Should I wait to level past 94?

Unknown

Personally i held all my 3 stars at 110 until i had 11 or more covers for at least 8 core characters that i could rotate around with. There was no impact to my scaling as i built my strength and the 110s were slightly better than the 94s.

Too many people here have taken 1 character to 166 and instantly struggled in pve to believe it has no impact

Arondite

theHappyDance wrote:

You're right. I freely admit its not to the level of scientific credibility. It's simply reasonable observation based on my limited playtime on both accounts (neither of which has involved much / if any PvE or PvP).

If you want something more scientific, feel free to test it yourself instead of mouthing off. My experience is that claiming something is "unscientific" is what people do when they want to disagree with someone but don't want to actually be bothered to prove anything. To the best of my ability, I've presented what information on this I have available to me. If you want to spend a week or more doing thorough, truly scientific testing, at a variety of different character levels, roster strengths, placements, etc.. carefully documenting everything, I'll happily concede to your findings. I don't have time for that myself, but I'd be happy to see the results if somewhat wanted to perform actual rigorous testing.

What a cop out. You don't come forward to a community with a hypothesis, hear disputes or rebuttals and then say "well you test it I don't want to".

It's your hypothesis. If you want it proven, prove it. Otherwise (as previously stated) it's just a cool story.

Unknown

No, your idea of scientific testing isn't even remotely scientific.

What we know about scaling is that it's set primarily when you enter an event, but does change somewhat during the event if you level your heroes a lot. But in general, you notice scaling more in subsequent events than whatever one you're currently in.

If you want truly scientific testing, you need a reset button. Then re-enter with a variety of different circumstances - levels all the same at different levels. Levels varying. Leveling up mid-event vs. not at all. Modify the number of days played. Try it in / out of alliances. Try it in different rank alliances. Try entering at different times during the event. Change the activity co-efficient or whatever it is for how active the bracket it is. Populate the bracket with a variety of different scores to see if that makes any difference.

Modify the placement and scoring history. Modify the performance history (if it exists). Test again with all the above variables on multiple events in sequence, playing through events vs. not playing through them.

Try it with every combination of all of the above variables, and probably others I'm not thinking of or that we might not even know about. This isn't just a simple case of collecting stats data or leveling costs.

The point is that if you want scientific, you can't just enter an event, level all the heroes and say "look no effect" just because it works that way under one very specific circumstance. You want to be able to claim that leveling has no impact, you have to actually rule it out. Doing a perfunctory test under a single condition doesn't do that. You want to rule it out, it'll take the amount of time that a proper scientific analysis takes, with the testing tools available to do it, and the time.

It'll take you probably something like a week to several weeks full-time, thoroughly testing your hypothesis from a variety of different angles, and meticulously documenting your results. THIS is how science actually works. Not just "I did a test one way, so now I've unconditionally proven it to be the case in all circumstances, for all time".

We also know that scaling varies by event, and by node, and sometimes even the same event repeated at different times, or within the event, they've changed the scaling.

I'm a thorough person, and I'm not impressed by half-hearted "science". My observations are anecdotal, based on what I've observed on my main account and two alts. It's far from scientific, but I consider it as scientific as Madrox's testing (which is not very at all). Based on what we know, we know that scaling combines a number of variables and isn't necessarily linear or with a direct a->b relationship, given its multi-variable relationship. It doesn't actually matter if you show that under one circumstance or subset of all circumstances that it doesn't have an impact. You want to be able to make that claim, you have to actually prove in an academically rigorous manner, that scaling has no impact under any circumstances. When you have multi-variable equations, elements that have slight impacts in isolation or under certain conditions can have significant impact in conjunction with other variables or under different conditions. You want to prove that a variable doesn't have any impact, you have to actually prove it.

Arondite

They took away downvoting at just the wrong time, because if I've ever seen someone deserving of red thumbs it's you, mate.

Everyone knows that you test one variable at a time and the rest of the experiment must be comprised of controls. Jamie's entry at least met that criteria. Your two accounts are wildly different from one another in many ways. Your neat little story is not proof of anything, yet you came here lauding it as proof. When I pointed out that it's not, you now state that my rebuttal is a claim in and of itself and that burden of proof rests with me, which is absolutely absurd.

"My cat can fly."

"I really doubt that. Cat's usually don't fly. But if you want to prove it to me...go ahead."

"If my cat can't fly, prove my Cat can't fly."

Unknown

Yet again, you don't understand the scientific method.

For a multi-variable problem, you test each variable in isolation and continue proceeding, introducing new variables until you've definitively proven that a variable doesn't have any under any conditions. There are many players with data in a wide variety of states. If they want advice, you need scientific data that's actually tested the whole range of rosters possibilities under different conditions, instead of just doing a basic test and saying "good enough".

You also ignored the most salient point, that it's generally understood scaling has a greater impact on subsequent events than on whatever your current one is (or at least used to be the case - it's possibly they changed that part during my absence).

Treating a multi-variable problem as a single-variable one will not get you accurate results. Period.

Arondite

If your cat can fly, you prove it. I'm not doing your legwork for your hypothesis.

I'm fully aware it's a multi-variable issue, but the one that's in the spotlight is the effect of roster levels vs initial node levels / initial scaling. It's commonly known (and developer-stated) that individual scaling is determined by node-performance, so that's a non-issue.

The fact of the matter is, you are the one who came with a claim. I am the one who asked you to prove it. Your response was to tell me to prove my claim. The problem with this is that I haven't made a claim - I've stated that yours was unsupported. So either prove your claim or **** off.

Unknown

OP was asking about scaling, not initial scaling, and even your assumption that initial scaling is the same under all circumstances is just that - an assumption that hasn't been concretely proven, and that my own experience suggests is inaccurate.

Arondite

theHappyDance wrote:

OP was asking about scaling, not initial scaling, and even your assumption that initial scaling is the same under all circumstances is just that - an assumption that hasn't been concretely proven, and that my own experience suggests is inaccurate.

Ah, well. If he was asking about scaling over the course of the event, that's actually already been developer-stated and time-proven to not be a function of roster levels.

Unless of course, your assertion is that IceIX is a bold-faced liar, in the company of all those other liars who made the same claim.

Unknown

Arondite wrote:

theHappyDance wrote:

OP was asking about scaling, not initial scaling, and even your assumption that initial scaling is the same under all circumstances is just that - an assumption that hasn't been concretely proven, and that my own experience suggests is inaccurate.

Ah, well. If he was asking about scaling over the course of the event, that's actually already been developer-stated and time-proven to not be a function of roster levels.

Unless of course, your assertion is that IceIX is a bold-faced liar, in the company of all those other liars who made the same claim.

I interpreted OP's question as "how will leveling characters affect my scaling, both initially, over the course of the event, and on subsequent events" because that's usually what people mean when they're asking about scaling (which typically doesn't even factor in the fact that other variables like time played, placement history, etc.. likely interact with roster levels in some way).

Assuming you have a link to his post (which I'd be quite interested to see for my own better understanding), that still only addresses a subset of that overarching question, and likewise assumes that they fully understand all their own systems, which isn't entirely clear, based on past experience. They don't have to be liars, they just need to have sufficiently complex systems, and not thought about how every variable interacts (and that most definitely has been shown to be the case in the past).

mr_X

Vhailorx wrote:

I see difficult but reasonable scaling. I have finished mostly top 50 since February, with the occasional top 10 or top 150. Lone hulk in this recent hulk event started at 395, but that's the only 395s I typically see. In thick as thieves, the cMags/hood node started at 270. Very hard goon-only nodes start around 270 and end up around 320. Deadly nodes with 3 villain enemies start in the 220s.

Same here. Really cautious with levelling as I don't want to break the game. On my way to becoming an iso millionaire. Which is a crazy situation and demonstrates that the scaling model either needs fixed or at the very least more clarity. Surely can't be good for d3 that players don't want to spend their iso.

I was wandering about that level 395 hulk. Only two of my roster are just over a hundred so was a bit puzzled why he was scaled so high. He can't of been universally scaled at that level. One small consolation is he did not scale higher when defeated.

Hopefully we never have to face a level 395 Juggernaut!!

TxMoose

Lee T wrote:

Vhailorx wrote:

I just crossed path with a 395 Hulk, my xMags/Hood were at 230 and I can pretty much copy paste your last sentence subtracting about 30/40 levels each time.

FYI - everyone I know had hulk at 395 for that one and most of my group is at 104. I do have a maxed Lthor and a loki at 142 and I added a 104 steve to add stun and actually a stronger red even at 104, especially if the shield comes back.

edit: OP - just to add my story. about 2 seasons ago I added covers to thor and loki and pushed them higher. all of my other ok covered ***s were ~100 but since I was starting my transition, I really wanted to add the extra *** cover every 3 days to supplement ddq and pve finishes. I started with Lthor at ~140 and loki ~120. it worked like a charm and within a season my thor was maxed and loki at 140 and I've only failed finishing t100 twice since then. once I tried the strategy of joining late and making a late push - finished 104. didn't have the team for that - its slow (but powerful) and requires health packs every few fights. the other one I just had too much other stuff going on. didn't push late and didn't care about the character (can be said for most of this season, other than blade and dp).

so up until ultron, I strictly kept Lthor on the bench in pve, worried about possible scaling. when the hardest of nodes scaled to 150+ (all I was using was 94s and a few 104s), I would skip the node and just be happy with t150 and sometimes t50. Ultron came along and all bets were off and I used thor extensively (nearly every node, but not quite). Since ultron I figured it made no since having to skip nodes and leave thor on the bench. if nodes are going to scale out of my range anyway, screw it, i'll use him, but just on the last node or 2 - the ones that scale really hard and sometimes on an essential at sub end during the grind, since those don't scale hard anyway. I have found that the top nodes have gone up a little but not to anything that I haven't muddled my way through - nodes with 3 characters top out at about 180, characters with goon feeders top out at about 200 and I've seen goon only nodes at 320 (totally different reason/strategy - will get to that in a bit). I have wiped out from time to time but haven't wiped so hard that I couldn't take another shot at it and make it (some boards are just unworkable). full clears plus now put me just outside t10 - still trying to get over that hump where before, t50 was all I could hope for. we'll see what happens going forward but I don't see any reason to not use my guy at this point. I don't know if ultron was factored into personal scaling history for pve but if I screwed up my scaling for using him, might as well use him moving forward. plus, the first pve was for another hulkbuster, so to me that justified it too.

now - goon only nodes. I've never seen goon nodes that high until I got my 5th red on steve. however, once that happened, when grouped with Cyclops and mnmags, unless you get a terrible board, goon nodes become trivial, even in the 300s. it just takes a while to get through because my steve and cyc are 104 and mnmags is 94. I've never wiped with that team, although I have retreated and restarted. and 80% of the time they take no damage, so I think that is why those nodes scale so hard because not only am I coming out unscathed, but a large chunk of the community is as well.

so - all depends on what you are wanting out of the game. if you are all about pve and want to avoid scaling at all costs and forego any chance of pvp placement, keep them at 94-100. if you want the extra covers for pvp t100, you will need at least 1 maxed (or close to maxed) top-tier guy, and probably one support guy pushed a little higher too - those tend to be fragile. then with the boosted characters, you could probably do pretty well. my experience so far this season is 600 is required to nearly guarantee a t100 finish. it used to be lower or maybe my brackets have been harder but 400 wouldn't sniff t100 for me lately.

currently my plan is to bring everyone up to 104 and only push higher top-tier guys when they're ready. we'll see what happens then. those I consider top-tier that are within a few covers of being ready include fist, cage, steve, grocket, doom, hood, Cyclops, and magneto.

ps. if you keep them all at 94 and do really well in pve (regular t50), consider a t100 pve alliance. pretty sure I should have done several months ago and I would have been much better off if I did. I've stuck it out in my current alliance and we do both ok and finish ~200 on everything. I wish we could clone the top half of the alliance and replace the bottom half, then we'd be pushing t100 in both. anyway, hope all this helps.

Unknown

mr_X wrote:

Vhailorx wrote:

I see difficult but reasonable scaling. I have finished mostly top 50 since February, with the occasional top 10 or top 150. Lone hulk in this recent hulk event started at 395, but that's the only 395s I typically see. In thick as thieves, the cMags/hood node started at 270. Very hard goon-only nodes start around 270 and end up around 320. Deadly nodes with 3 villain enemies start in the 220s.

Same here. Really cautious with levelling as I don't want to break the game. On my way to becoming an iso millionaire. Which is a crazy situation and demonstrates that the scaling model either needs fixed or at the very least more clarity. Surely can't be good for d3 that players don't want to spend their iso.

I was wandering about that level 395 hulk. Only two of my roster are just over a hundred so was a bit puzzled why he was scaled so high. He can't of been universally scaled at that level. One small consolation is he did not scale higher when defeated.

Hopefully we never have to face a level 395 Juggernaut!!

Because it is only one opponent, a 395 Hulk is an average level of 132, which is the logic behind virtually everyone seeing a 395. If there was not a cap on levels, many of us would likely have seen Hulk go up past level 1000

anamosity

I'll throw my 2 cents in on the leveling and difficulty.

I at one time had a 220 3/5/5 xforce, 166 hulk and thor, 120 Deadpool and the rest at 110 or below. I brought up my xforce to help out in pvp when that was what you needed to win. I never used him in PVE nor did I ever use my highest level guys. My PVE got distinctly harder after leveling him from 70 to 220 however and then with the PVP MMR change that got ugly too. When the sale was on I unloaded my Xforce hoping it would either fix the problem or I'd just leave the game as it wasn't playable at that point anyways.
The very next PVE was noticeably easier on initial levels of opponents, the essentials going from mid 100's to the 90's. The game also more correctly labeled the fights as easy/normal/hard whereas before a brutal fight was labeled as normal.
PVP is also better (though that could be due to background tweaking) I now face 2* opponents up to about 400 and then start facing higher 3* with maxed 3* I can't skip away from around 500 points.

To the OP I'd suggest keeping your leveling in check, but bring up one or two key characters that you have well covered so that you can take down the big enchilada easily. Then don't use them in PVE. From what I can tell they will bring up your initial difficulty a little but by not using them to roll over easier nodes it will keep your personal scaling in check.

optimus2861

My nickel's worth, having never yet had a max 3* toon.

I have half a dozen 3* toons now in the low 100 range, with the highest pair being 114 ). I think I'm going to use that number as my soft cap for a while, until I have enough 3* toons able to move significantly higher than that. That way whatever PvE scaling I do face shouldn't grossly outpace my 2* toons, who still tend to be my workhorses.

It'll probably be many weeks if not months before I get enough 3* toons up to par to go beyond my soft cap anyway. My playtime has taken a real dive since the birth of my son, so T150 in PvE has been beyond me lately. That leaves DDQ & tokens as my only source of 3* covers, and that's pretty slow.

Oh well. I'm in no rush.

Malcrof

optimus2861 wrote:

My nickel's worth, having never yet had a max 3* toon.

I have half a dozen 3* toons now in the low 100 range, with the highest pair being 114 ). I think I'm going to use that number as my soft cap for a while, until I have enough 3* toons able to move significantly higher than that. That way whatever PvE scaling I do face shouldn't grossly outpace my 2* toons, who still tend to be my workhorses.

It'll probably be many weeks if not months before I get enough 3* toons up to par to go beyond my soft cap anyway. My playtime has taken a real dive since the birth of my son, so T150 in PvE has been beyond me lately. That leaves DDQ & tokens as my only source of 3* covers, and that's pretty slow.

Oh well. I'm in no rush.

Not sure if it means anything, i am using 120 as my soft cap for my 3*'s (other than my duo of 153's) as that is where the iso leveling jumps from 600 to over 1k a level. I have a feeling this is significant somehow.

Unknown

Malcrof wrote:

Not sure if it means anything, i am using 120 as my soft cap for my 3*'s (other than my duo of 153's) as that is where the iso leveling jumps from 600 to over 1k a level. I have a feeling this is significant somehow.

I was collecting my guys at 110 until I had a bunch ready to roll. After anniversary week, I had a ton of iso and moved them all to 120. The impact to my PvE scaling was noticeable, as my 94s really started to struggle on the higher end nodes where they use to be full event workhorses. Since I had bumped up 10 or 12 guys I had options for the high end, but it was very noticeable that the 94s were no longer on equal footing.

Unknown

anamosity wrote:

To the OP I'd suggest keeping your leveling in check, but bring up one or two key characters that you have well covered so that you can take down the big enchilada easily. Then don't use them in PVE. From what I can tell they will bring up your initial difficulty a little but by not using them to roll over easier nodes it will keep your personal scaling in check.

My 94 MsMarvel/Stormneto handles Big Enchilada just fine. So no need to overlevel for this reason alone.

TxMoose

shurak wrote:

anamosity wrote:

To the OP I'd suggest keeping your leveling in check, but bring up one or two key characters that you have well covered so that you can take down the big enchilada easily. Then don't use them in PVE. From what I can tell they will bring up your initial difficulty a little but by not using them to roll over easier nodes it will keep your personal scaling in check.

My 94 MsMarvel/Stormneto handles Big Enchilada just fine. So no need to overlevel for this reason alone.

yes, BE is not a reason to level anyone since there are multiple teams/approaches with 94s to do the job there. you cannot finish t100 on any kind of regular basis in pvp with 94s (note: this was my experience when I was playing only 94s, no idea how the recent changes have affected that). leveling a pair, to me, would be to finish t100 in pvp - that's why I did my pair.

Roswulf

TxMoose wrote:

shurak wrote:

anamosity wrote:

To the OP I'd suggest keeping your leveling in check, but bring up one or two key characters that you have well covered so that you can take down the big enchilada easily. Then don't use them in PVE. From what I can tell they will bring up your initial difficulty a little but by not using them to roll over easier nodes it will keep your personal scaling in check.

My 94 MsMarvel/Stormneto handles Big Enchilada just fine. So no need to overlevel for this reason alone.

yes, BE is not a reason to level anyone since there are multiple teams/approaches with 94s to do the job there. you cannot finish t100 on any kind of regular basis in pvp with 94s (note: this was my experience when I was playing only 94s, no idea how the recent changes have affected that). leveling a pair, to me, would be to finish t100 in pvp - that's why I did my pair.

Speaking as someone mid transition (3 cover-maxed 3*, with a bunch of others useful to semi-useful) who has soft-capped everyone at lvl 96, that seems to have changed. I invariably have enough buffed characters to put together one or two viable teams, and when I climb a couple hundred points early and put in a reasonable amount of effort in the hour before crunch time, I can usually hit top 100.

Top 50 is not however possible.

puppychow

dathremar wrote:

The questions “What affects scaling” and “What is the meaning of life” are interchangeable it seems. Even the most hardened veterans don’t agree on what the contributing factors are. Personally, as a player who’s soft-capped his max covered 3* til the Iso cost reduction patch, levelling my roster past 94 makes pve harder. I could see it especially in Goon damage apparently. Maybe that’s because I pushed 25 characters to level 120 at once. I think it’s a mental effect as well because we’ve all been indoctrinated to believe that levelling roster is equivalent to harsher scaling.

To be fair, I also leveled up a lot of my characters to 120, and for recent pve events I haven't had more or less difficulty than usual. I think it's a matter of the weekly buffs that affect your scaling more, to be honest.