Figuring out scaling

rossmon

So, I know this is probably a lost cause, but...I'd like to make a project for getting a general distribution for scaling. This will probably be a long term project, but here goes....
Anyone who is willing, please take screenshots of nodes in PVE before you play them. Then, please post for each node, before first clear (to prevent personal scaling, and at end of grind (just because):
Node name/number, Opponents level, difficulty.
and your top 6 boosted, and unboosted level characters.

https://docs.google.com/spreadsheets/d/ ... sp=sharing

I will post findings eventually. In the mean time, I'm hoping to collect as many players data as I can, in order to get best data.

Thanks to All!

In order to preserve some semblance of a scientific method, please only put in levels before you start the clear. I'm going to start taking screenshots as I do my first clear, and then go back later with the data, so it doesn't mess with my timing. To figure this out accurately I'm going to need to compare the same node, so all data is appreciated, but It'll probably be The next event before I have enough people giving data for this to work. Hopefully I can in some way decode part of the madness.

Apparently I am sick and twisted, and enjoy crazy math.....


EDIT: So, I've basically abandoned this project. A little bit of work has disproved my possibility and proven the secret 'scaling' variable exists.
End result:
Trivial nodes are the same for all.
Scaling changes based on too many variables for anyone to be able to figure out the secret
Leveling your characters hurts only if you don't do it evenly.
Character/Roster level is a factor, but not the only factor.

rossmon

Example, and mine for hulk 1 Beginning.:
Nevada 1: 34, Trivial
Nevada 2: 43, Trivial
Nevada 3: 55, Trivial
Nevada 4: 108 Easy
Nevada 5 124 Easy
Nevada 6 395 Normal
Storm: 89, Easy
Blade: 89, Easy
Thor: 86, Easy
Roster: 94 unboosted (+patch @104)
Boosted: Falcap:165; MrF: 161, Thoress: 160, Ares/HE/Storm: 150, Blade: 134, Doom@129

alphabeta

Feels like a google docs spreadsheet where we could direct input would be useful (plus then the data is in an easier format to analyse

Happy to share mine but didn't record it for first two subs of hulk - will do for the 3rd sub when it starts

rossmon

alphabeta wrote:

Feels like a google docs spreadsheet where we could direct input would be useful (plus then the data is in an easier format to analyse

Happy to share mine but didn't record it for first two subs of hulk - will do for the 3rd sub when it starts

Thanks. That's a great Idea. I'll be putting it into an excel sheet anyway, so, why not...

rossmon

https://docs.google.com/spreadsheets/d/ ... sp=sharing

fmftint

There's more to it than what you're asking for, you also need
Team used and levels
Health going in/out of battle

lukewin

There was a thread not too long ago that had some very in-depth analysis of scaling. Can't remember who it was by, but I was impressed and their findings were similar to what my opinions on scaling are.'

EDIT: found it. Feedback & Analysis of PvE Personal Scaling Over 471 Battles by Vynyv

rossmon

fmftint wrote:

There's more to it than what you're asking for, you also need
Team used and levels
Health going in/out of battle

I'm sure that I'll move on towards that eventually, Right now I'm more just trying to figure out the starting point.

mindsuckr

I put my reply in as Death. What's interesting is we seem to have almost the same levels when I have higher unbuffed characters than you. I seem like scaling is based on your buffed levels based on our extremely limited, two person data.

rossmon

This is interesting so far. I'm going to wait until a decent amount of data is collected before I do much calculations. So far, I find it interesting that my 94 roster, and a 166 roster both have the same trivial node levels. Also I've heard some talk of tiers, which could also make sense...94-108 get the same levels for example. It would be nice to figure out where this starts. If this is the case, I could level my 94s to 108 without any adverse effects, and my 2* would not be hurt by higher levels.

mindsuckr

Just updated Greenland for you. We have the same level nodes, but our essentials have a variance. I am taking this event easier though and only hitting ~10k each day to get the legendary token since none of the rewards will help me.

XandorXerxes

rossmon wrote:

So, I know this is probably a lost cause, but...I'd like to make a project for getting a general distribution for scaling. This will probably be a long term project, but here goes....
Anyone who is willing, please take screenshots of nodes in PVE before you play them. Then, please post for each node, before first clear (to prevent personal scaling, and at end of grind (just because):
Node name/number, Opponents level, difficulty.
and your top 6 boosted, and unboosted level characters.

https://docs.google.com/spreadsheets/d/ ... sp=sharing

I will post findings eventually. In the mean time, I'm hoping to collect as many players data as I can, in order to get best data.

Thanks to All!

In order to preserve some semblance of a scientific method, please only put in levels before you start the clear. I'm going to start taking screenshots as I do my first clear, and then go back later with the data, so it doesn't mess with my timing. To figure this out accurately I'm going to need to compare the same node, so all data is appreciated, but It'll probably be The next event before I have enough people giving data for this to work. Hopefully I can in some way decode part of the madness.

Apparently I am sick and twisted, and enjoy crazy math.....

I've actually got a decent log that I kept track of when I jumped my characters to 166. I wanted to know what effect level (alone) had on node level. Short answer - level didn't matter.

Biggest finding that I found was that the only real jump I saw in levels was when I used Luke Cage on a clear. When I stopped using him, levels came down a little bit and I've been seeing a small, gradual rise since which I imagine is based on my clear times, etc.

I'll dig up the data when I get back home.

rossmon

Thanks to all who are inputting so far. What I see so far:
1: first 3 nodes seem to be the same to all.
2: I've screwed my scaling.

I'm going to let this continue to build, so I have more data and look at it all later.

Thanks again everyone