event until they can get the distribution that they want for that one cover. That could pave the way to the OPs initial suggestion of changing all events to this structure, at least from a technical standpoint. There would also have to be enough increased revenue off of the changed structure to justify converting every single PvE to this structure, so hopefully the gauntlet did that.

NorthernPolarity

Phantron wrote:

The current (heroic DA) isn't really that time zone sensitive because it's so grindy that it's a lot more important whether you get your 15 cycles' worth of nodes in as opposed to doing them in the most optimal pattern. Yes missing the end of sub still sucks (I miss them both too) but if I was actually grinding like past events it wouldn't have mattered much because everyone gets very little from rubberband, even on the last cycle. Now I chose to not grind much for heroic DA and my place reflect that (started out #100 after first sub, went up to about 20 yesterday but probably fell back below 100 again since I had a block of 16 or so hours without playing).

I'm at 80k points having grinded relatively sub-optimally for both subs, so to have a 20k lead over RIO who grinded optimally but missed the end of the sub because of EU times probably means that grinding the "end of the sub" matters a lot more than you think. I think for EU players, the end of the sub is like the last 8 hours of it since they need to sleep, so they're missing out on 2-3 clears at the end which probably accounts for that extra 10k I got per sub.