forum

Removing success rate stats.

posted
Total Posts
36
This is a feature request. Feature requests can be voted up by supporters.
Current Priority: +71
Topic Starter
Wishy


This stat is useless and misguiding. On most cases you get a low success rate even when the map is easy since most people retry/wait until fail because they missed and they wanted a full combo. Plus you can get tons of fails/retries because FL players, DT players retrying a lot too, same with every other mod. Meaning you can get a really low success rate even on a very easy map.

Then again you got very hard maps like FREEDOM DiVE where the success rate is actually ridiculously high since truth is that map has been cleared no more than 5 times (I'm pretty sure of this), and most clears are because of no fail/half time/easy, which doesn't really make any sense.

Plus, some maps have a very low HP Leech, so even when they may be completely insane, clearing wouldn't be much of a challenge.

Examples:

http://osu.ppy.sh/b/129891&m=0

Success rate on Another is almost the same as FOUR DIMENSIONS, one being just a very tiring stream map, while the other is by far the hardest map ever ranked.

http://osu.ppy.sh/s/24313 and http://osu.ppy.sh/s/48979

Airman has a higher success rate, and it's obviously WAY HARDER. But:

On Airman most people don't even try to FC and play it just for fun, HP Leech is low so clearing isn't that hard. Meanwhile, Mythologia's HP Leech is higher, you got lots of streams meaning some epic fail on any of those means failing. Plus it's actually possible to FC/easy to get a high combo on so you may get lots of retries/intentional fails after a miss.

tl;dr: Misguiding data and almost useless since it doesn't really say anything, and you can get a better reference of how hard the map is by looking at the top scores.
Leechertyper
I feel they should go with Score Distribution
deadbeat
might be interesting having it as a personal statistic somewhere
Kuro

Wishy wrote:



This stat is useless
It doesn't even show anything when you click taiko so I'm pretty sure the entire taiko community wouldn't care what happens to it.
XPJ38
I support.
My two cents: Points of Failure graph should be removed as well because it's also quite useless. I mean, I'm pretty sure that the majority of people don't look at it.
Aqo
I also agree that this graph is quite possibly the most useless thing ever for anybody who knows how this game works and it does no more than to mislead beginners. Removing it would trim away unnecessary clutter.
Flanster
It could be replaced by map rating IF removed that is.
Aqo
Map rating is just as useless and should be removed as well :b

The fact that map rating is actually *mapset rating* makes no sense since every map in a mapset is completely different and often even made by different mappers with completely different styles, and yet people get to vote after playing only one map; there is also the fact that if people didn't perform well on a map it's because they didn't understand it well, so they'll end up rating it only by the song and not by the quality of the map itself. If map rating was per-map based (as opposed to per-mapset) and you were only allowed to vote after getting at least an A rank (A/S/SS) on the map, only then it would remotely make sense. As of right now it's a useless stat as well.
theowest

Lolicore Flandre wrote:

It could be replaced by map rating IF removed that is.
Difficulty specific rating? nope
TheVileOne
You shouldn't suggest removing useless stats. They are useless, there is no benefit to having them removed, and even if there was a possibility of there being a use, that would be more beneficial to keep it there instead of removing it.

Besides it isn't useless, you can tell where most people fail on a song. Also if more people fail than retry. That stat obviously wont be perfectly accurate but it sure isn't so inaccurate that the results would be wrong in 90% of cases. Edge cases don't matter, because such a system can never be perfect, so stop complaining about the edge cases being misleading.
MillhioreF
I'm gonna have to agree with TheVileOne - it isn't COMPLETELY useless. As long as it has a small use (it tells where most people fail or retry) then there's no reason to remove it for the sake of removing it.

Of course, if we get some better, more important information, then that would be a good reason to consider removing it for the sake of space to fit the new information in. This doesn't seem to be the case at the moment, though.
theowest
so let's improve it.
Topic Starter
Wishy

TheVileOne wrote:

You shouldn't suggest removing useless stats. They are useless, there is no benefit to having them removed, and even if there was a possibility of there being a use, that would be more beneficial to keep it there instead of removing it.

Besides it isn't useless, you can tell where most people fail on a song. Also if more people fail than retry. That stat obviously wont be perfectly accurate but it sure isn't so inaccurate that the results would be wrong in 90% of cases. Edge cases don't matter, because such a system can never be perfect, so stop complaining about the edge cases being misleading.
It is a problem because it is misguiding, according to success rate data every map, even easy ones.

People sometime just stop playing until fail when they miss. Only use you can get from that graph is "where the hard part is", but then again I don't really know anyone who actually looks at that. It's almost useless and misguiding.

AR is still missing, both In-Game and on beatmaps pages.
Hiryuu_old
+1 support

When I started playing osu! on 2011.. back when I used to browse for songs before I gained supporter, I saw that there was a graph there but I never really saw it to be any use of information towards me even as a beginner. I used to think that yeah, that graph doesn't tell me anything..so I just downloaded the song and saw it for myself if it's right for me to practice on it or not.
Also, I'm quite certain that most users who have supporter don't even look at the graph nor care about it (I don't). If anything, then maybe some users do who like to observe it, but I don't think the majority do.
SunBurn
Excellent post!
Support from me!
winber1
... what success rate?

does it even exist
theowest

winber1 wrote:

... what success rate?

does it even exist
Topic Starter
Wishy
Apparently that map is really easy to clear. :s
TheVileOne
Apparently 4 percent success rate is considered easy to clear to some people. Remember that's 4 percent of plays in general. These plays can be from the same people.
jemhuntr
But if i'm a beginner, I'll think that a 4% success rate is so low. And that makes me think that 96% plays are fails, even if around 80% of those plays are actually just retries.
Success rate shouldn't include retries and quits since technically, you didn't fail in those conditions.

Apparently people adapt to the environment. Having this system for so long made people adapt to how it currently works, so old players think it's normal, but new players would think it's weird.
TheVileOne
I think in the OP's example 92 percent are fails and the other 4 percent is retries.

I agree though, showing only fails would be a more accurate depiction of difficulty.

Anyways if we're talking about useless panels here, I think the rating spread graph is a lot more useless than this. The rating spread graph doesn't even have numbers attached to it. It should be removed before this panel.
Winshley
The only stats that is being used here is how many time players have played into the map. You can see the "[Artist] - [Song Title] [Diff Name] has been played 10,000 times!" at #announce channel. Maybe it's just to show you which map is being mainstream right now... :lol:

Also, the Success Rate is already per difficulty-specific, not per mapset.

TheVileOne wrote:

Anyways if we're talking about useless panels here, I think the rating spread graph is a lot more useless than this. The rating spread graph doesn't even have numbers attached to it. It should be removed before this panel.
We don't even get to see how many players have voted the map, though the "Like/Dislike" spread already showed you.
jemhuntr
Hell, I don't even know how that user ratings are computed. I man that red-green bars on top of rating spread.
winber1

theowest wrote:

winber1 wrote:

... what success rate?

does it even exist
there is something called sarcasm; you might have heard of it.
Dexus
In my honest opinion the whole section that shows detailed information should be turned into some sort of drop down box or showable area like in user profiles. It should be a global setting that way it's not something you have to keep hitting to hide or show making it useless. This would make it easier on your eyes that way you don't have to see this statistic that you seem to not like very much. To be frank I don't even pay attention to the whole section unless I'm looking for something specific every now and then. I'm certain this happens with a lot of others as well.
Topic Starter
Wishy

TheVileOne wrote:

Apparently 4 percent success rate is considered easy to clear to some people. Remember that's 4 percent of plays in general. These plays can be from the same people.
4% for the hardest map is actually extremely high, since super easy to clear maps get like 8%.

This thing also takes into account No Fail plays (lol), same with all other mods. Freedom Dive clear rate should be way below 1%, but it is way higher because of mod plays. There is no real way to discuss this it's just misguiding data. Even on easy difficulties you see people retry a lot, it's not that they retry because it's hard to FC, they retry because they are playing DT HR HD FL and get a 100.
Aquamarine
Ya it's needs a revamp. I don't agree with removing it.
Oinari-sama
The Success Rate doesn't make much sense, but I like the Point of Failure graph. Maybe it would've made more sense to show Failure Rate (Fail only, excluding retry)?

EDIT: On second thought, Failure Rate is dodgy too since some players (myself included) sometimes fail on purpose if they miss FC.
Topic Starter
Wishy

Saph1 wrote:

Ya it's needs a revamp. I don't agree with removing it.
People still fail maps intentionally after missing or getting bad acc so it would indeed keep being misguiding. Then again you get some people who pause and retry when they are gonna fail so in any case this would never be any accurate.
endlessHorizon
Keep the graph! But remove the Success Rate
Zare

Wishy wrote:

TheVileOne wrote:

Apparently 4 percent success rate is considered easy to clear to some people. Remember that's 4 percent of plays in general. These plays can be from the same people.
4% for the hardest map is actually extremely high, since super easy to clear maps get like 8%.

This thing also takes into account No Fail plays (lol), same with all other mods. Freedom Dive clear rate should be way below 1%, but it is way higher because of mod plays. There is no real way to discuss this it's just misguiding data. Even on easy difficulties you see people retry a lot, it's not that they retry because it's hard to FC, they retry because they are playing DT HR HD FL and get a 100.
Exactly my thoughts, especially the fact that No Fail plays are considered "successful"
Big Black has a success rate of almost 10%, which is insanely high, it's still probably the 2nd hardest map..
It's so high because a LOT of players play it for fun with No Fail (btw, these are the same people who voted for it for the Best of 2012, but yeah)
Also the amount of people retrying because 100 or Miss.
This stat is just misleading for beginners, which is a bad thing, which then again means removing it would be a good thing and would improve the current situation.

support

edit: oh yeah, not only No Fail plays, but also Halftime plays. That's just stupid.
TheVileOne
Beginner players don't look at a map and see a 10% pass rate and thing that's the song for them. It's not a bad thing, because normal songs average a 20% pass rate and the harder a song is, the more red there is in the graph. so if I were an intelligent beginner who doesn't manage to read the difficulty names for whatever reason (not so intelligent), i would pick up that the more blue in the graph, the less intense that map is usually. This will be true for the large majority of approval maps.

Also it's not useless, because I can tell without even playing Big Black that it has a deathstream or deathjump related element in it by looking at the massive red spike right at the start. It may not have an official use, but the data is useful to know. If I were wondering where people were having problems in my map, I could look at that graph to find clues. If we didn't have it, there would be no easy way to figure out such information.
Friggles
I think it should be replaced with some graphic showing what percentage of players can achieve a certain rank. I.e. if a player got a B he would also count as a person who can get a C. Plays with difficulty reduction mods should be excluded, except for maybe No Fail. I think that would be pretty useful and fun to look at :D.

Difficulty increase mods could be a problem, there would have to be some distinction between a normal rank and a HR,DT,FL,HD rank.
boat
das bump

I get that it's inaccurate and should either be reworked or removed, but I don't see the issue in it being misleading. I don't think new players actually care about any difficulty indicators or graphs, they'll download the map, realize that it's too hard and simply not play for now.
Friggles
I just thought maybe each rank could be looked at in detail, so for a certain rank you can see how many of each mod there are - difficulty increase and decrease mods. But that might get a bit complicated.
Babosu
And it still exists
Please sign in to reply.

New reply