Old 08-24-2019, 10:01 PM   #1
One Winged Angel
Anime Avatars ( ◜◡^)っ✂╰⋃╯
Retired StaffFFR Simfile AuthorD7 Elite KeysmasherFFR Veteran
 
One Winged Angel's Avatar
 
Join Date: Mar 2007
Location: Squat Rack
Age: 29
Posts: 10,718
Default Scoring Data in Brutal Difficulty Range

I've mentioned a few times post-tourney I intended to create a thread that would act as a hub to discuss difficulty changes to be made to 100+ charts. I gathered some data manually and decided there's no harm in creating that thread now. We may not hammer out the exact numbers until later, but in an effort to find the best starting point for D8 on the leaderboards, I figured I could share this data to try and start grouping things appropriately.

Here's a spreadsheet of all 98+ difficulty charts in the game that have been released prior to this tournament, along with AAA counts, what rank a SDG will net you, what rank sub-50 will net you, and the release year of the chart. At first I only included 99+ but I know there's gonna be that one guy that brings up oppIrish so I dipped down to 98. I believe it's fair to say nothing rated 97 or lower needs to be considered being moved to 100+. Yes, I'm aware some charts are tokens and the scoring data is impacted due to these charts not being immediately accessible.

https://docs.google.com/spreadsheets...it?usp=sharing

I have my own ideas regarding what should move but I don't want to guide the discussion. I've seen a lot of people discussing this range on Discord and I'm sure they'll comment here.

What I will say is that the skill rating algorithm needs to be adjusted. There are currently 19 players that have a AAA on at least one level 100+ chart (quite a few less if Rave 7 is removed from consideration), but a good handful more than that already in or approaching the 100+ range on the leaderboards. The skill rating system approximates the highest level you're capable of AAA'ing, and when a player's personal rating is higher than any AAA they've ever achieved, something's up. Current scaling awards a higher equivalency for 2-0-0-1 on a file rated x+1 in difficulty compared to a AAA on a file rated x, and I feel many players would agree the AAA in this instance is likely more impressive, assuming the charts are both rated appropriately.

In my opinion, charts rated up to 104/105 should be adjusted to scale in such a way that a BF on x+1 is worth approximately the same as a AAA on x. It'll make reaching level 100+ reflect the fact that the player can actually AAA charts in that range as intended by the system, not just score relatively well. Beyond that difficulty threshold an entirely new scaling formula likely needs to be developed for the charts that, less two or three players at the moment, no one is going to be approaching a AAA. But that's a separate discussion (which could honestly be discussed here as well, I don't mind)

Interested in both thoughts regarding chart difficulties in this range and ensuring appropriate equivalency scaling.
__________________


Quote:
Originally Posted by ilikexd View Post
i want to be cucked by cirno

Last edited by One Winged Angel; 08-24-2019 at 10:10 PM..
One Winged Angel is offline   Reply With Quote
Old 08-24-2019, 11:40 PM   #2
s1rnight
( ŻuŻ)-b
FFR Music Producer
 
s1rnight's Avatar
 
Join Date: Aug 2011
Posts: 290
Send a message via AIM to s1rnight Send a message via Skype™ to s1rnight
Default Re: Scoring Data in Brutal Difficulty Range

hehe, i actually wonder if it's possible to auto-rate these charts... if you re-base the user's levels on a "normal distribution" that takes scores on each file within a group of users (the hierarchy of which should stay constant across all files, AAAs notwithstanding), you can then use that to weight how difficult each of the charts are... you'd then have a whole system that auto-updated on scorings alone... "difficulties" on charts then would be more like bounties, bc they'd go down if more people DEFEATED one

i remember how fun the "goat score calculator" was to make... i may try to make something for this too lol

edit: i know how im gonna do this, brb
__________________


 a  'skill  boost'  .  .  .  !  one  in  the  flesh

Last edited by s1rnight; 08-25-2019 at 12:19 AM..
s1rnight is online now   Reply With Quote
Old 08-25-2019, 12:23 AM   #3
Herogashix
FFR R^3 Hokage
FFR Music ProducerD7 Elite KeysmasherFFR Veteran
 
Herogashix's Avatar
 
Join Date: Apr 2009
Age: 22
Posts: 2,158
Default Re: Scoring Data in Brutal Difficulty Range

not going to lie, i didn't realize till looking at the spreadsheet that metro hasn't gotten a AAA yet when it feels totally doable for me (which means it should be for other d8 players). that's an interesting piece of info

also can we agree Almost There is more like a 100 or something
__________________
Herogashix is offline   Reply With Quote
Old 08-25-2019, 12:32 AM   #4
RenegadeLucien
FFR Veteran
Skill Rating Designer
DeveloperFFR Veteran
 
RenegadeLucien's Avatar
 
Join Date: Jan 2016
Age: 23
Posts: 255
Default Re: Scoring Data in Brutal Difficulty Range

Quote:
Originally Posted by s1rnight View Post
hehe, i actually wonder if it's possible to auto-rate these charts... if you re-base the user's levels on a "normal distribution" that takes scores on each file within a group of users (the hierarchy of which should stay constant across all files, AAAs notwithstanding), you can then use that to weight how difficult each of the charts are... you'd then have a whole system that auto-updated on scorings alone... "difficulties" on charts then would be more like bounties, bc they'd go down if more people DEFEATED one
I feel like this would add way too much variability to a player's skill rating, in that you could suddenly drop rating just because someone else did well on a file. Not to mention, it would be entirely conceivable under this system that your skill rating would go down by improving your score on a file. And on top of this, there aren't enough people at the top end to accurately imitate a normal distribution for 100+ diff songs.
__________________


Zenith Ultimate Struggle: 3rd (D2)

All public 1-36s AAA'd

Last edited by RenegadeLucien; 08-25-2019 at 12:33 AM..
RenegadeLucien is offline   Reply With Quote
Old 08-25-2019, 01:20 AM   #5
rayword45
Local Teenage Wastebasket
FFR Simfile AuthorD7 Elite KeysmasherFFR Veteran
 
rayword45's Avatar
 
Join Date: Feb 2007
Location: My bed
Age: 22
Posts: 2,419
Default Re: Scoring Data in Brutal Difficulty Range

Quote:
Originally Posted by Herogashix View Post
also can we agree Almost There is more like a 100 or something
you say this in spite of it having the best stats by far of all 99s

and tbh i still agree
__________________
The above post has a 50% chance of being useless. Potentially. Maybe.

BEST AAAs: WANDERLUST, Pandora, Necropotence, Mourning The Lost, Eradication, Feldschlacht

Hey, we need some users on this site. Please join.

And if you have not recommended any albums yet, do so. Please. I have a goal to reach. Here.
NO WAIT THAT SHIT'S OLD GO HERE INSTEAD.
rayword45 is online now   Reply With Quote
Old 08-25-2019, 01:37 AM   #6
Precarious
Unacceptable
FFR Veteran
 
Precarious's Avatar
 
Join Date: Mar 2014
Posts: 190
Default Re: Scoring Data in Brutal Difficulty Range

I saw you bring this up in the Odd-22 thread. Obviously I'm going to be a bit limited in what I can suggest, simply because I'm orders of magnitude below the relevant ability level here, but I *can* say this isn't a Brutal+ difficulty-exclusive issue. AAAs on difficulty x tend to be more impressive than flags on x+1 almost across the spectrum, and that's before accounting for frequent errors in difficulty assignment.

But I feel like you're getting at a larger point that doesn't have an answer. One reason people seem to talk about individual scores more than average level is that any algorithm is going to be somewhat inconsistent, because songs don't always scale the same way. I've heard that Husigi, for example, is relatively easy for a very high level player to SDG, but very hard to AAA. The spreadsheet stats seem to bear that out. A formula that accurately rewards performance on that song would probably need to penalize raw goods more harshly than on another song that is equally difficult to AAA, but harder to score on generally. Since adjusting every song for its quirks is impossible (or at least impossibly impractical), we can only do our best.

In any case, though, I agree with your basic premise. A lot of people seem to disdain playing for AAAs, since Etterna, osu!, etc. don't, but that's what FFR is built around, and I don't see a problem with maintaining that focus. I think accelerating AAA-decay is probably a reasonable approach in general. I'd also like to see difficulties reviewed, and not just in the 98+ realm, although I realize that's beyond the scope of this discussion.

Last edited by Precarious; 08-25-2019 at 01:39 AM..
Precarious is offline   Reply With Quote
Old 08-25-2019, 06:37 AM   #7
s1rnight
( ŻuŻ)-b
FFR Music Producer
 
s1rnight's Avatar
 
Join Date: Aug 2011
Posts: 290
Send a message via AIM to s1rnight Send a message via Skype™ to s1rnight
Default Re: Scoring Data in Brutal Difficulty Range

hello everyone, i am back from making a cool script. this one takes a normal distribution of the top 100 "grand total" users (meaning they prolly played a lot of files, and are at a... variable? maybe, skill level. prolly no noobs tho), and then takes the standard deviation of a hypothetical "AAA" score. theres prolly lots of important omissions, i am wondering if adding EVEN MORE USERS to the sample would help



(important thing to note is that MOST of the d8 and d7 users are not anywhere near the top 100 grand total, lol. so i am going to make a script that have a lot of those guys at least...)
__________________


 a  'skill  boost'  .  .  .  !  one  in  the  flesh

Last edited by s1rnight; 08-25-2019 at 06:48 AM..
s1rnight is online now   Reply With Quote
Old 08-25-2019, 06:39 AM   #8
s1rnight
( ŻuŻ)-b
FFR Music Producer
 
s1rnight's Avatar
 
Join Date: Aug 2011
Posts: 290
Send a message via AIM to s1rnight Send a message via Skype™ to s1rnight
Default Re: Scoring Data in Brutal Difficulty Range

this script would most likely be EVEN BETTER if i could take all the users on a given leaderboard, since right now it just constructs a virtual leaderboard with (any of the top 100 grandtotal users that happened to play the chart to teh end), which isnt exactly exhaustive. i may experiment with other things...
__________________


 a  'skill  boost'  .  .  .  !  one  in  the  flesh
s1rnight is online now   Reply With Quote
Old 08-25-2019, 09:43 AM   #9
rushyrulz
Digital Dancing!
Multiplayer ModeratorEvent StaffFFR Simfile AuthorFFR Music ProducerD7 Elite KeysmasherFFR Veteran
 
rushyrulz's Avatar
 
Join Date: Feb 2006
Location: 72 billion club, NE
Age: 27
Posts: 12,349
Default Re: Scoring Data in Brutal Difficulty Range

oppIrish could easily be 100, Stress Free Style and Call me It as well. Agree with Almost There because anchory as fuck 300 BPM stream is just as bad as regular 333 BPM stream.

Powerflux could use a bump because mid D7 and below can't even approach that file without tiring out midway through.

bump revo up to 110 honestly

nerf DISCHARGE RUSH maybe

I kinda wanna say nerf do i smile?
__________________

rushyrulz is offline   Reply With Quote
Old 08-25-2019, 10:33 AM   #10
Matthia
D8 Unicorn
FFR Veteran
 
Matthia's Avatar
 
Join Date: Nov 2017
Location: Pacific Timezone, USA Age: 17.7
Posts: 146
Default Re: Scoring Data in Brutal Difficulty Range

I'm also on the + side for bumping oppIrish 1 or 2 levels

buff Powerflux
buff violent arcade
buff both level 100 OWA Raged charts
buff TBR
buff /a/ (or \a\ is what most people call it and idk why)

nerf WWE
nerf Almost There
maybe nerf whites walls 2 but its 7 minute length kind of brings the difficulty up again

Like said countless times already the 98+ charts will be reworked post tourney
Matthia is online now   Reply With Quote
Old 08-25-2019, 11:44 AM   #11
Shxfted
D8 Unicorn
FFR Simfile AuthorD7 Elite KeysmasherFFR Veteran
 
Shxfted's Avatar
 
Join Date: Nov 2014
Age: 24
Posts: 200
Default Re: Scoring Data in Brutal Difficulty Range

Both OWA 100s: +1
Make The Fire Burn: -1? (though this is one that is just hard to AAA but not super hard to get near AAA)
Powerflux: +1
Violent Arcade: +1
/a/: +1
WWE: -1
WW2: -1
__________________
Lemon's Fruit Snacks 2
Shxfted is offline   Reply With Quote
Old 08-25-2019, 12:15 PM   #12
One Winged Angel
Anime Avatars ( ◜◡^)っ✂╰⋃╯
Retired StaffFFR Simfile AuthorD7 Elite KeysmasherFFR Veteran
 
One Winged Angel's Avatar
 
Join Date: Mar 2007
Location: Squat Rack
Age: 29
Posts: 10,718
Default Re: Scoring Data in Brutal Difficulty Range

I agree with both those last two posts less nerfing MtFB...I'd say it has the edge in difficulty over any 102 and if Powerflux and VA go up that would leave Odd-22 as the sole 103 with multiple charts in the sub-tiers above and below, which is kind of ugly. The purpose of posting the spreadsheet was hoping high tier players can come to a consensus (or as close to one as possible) regarding the order of difficulty for those charts, and then difficulty boundaries can be drawn from there, so as to retain an acceptable number of files per subgroup. I probably should have noted that in the OP. Otherwise yeah sure there's an argument to move up most 99s but then what's left in 99? etc.

@s1rnight I was toying with the idea of automating difficulties based on scoring data alone (manipulating the spreadsheet data points as inputs) but I'd need a way to procure this information across all charts and I have no intention to do that manually (and wouldn't be able to publically access sdg/sub-50 cutoffs for easier charts anyways). A few issues arise when comparing charts from years back where the unique player count is substantially higher than charts released nowadays but I think a modifier can be applied to make the order truer to what it should rightfully be.

There will always be charts that spit out values much higher than they should because no one wants to give them more than one or two serious runs due to sheer length, or they're extremely backloaded in difficulty. Something like Starbound would easily produce a value over 100 when it definitely shouldn't (though it's certainly a bit higher than where it currently is).
__________________


Quote:
Originally Posted by ilikexd View Post
i want to be cucked by cirno

Last edited by One Winged Angel; 08-25-2019 at 12:21 PM..
One Winged Angel is offline   Reply With Quote
Old 08-25-2019, 01:04 PM   #13
TC_Halogen
Rhythm game specialist.
Retired StaffFFR Simfile AuthorFFR Music ProducerD7 Elite KeysmasherFFR Veteran
 
TC_Halogen's Avatar
 
Join Date: Feb 2008
Location: Bel Air, Maryland
Age: 28
Posts: 18,577
Send a message via AIM to TC_Halogen Send a message via Skype™ to TC_Halogen
Default Re: Scoring Data in Brutal Difficulty Range

The solution to this problem is going to be whether or not we inevitably shift away FFR's mindset of rating based off of AAA difficulty and take more physical capability into account. Other rhythm games don't arbitrarily skyrocket entirely playable charts simply because of having concepts that make their chart tougher to perfect. Songs like Crowdpleaser continue to be treated as an anomaly in any case, though. FFR obviously cannot rely on passability for the purpose of difficulty scaling considering that the game mechanics can promote endless mashing to stay alive - though that's not really impactful for the leaderboard system.

On top of it, we're contending with the game's process of rendering charts where songs that are LOWER in effective tempo can be harder than songs that are higher in tempo due to overall frame distribution (see: Almost There versus Magical 8-bit Tour).

Wanderflux and Gamma both feel substantially harder than RATO does to me, and yet they fall lower down on the difficulty scale simply because they don't theoretically have parts that are harder to AAA compared to RATO. My average score on RATO will be better than Wanderflux/Gamma nearly 100% of the time because the physical requirement to play the chart is substantially lower on account of all of the difficulty being contained within jumptrillable walls. Yet, RATO sits at 108 because of this. Naturally, not enough time has passed to make a claim about overall scoreboard spread between these two songs, but I'd be willing to wager that a good majority of players will have a similar relationship between these two songs.

On the flip side: FFR values difficulty spikes way too much due to the skill rating algorithm placing an exceptional weight on scores that are near or at the AAA-level. Two quick examples:

- Husigi should not be a 101, it should be a hard-to-AAA-(number lower than 100) because of the large wall and short [14] jack sequence. Aside from those two bits that very literally account for less than 5 seconds of the total length, the chart does not have much that puts it beyond the low-mid 90s. Full-run single digit good counts (not necessarily effective SDGs) can be found all the way down to rank 70 for a song whose difficulty level actually eclipses the entry requirement for FFR's D8 in this current official.

- Winter Wind Etude doesn't have quite as deep of a score spread for a 101, but the chart has virtually no physical boundaries compared to anything that's 90+; the chart is rated as high as it is because of the brief jack/polys and the ending (which isn't all that difficult even for 100 standards). Numerous D7-level competing OT players (even some that might not have been thought of as front-runners) have extremely good scores on this and might be a tell-tale sign that a rating drop might be required.

I won't go through the entire list of files with explanations behind everything, but here are my suggestions (keeping in mind, which come from a bit of a different angle):

- System Doctor: 101 -> 98*
- Husigi Usagi Milk Tei: 101 -> 99*
- White Walls Part 2: 102 -> 100*
- Winter Wind Etude: 101 -> 100
- Violent Arcade: 103 -> 102*
- RATO: 108 -> 106*

[...etc]

The disparity between Revo and every other 102 is absolutely obnoxious, by the way...

* reiterating that this is a change reflective physical capability, not necessarily difficulty to AAA. Changes like this absolutely would not work with our current skill rating algorithm and would likely cause some annoyances.


With respect to our current system and putting aside the concept of physicality, though - my thoughts:
- Snafu: 100 -> 99
- Powerflux: 103 -> 104
- Vegas/OWA-Skeletor: 100 -> 101
- Magical 8bit Tour: 99 -> 100 (frameslol)

...with respect to not using that to guide the discussion:
Quote:
The skill rating system approximates the highest level you're capable of AAA'ing, and when a player's personal rating is higher than any AAA they've ever achieved, something's up. Current scaling awards a higher equivalency for 2-0-0-1 on a file rated x+1 in difficulty compared to a AAA on a file rated x, and I feel many players would agree the AAA in this instance is likely more impressive, assuming the charts are both rated appropriately.
It has become increasingly clear that there needs to be some sort of logrithmic scaling of AAA equivalency where the higher the difficulty gets, the more lenience it should take for a score to be considered "as impressive" as you mentioned. This in turn points back to the difficulty scale, though: how do we use a mathematical equivalence for comparing the results of a song at x difficulty to a AAA at x-y difficulty when certain disparities clearly exist within a point of difficulty? A length factor might be needed on top of the score for AAA equivalence determinance (as in: slow the exponential decay rate as the length of the song grows).

Take where is my balls (100) and A Dichroic Glass Snafu (100) -- both released within a week of each other. The difference in score spread is pretty quickly apparent. Nearly the same number of players, and very close number of times played (within a few hundred), yet the score spread clearly favors the shorter song. That might need to be another point to look into for developing a new skill-rating algorithm.

Last edited by TC_Halogen; 08-25-2019 at 04:41 PM..
TC_Halogen is offline   Reply With Quote
Old 08-25-2019, 01:44 PM   #14
devonin
Very Grave Indeed
Community Manager, Global Moderator, User Support
Global ModeratorFFR Simfile AuthorFFR Veteran
 
devonin's Avatar
 
Join Date: Apr 2004
Location: Ontario, Canada
Age: 35
Posts: 10,036
Send a message via AIM to devonin Send a message via MSN to devonin
Default Re: Scoring Data in Brutal Difficulty Range

Quote:
Originally Posted by One Winged Angel View Post
The skill rating system approximates the highest level you're capable of AAA'ing, and when a player's personal rating is higher than any AAA they've ever achieved, something's up. Current scaling awards a higher equivalency for 2-0-0-1 on a file rated x+1 in difficulty compared to a AAA on a file rated x, and I feel many players would agree the AAA in this instance is likely more impressive, assuming the charts are both rated appropriately.
Extreme low skill player opinion incoming, disregard as you like.

A factor to consider here though alongside drawing conclusions about the formula and the way it treats AAAs, with regards to "when a player's personal rating is higher than any AAA they've ever achieved" is that since the introduction of the skill rating system, there is no incentive at all to care about ever getting a AAA unless a) You care strongly about maximizing tier points or b) You're at the extreme top end where improving to a AAA where possible is one of the only ways to actually improve your skill rating.

I'm level 33, my best AAA Equiv is a 34.98 and my best AAA is my 68th highest score, and is on a 17. Could I AAA a 33? Depends hugely on the 33 and it would take a ton of time if at all. But I can -absolutely- AAA much higher than a 17, there's just no benefit for me to put in the replay time to do it, or cultivate the accuracy over speed/reading.
devonin is offline   Reply With Quote
Old 08-25-2019, 03:10 PM   #15
TheSaxRunner05
The Doctor
FFR Veteran
 
TheSaxRunner05's Avatar
 
Join Date: Apr 2006
Age: 30
Posts: 5,917
Default Re: Scoring Data in Brutal Difficulty Range

The main trouble I see in increasing the AAA equivalency decay is it really advantages those that don't typically play for AAAs, for tournaments. You could end up with people with 30g on Tageri not even count as D5, but they clearly have the speed. I know that's difficult for any system to handle, but it may make it worse.
__________________


TheSaxRunner05 is offline   Reply With Quote
Old 08-25-2019, 05:47 PM   #16
Matthia
D8 Unicorn
FFR Veteran
 
Matthia's Avatar
 
Join Date: Nov 2017
Location: Pacific Timezone, USA Age: 17.7
Posts: 146
Default Re: Scoring Data in Brutal Difficulty Range

Quote:
Originally Posted by PrawnSkunk View Post
Looked back at the skill rating thread (the one that contains the AAA Equivalency formula: http://www.flashflashrevolution.com/...d.php?t=140927),
we can clearly see how 1-good-input AAA equivalency decay diminishes as the difficulty of a file inflates. (Disregard to the fact that the graphs are a little inaccurate today due to some occasional adjustments over the years)

I spent time iso-ing the last ten notes of 100+ files and purposely hitting blackflags to absorb more data on how that one good affected the equivalency score relative to the varying difficulties of the files
Code:
100 - First good costs 0.42 equivalency
101 - 0.42
102 - 0.41
103 - 0.41
104 - 0.40
105 - 0.39
106 - 0.39
107 - 0.38
108 - 0.38
109 - 0.37
110 - 0.37
*I have already tested to see whether or not dropping more goods produces different change of equivalency loss, and it is quite linear*

The idea of punishing less as difficulty increases is fine, not much of the concern here. Maybe (if possible) rework the AAA equivalency formula (or perhaps even better just implement a new method that will carry on after an output from the current formula was generated) in a way that robs more out of your equivalency score for the first few goods of a 100+ file, and then adjust to the normal speed of -0.xx per raw good you take. AAA'ing something should be more rewarding honestly:


You Universe

Difficulty= 104

{PA} - {Old Equivalency Score} {Change}--> {New Equivalency Score} {Change}
AAA - 104.00 N/A--> 104.00 N/A
1g - 103.60 -0.40 --> 103.40 -0.60
2g - 103.20 -0.40 --> 102.85 -0.55
3g - 102.80 -0.40 --> 102.35 -0.50
4g - 102.41 -0.39 --> 101.90 -0.45
5g - 102.01 -0.40 --> 101.50 -0.40
6g - 101.62 -0.39 --> 101.11 -0.39
7g - 101.23 -0.39 --> 100.72 -0.39
8g - 100.84 -0.39 --> 100.33 -0.39
9g - 100.46 -0.38 --> 99.95 -0.38
10g - 100.07 -0.39--> 99.56 -0.39

Notice the change of new equivalency


The example in this spoiler is there to give you an idea on how my solution might work out, the values I gave for the "new equivalencies" are somewhat arbitrary and shouldn't be taken seriously plz. This will surely take much time to put together, making sure everything is balanced well down to each single difficulty level.

Quote:
Originally Posted by TheSaxRunner05 View Post
The main trouble I see in increasing the AAA equivalency decay is it really advantages those that don't typically play for AAAs, for tournaments. You could end up with people with 30g on Tageri not even count as D5, but they clearly have the speed. I know that's difficult for any system to handle, but it may make it worse.
This could be handled just by letting the system/method only affect D8 territory.

Last edited by Matthia; 08-25-2019 at 05:56 PM..
Matthia is online now   Reply With Quote
Old 08-26-2019, 12:18 AM   #17
One Winged Angel
Anime Avatars ( ◜◡^)っ✂╰⋃╯
Retired StaffFFR Simfile AuthorD7 Elite KeysmasherFFR Veteran
 
One Winged Angel's Avatar
 
Join Date: Mar 2007
Location: Squat Rack
Age: 29
Posts: 10,718
Default Re: Scoring Data in Brutal Difficulty Range

Did a quick midnight skim, lots of good points being made and I appreciate the discussion. Posting to remind myself to respond in greater detail tomorrow because there's several sentiments I agree with while others would compromise some direct comparisons between charts under the current skill rating system.

Honestly this probably mostly boils down to adjusting the scaling for certain edge cases.
__________________


Quote:
Originally Posted by ilikexd View Post
i want to be cucked by cirno
One Winged Angel is offline   Reply With Quote
Old 08-26-2019, 02:24 AM   #18
Untimely Friction
D6 Challeneged
FFR Veteran
 
Untimely Friction's Avatar
 
Join Date: Aug 2012
Age: 27
Posts: 1,215
Default Re: Scoring Data in Brutal Difficulty Range

nerf milk tei its too easy to get 20g on

Edit: Reflecting on this and Halogens mention of it being the difficulty of the entry level for D8 or so, I wonder at if d8's can AAA this. That's prolly important since the whole point was to guage AAA equivalency, and therefor moooost d8's should competently fesibly AAA Milk tei, and if they dont, somethings weird, and considering AAA's past 100 just isnt reasonable? Or something...

Last edited by Untimely Friction; 08-26-2019 at 02:27 AM..
Untimely Friction is offline   Reply With Quote
Old 08-26-2019, 08:35 AM   #19
Dynam0
The Dominator
D7 Elite KeysmasherFFR Veteran
 
Dynam0's Avatar
 
Join Date: Sep 2005
Location: North Bay, ON
Age: 30
Posts: 8,658
Default Re: Scoring Data in Brutal Difficulty Range

I propose difficulty rating be composed of the sum of two factors:

A = General Difficulty Factor
B = Upper Tier Scoring Factor

Total AAA Equivalency = A + B

Yes, yes I know this means we need two separate difficulty values for each chart now, how annoying! But the idea is that for charts with a consistent difficulty, B would tend to 0 and for charts like Husigi it would be on the higher end.

Now for calculating the skill points, I have messed with this in excel a bit and this is where I am at now:

Total Skill points = Ax + B^y

where x = total raw score / max raw score, and;
where y = 1/(effective goods + 1)


Did some fiddling in excel with those and the effect in principle seems promising, but it needs more fine tuning.

For Husigi:

Using A = 85 and B = 16

Example 1 - Raw Score (0/0/0/0) - Total Skill Points: 101
Example 2 - Raw Score (0/0/0/1) - Total Skill Points: 95.07
Example 3 - Raw Score (1/0/0/1) - Total Skill Points: 88.50
Example 4 - Raw Score (5/1/0/2) - Total Skill Points: 86.23
Example 5 - Raw Score (45/12/4/16) - Total Skill Points: 84.15

Someone wanna play around with that idea some more? For instance I find the decrease in the "B" component is a bit too punishing in the above example while the decreases in the "A" component are not harsh enough.

Last edited by Dynam0; 08-26-2019 at 08:49 AM..
Dynam0 is offline   Reply With Quote
Old 08-26-2019, 01:08 PM   #20
One Winged Angel
Anime Avatars ( ◜◡^)っ✂╰⋃╯
Retired StaffFFR Simfile AuthorD7 Elite KeysmasherFFR Veteran
 
One Winged Angel's Avatar
 
Join Date: Mar 2007
Location: Squat Rack
Age: 29
Posts: 10,718
Default Re: Scoring Data in Brutal Difficulty Range

Ok, a few thoughts:

FFR historically has always been a game favoring accuracy. There aren't other letter grades players are aiming for like AA which requires meeting a specific percentage threshold or other mechanics similar to that, and as such players will gravitate towards playing charts they can score relatively well on. Even with incredibly difficult charts being added now, outside of vrofl and the very newly released HQR, every chart has at least been SDG'd (and HQR will get there by round's end). By tourney's end I'm projecting around 5 charts at maximum that won't have SDGs as top scores, so the difficulty ceiling will remain quite low comparatively to what's found in Etterna and played by top tiers.

I don't have a major issue with AAA difficulty being largely represented in a chart's difficulty because I feel nerfing some charts that are extremely hard to AAA but understandable at a much lower skill level penalizes what a AAA is worth on those charts.

Take Husigi vs. Jamais Vu. JV at time of release was viewed as an entry level 13 and it likely remains as such to this day. Most players are going to experience a slingshot effect of sorts when scoring on both charts. Players approaching the D6 level might be able to score in the 20-30g range on Husigi without even being able to read Jamais Vu. Eventually they'll hit a higher reading/speed/stamina threshold that allows them to read JV, and now they're reaching that same scoring range there. Meanwhile they've only made marginal gains on Husigi because the relevant parts of the chart represented in its difficulty require technical consistency and control not demonstrated as effectively as raw speed at higher levels. Players will continue making greater gains in speed and stamina and get to a point where they're consistently near AAA'ing JV on any playthrough. It's at this point in skill where I'm willing to bet a majority of players will say a Husigi AAA is much more impressive than on a chart like JV, and certainly more impressive than anything under 100.

The entry requirement for a top 20 rank on Husigi, a chart that's been out for 8 years, is 6g. Rave 7, also viewed as an entry level 13 released in the same tournament as Husigi, requires 1-0-0-1. Miku at 102 even has a stricter top 20 entry requirement at 4g. Dropping Husigi's difficulty to the FGO range penalizes the rating value that should rightfully be gained by any player capable of perfecting a chart like that, as well as any other charts that exhibit similar scoring trends. I think it's best to identify these files and create separate equivalency decay formulas for them instead of outright nerfing the difficulties. I feel only a small minority of top players would value a Husigi AAA consistently less than the charts sitting in the 100s bracket.

The extreme high end (105+) can be adjusted to better account for physical capabilities required for playing these charts. Again, these are charts that even the top players (less two or three atm) will not be approaching AAAs on, and as such should have separate equivalency decays. Up until this point there have only been a small handful of charts that any D7+ player might not be able to keep up with, and they're all 100+ in rating already anyways (other offenders in the 90s range might be charts like Serious Shit or thinking of you but....that's basically it?). RATO was placed where it was in difficulty without having anything to compare with for a decade. I have no qualms with nerfing its difficulty on the basis that it's much easier to make it through with a semi-respectable score compared to Wanderflux or HQR (and I'm sure more charts to come throughout the tourney). Establishing a separate decay formula for charts of this difficulty needs to happen. It's fair to say 20g on Wanderflux is deserving of at least 100+ equivalency, but you'd need to shoot its rating up a few points further to hit that point with current scaling.

More thoughts later Maggles is annoyed I'm ignoring her.
__________________


Quote:
Originally Posted by ilikexd View Post
i want to be cucked by cirno

Last edited by One Winged Angel; 08-26-2019 at 01:35 PM..
One Winged Angel is offline   Reply With Quote
Reply


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump



All times are GMT -5. The time now is 02:58 PM.


Powered by vBulletin® Version 3.8.1
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Copyright FlashFlashRevolution