the "Average" Scorehero player

**dnaygs** · Joined: 22 Mar 2006 Posts: 101

if the scores are stored on a database, the average score for each song could easily be calculated. I don't think randomly guessing how many FCs someone might have is very accurate. Also you have to take into account people that don't play on expert.

**Dynasty99** · Joined: 01 Nov 2006 Posts: 911

**jonnyg111** · Posted: Tue Nov 04, 2008 11:23 pm Post subject:

My perception before GH3 came out was : If you can just barely beat Jordan, then you were exactly average. 5*'d, or above 350k, you were pretty good. 400k+, you are a legend. FC : hadn't happened when i first came up with this :P (you guys would be surprised how long I lurked before I joined, I was scared to join until I had passed Jordan

)

My perception now : If you got world tour near or on release day, and 5*'d the whole game (maybe without satch boogie) on sightread, then you are above average. If it took you a couple of tries to 5* everything, then you are like me, still above average, but not recognizable.

Another way I perceive it : If you can pass TTFAF, just barely, then you are exactly average. This is the easiest way I can think of figuring it, Balancing out numbers of FC's and rank and scores just doesn't appeal to me... I have NEVER used a SP path in my life, and I'm top 400 GH3 wii I think... I haven't updated that stuff in a long time though. If you base it on score or rank, that's just their ability to look up how to use SP on the internet, not their ability to play. Passing TTFAF, now THAT'S an achievement (literally) that we can go by.

I'm probably wrong in most of your eyes, but this is the way I've always thought about it.
_________________

You have died of dysentery. Put in your sig if lemons.

**Skysore** · Joined: 09 Oct 2007 Posts: 582

I think the only way you can do this for expert is to take all of the carreer scores and average them
_________________

SoCal Get-Together anyone???
GOALS:
900K TTFAF []
700K TDWDTG [x]
Start Playing Again []
My Accomplishment Thread!
Check out my Youtube!

**Seraphixla** · Posted: Wed Nov 05, 2008 1:22 am Post subject:

**p41** · Joined: 27 Jul 2006 Posts: 439 Location: Boston, MA

**YetiPirate** · Joined: 13 Aug 2007 Posts: 218 Location: Oregon

**g0dly** · Joined: 07 Feb 2008 Posts: 635

well i would assume that the average sher would be determined by what songs the average person can pass. if that is the case, you would need some way of determining which users can pass which songs. the hard part about this would be that there are alot of users like me that only input some of their scores even though they can pass all of the songs. i dont see how you could do it with the info on the site.

**RBurgandy** · Joined: 27 Mar 2008 Posts: 186 Location: Vermillion, SD

**NfskMjmMal2** · Joined: 27 Aug 2008 Posts: 1000

True, we could only have "the average [insert game] player". But I don't think that going by rankings is a sure enough method, because not everyone submits scores for every song. I agree that in order to find out how "good" the average player on each game is, the average score on any and every given song would be needed.

I also hadn't come to the realization that the reason why the average skill level seems so high is because the level of the frequent posters is generally much higher than those who don't post. There are probably more people with exactly 0 posts than people with 200 or more. I too think that Sil3nce is a fairly accurate representation of the average, but I guess we'll only find out with the full data, or not find out at all.

**Bearclaw67** · Joined: 12 Jun 2008 Posts: 79 Location: Ohio, United States

**YetiPirate** · Joined: 13 Aug 2007 Posts: 218 Location: Oregon

**hydrapower** · Joined: 28 May 2006 Posts: 1708 Location: Rochester, MN

WARNING: MATH/STATS AHEAD!

I attempted to come up with a scientific answer to this question, rather than just coming up with an educated guess. I took the population of GH3 Expert scores (20,142 users) and from there extracted a sample of 100 random users by randomly generating numbers from random.org's Integer Generator. In statistics it's not usually necessary to use the entire population's data to find out characteristics of that population - a large enough sample will do. e.g. when they poll people to find out what the President's approval rating is, they don't ask everybody in the country, they only ask about 1000 people. I figured 100 was a large enough sample for this.

The question that arose is what do you do with people that haven't entered all their scores, for one reason or another. I didn't want to assume that they couldn't pass all the songs, so I took a simple projection of what they would have had they entered scores for all 70 songs, by simply calculating the average score they had across what they entered and multiplying it by 70. Naturally, this leads to some wacky projections when you have people that just enter their TTFAF score and leave. In those cases, where the projected score was higher than any real person's actual score (>24 million), I threw those away. There's probably a better way to go about it, but I just used this method for simplicity.

Then the question of FC's arises: do we project the number of FC's someone could get? I say no, because:

1) Someone may FC a number of songs on the first half of the game (difficulty-wise) and it would not seem likely that they could FC the second half of the game at the same rate.
2) My perception is that when someone FC's a song they are more likely to be proud of that score and enter it into SH than to sit on it.
3) Some people entered only a few scores and they were all FC's, so they would project out to 70/70 FC's

Accordingly, I decided that unless you've entered the FC into the database, you don't have it.

Okay, so with that explained, here is what I found:

The mean number of scores entered was 31.07.
The mean total score was 8,069,794.
The mean projected total score was 17,995,613.
Of the 31.07 scores entered, 26.79 were 5-stars (86.2%). This doesn't mean the average user can 5-star 86.2% of the setlist, it means that when the average user enters a score, 86.2% of the time it's a 5-star score.
The mean number of FC's was 7.

Thoughts, comments, criticisms? If you've read all the way down here, thanks.

_________________

My Backloggery
My Drum Accomplishments: Guitar Hero / Rock Band

Now Available in e-book format! Book 1 Book 2
Otharia - Science Fiction/Fantasy from some aspiring writers, check it out! Updates weekly!