| View previous topic :: View next topic |
| Author |
Message |
chronoboy
Into the past with a splash
Joined: 04 Jun 2010
Posts: 162
Location: Canada
|
 Posted: Sun Nov 21, 2010 11:40 am Post subject: JSH357 Top 30 for Winter 2010?!? |
 |
|
Are we going to have a Top 30 list of Winter 2010 which may include the recent additions coming from the Heart of OHR contest? You seemed to have stopped completely after 2009, it's kind of sad to see, those lists were great to have.
Perhaps even starting a Top 30 RPG list, which would just include games in the RPG genre as a tribute to the Heart of OHR.
For those who don't know what lists I am referring to:
http://www.jsh.illuminative.org/top30.html
_________________
Current project: Chronoboy Adventures
Website: http://www.chronoboy.com/ |
|
| Back to top |
|
 |
JSH357
Joined: 02 Feb 2003
Posts: 1705
|
| Posted: Sun Nov 21, 2010 1:50 pm Post subject: |
|
|
I have lost interest in it due to lack of interest and people insulting/whining about the lists whenever I make them.
If I can have a guaranteed 15 or so people participate and actually submit quotes, I will consider running another poll. That doesn't seem likely. |
|
| Back to top |
|
|
Fenrir-Lunaris
WUT
Joined: 03 Feb 2003
Posts: 1747
|
| Posted: Sun Nov 21, 2010 10:45 pm Post subject: |
|
|
I'll vote.
If you include votes from people who AREN'T in the OHR community, that might be interesting, to say the least... |
|
| Back to top |
|
|
JSH357
Joined: 02 Feb 2003
Posts: 1705
|
| Posted: Mon Nov 22, 2010 7:43 am Post subject: |
|
|
| Fenrir-Lunaris wrote: | I'll vote.
If you include votes from people who AREN'T in the OHR community, that might be interesting, to say the least... |
"Outsiders" tend to have only played one or two games and would be too big of an outlier to make any statistical sense in charting. I'm not saying I don't value their opinions; they just tend not to have a wide enough knowledge of the medium. This can also lead to vote-stuffing potentially. |
|
| Back to top |
|
|
chronoboy
Into the past with a splash
Joined: 04 Jun 2010
Posts: 162
Location: Canada
|
| Posted: Mon Nov 22, 2010 12:12 pm Post subject: |
|
|
Doesn't the CP gamelist and other gamelists have a popularity gauge? Instead of using a voting system, count how many people likes the game in the specific time frame. Say before the time frame, 80 people listed as liking it, then in the time frame, this number jumped to 95. Now 15 people liked it during the winter season.
To think about, I am quite sure that CP gamelist has a rank system for all the games, I'm sure if you had access the database data, this entire process I explained could even be automated. It's simple math really. Who maintains the CP gamelist database here? I could send them a concept of how this would work for they could implement it into CP. At the moment, the ranking system is lifetime wide, it does not provide statistics based on season or years for example.
_________________
Current project: Chronoboy Adventures
Website: http://www.chronoboy.com/ |
|
| Back to top |
|
|
JSH357
Joined: 02 Feb 2003
Posts: 1705
|
| Posted: Mon Nov 22, 2010 12:15 pm Post subject: |
|
|
| CP does have those things, but almost nobody uses them. The highest ranked game on the site isn't even UPLOADED anymore. Manual polls are actually more accurate in that light (and also more fun when they actually work). |
|
| Back to top |
|
|
JSH357
Joined: 02 Feb 2003
Posts: 1705
|
| Posted: Sun Nov 28, 2010 5:45 pm Post subject: |
|
|
I got to thinking about this again yesterday, actually.
Chronoboy, would you be interested in taking over the Top 30 polls? You seem to know your way around web design, and you might be the perfect man for the job.
One thing that is pretty important is to a) post in every OHR hub and b) bug certain folks occasionally. If you look up the past threads, you can see how I got the information.
All of the scoring was done with a simple (x + y) formula:
x = The score derived for a particular game from how people ranked their individual choices
y = The number of votes a game received
I found this to be the most accurate scoring method, although the low number of voters tended to make the charts skew wildly over the months. |
|
| Back to top |
|
|
Voltire
Yep.
Joined: 16 Oct 2010
Posts: 87
Location: A nice man's big, white van.
|
| Posted: Sun Nov 28, 2010 5:50 pm Post subject: |
|
|
| If Chronoboy cannot do it, then, JSH, I will vote as well. I'm new to the forums, but I've played the games for at least 2 years. Either way, it seems like a good idea to do again. |
|
| Back to top |
|
|
chronoboy
Into the past with a splash
Joined: 04 Jun 2010
Posts: 162
Location: Canada
|
| Posted: Sun Nov 28, 2010 8:37 pm Post subject: |
|
|
I would love to do this, however my only obstacle is that for some reason, I am unable to sign-up for SlimeSalad(a more active forum), during the registration, I seem to get a white screen of death with no error messages or anything... I went a PM to Moogle1 here on CP, as I believe he is Mogri on SlimeSalad, whom manages the website itself. Perhaps he could manually register me and PM me a temp password that I can change.
It's strange though, SlimeSalad is the only phpBB forum giving me so much trouble just to register. Lettuce does go in Salad, right? Or is it expecting something less generic like a Radish.
_________________
Current project: Chronoboy Adventures
Website: http://www.chronoboy.com/ |
|
| Back to top |
|
|
JSH357
Joined: 02 Feb 2003
Posts: 1705
|
| Posted: Mon Nov 29, 2010 1:06 am Post subject: |
|
|
| I'm pretty sure you aren't the first to have trouble. I'm a moderator there too; If Mogri doesn't get back to you (he will) I will bug him. |
|
| Back to top |
|
|
Spoon Weaver
Joined: 18 Nov 2008
Posts: 421
Location: @home
|
| Posted: Wed Dec 01, 2010 8:31 am Post subject: |
|
|
| I see you're on slimesalad now. I'm glad this worked out. |
|
| Back to top |
|
|
Moogle1
Scourge of the Seas
Halloween 2006 Creativity Winner
Joined: 15 Jul 2004
Posts: 3377
Location: Seattle, WA
|
| Posted: Wed Dec 01, 2010 4:33 pm Post subject: |
|
|
Most of the people who I know to have had problems registering were foiled due to the CAPTCHA. Your problem specifically seems to have been a server error or something, because it was broken and suddenly fixed again all while I was on vacation.
_________________
|
|
| Back to top |
|
|
TMC
On the Verge of Insanity
Joined: 05 Apr 2003
Posts: 3240
Location: Matakana
|
| Posted: Sat Dec 04, 2010 4:49 am Post subject: |
|
|
| JSH357 wrote: | All of the scoring was done with a simple (x + y) formula:
x = The score derived for a particular game from how people ranked their individual choices
y = The number of votes a game received
I found this to be the most accurate scoring method, although the low number of voters tended to make the charts skew wildly over the months. |
This is of course equivalent to 11 points for 1st ranking through to 2 points for 10th. But I think we can find a better scheme than that. I agree that the number of people voting for a game is important. I suggest 14 points for first place and 5 for tenth.
There are lots of people who only rank a few games, and they tend to skew the points of those games quite severely because there are few voters. I think last time you made 5 games the minimum, which worked quite well? Or did you let us vote for more than 10? Both?
_________________
"It is so great it is insanely great." |
|
| Back to top |
|
|
JSH357
Joined: 02 Feb 2003
Posts: 1705
|
| Posted: Sat Dec 04, 2010 6:24 am Post subject: |
|
|
| The Mad Cacti wrote: | | JSH357 wrote: | All of the scoring was done with a simple (x + y) formula:
x = The score derived for a particular game from how people ranked their individual choices
y = The number of votes a game received
I found this to be the most accurate scoring method, although the low number of voters tended to make the charts skew wildly over the months. |
This is of course equivalent to 11 points for 1st ranking through to 2 points for 10th. But I think we can find a better scheme than that. I agree that the number of people voting for a game is important. I suggest 14 points for first place and 5 for tenth.
There are lots of people who only rank a few games, and they tend to skew the points of those games quite severely because there are few voters. I think last time you made 5 games the minimum, which worked quite well? Or did you let us vote for more than 10? Both? |
I would argue that having different amounts of games in each vote really hurt the polls, actually. When I started doing that, I noticed a lot of the huge score differences between games appearing, since people who only rate 5 games tend to like the same ones (Wandering Hamster is a good example as it always has a significant lead over other games). I guess the system could still work, but I would change the minimum to 10. Unfortunately, that might cause fewer people to vote.
I agree that there are cases where my addition didn't work right. For instance,
10 people rank SOJ 5/10:
(6*10)+10 = 70
5 people rank OHR House 10/10:
(10*10)+5 = 105
Personally I think SOJ should either tie or score higher than OHR House in this case. 5 more people liked it enough to rank it. I was always afraid of changing the formula too much since people were already pretty critical of my lists.
Under your system it would be:
(5*14)+10 = 80
and
(10*10)+5 = 105
I like this system more myself. Only thing is, does it actually work, or does it require a significant amount of voters to work? Hmm... Only a poll would show the truth. Perhaps the # of voters bonus would be better if it were proportional to the number of total voters. |
|
| Back to top |
|
|
Pepsi Ranger
Reality TV Host
Joined: 05 Feb 2003
Posts: 493
Location: South Florida
|
| Posted: Sat Dec 04, 2010 9:24 am Post subject: |
|
|
| JSH357 wrote: | 10 people rank SOJ 5/10:
(6*10)+10 = 70
5 people rank OHR House 10/10:
(10*10)+5 = 105
Personally I think SOJ should either tie or score higher than OHR House in this case. 5 more people liked it enough to rank it. I was always afraid of changing the formula too much since people were already pretty critical of my lists. |
Why are so many people afraid of averages? Why not just take an average score per game and multiply that by the number of times it's been voted on?
10 people rank SOJ 5/10:
5*10 = 50
5 people rank OHR House 10/10:
10*5 = 50
Tie.
Here's a more likely list:
3 people vote Wandering Hamster 10/10
2 people vote it 9/10
5 people vote it 8/10
3 people vote it 7/10
total 13 people
3*10 = 30
2*9 = 18
5*8 = 40
3*7 = 21
Average: 109 / 13 = 8.38
2 people vote Vikings of Midgard 10/10
4 people vote it 9/10
3 people vote it 8/10
1 person votes it 7/10
0 people vote it 6/10
3 people vote it 5/10
total 13 people
2*10 = 20
4*9 = 36
3*8 = 24
1*7 = 7
0*6 = 0
3*5 = 15
Average: 102 / 13 = 7.85
Under this system, Wandering Hamster wins.
However, we all know that voter turnout varies, so you'll want to divide the average based on how many people vote, not on how many people vote on a specific game.
So, if 15 people vote in total, but only 13 vote on Vikings of Midgard and Wandering Hamster specifically, then you may change the average to:
Wandering Hamster
Average: 109 / 15 = 7.27
Vikings of Midgard
Average: 102 / 15 = 6.0
Wandering Hamster still wins.
And just to show contrast:
1 person votes Yo Ghost 5/10
2 people vote Yo Ghost 3/10
Without total voter turnout:
5*1 = 5
3*2 = 6
total 2 people
Average: 11 / 2 = 5.5
versus with total voter turnout:
Average: 11 / 15 = 0.73
So, now your rank so far is:
1st) Wandering Hamster (7.27 average)
2nd) Vikings of Midgard (6.0 average)
3rd) Yo Ghost (0.73 average)
I think this is the most fair, as it rewards both rank and voter number without skewing the numbers too drastically. It also permits you to allow however many or few votes the voter wants to submit.
If a player has only played Sword of Jade 2 and votes only on that, then Sword of Jade 2 gets an automatic 10. He'll add to the voter turnout average, but his single vote won't kill the leaderboard. And if he's the only one who votes on it, then that perfect score of 10 in perspective to total popularity still only scores:
10 / 15 = 0.67
Worse than Yo Ghost because at least two people thought Yo Ghost was worth putting on the list, where the perfect 10 game only had one supporter. (But the margin was still very close, as you can see.)
I think the system of averages with voter turnout could work. What does everyone think?
_________________
Progress Report:
The Adventures of Powerstick Man: Extended Edition
Currently Updating: General sweep of the game world and dialogue boxes. Adding extended maps.
Tightfloss Maiden
Currently Updating: Chapter 2 |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
