>> Calculate attendance outside of ratings

[lackattack]

Concise description:

• Show attendance % on the profile, remove attendance attribute from ratings system.

Specifics:

• Attendance is more objective than subjective, so there is nothing more annoying than an unfair attendance rating. We'll remove it from ratings alltogether.
• Unlike other suggestions to have fake stars based on how long it takes you to take your turns, this suggestion is to simply display % turns taken on the player profile (as suggested earlier here).
• It could be nice to factor this attendance % into your overall ratings score, but let's see how this round of changes to ratings shapes out before making the formula too complicated.

This will improve the following aspects of the site:

• Ratings will be based on attributes that are subjective, not objective.
• Players will not waste their time assessing something that can be automated (more or less ).

Please give your input on this suggestion!

[The Fuzzy Pengui]

I love it!

[wicked]

Please explain how you plan to figure the attendance percentage. Is it going to be a missed turn percentage I hope? IMO this should be turn-based, not time-based.

[lackattack]

Agreed. It should be the % of your turns that you clicked "Begin Turn".

[Incandenza]

So, if I understand correctly, there'll be something in my profile that says something along the lines of "Incandenza has taken 98 of 100 total turns (98%)". Not a bad idea, tho I have a feeling that all but the most serial of deadbeats and newest of noobs will have anything less that 98-99%.

[PLAYER57832]

I love the idea of automating attendance!!!!! No matter how it is done.


But as Incandenza mentioned, percentages would probably be too low to illuminate even "chronic" deadbeaters.

An "average turns missed per game" would be more likely to show something, but would blurr misses in short games with misses in very long games, unfortunately.

What about listing the numbers of turns missed versus game numbers just as straight numbers?

# turns missed/ # games

Ideally, there might be some distinction between turns missed in speed games, short casual games and games that go on for weeks or months. I take quite a different view of someone who misses 2 turns in a year-long game (or even a dozen!) and someone who misses even 1 in a shorter game without reason.

Example: # missed turns / speed games
# Missed turns / short games (under a month)
# Missed turns/ long games (over 1 month)


these could be listed as #/#/# with different colors for each number or just combined, with a "clickable" screen showing the breakdown.





I would also like to see some space -- perhaps in the member's profile, perhaps adjoining this rating -- where someone can list, BRIEFLY mitigating circumstances. (self- proclaimed)
example: computer issues in past ... now resolved
cannot play on Saturdays
etc.

[gloryordeath]

A great idea and fix for this problem!

[lackattack]

So, if I understand correctly, there'll be something in my profile that says something along the lines of "Incandenza has taken 98 of 100 total turns (98%)"

Correct. It is what jiminski suggested here: viewtopic.php?f=4&t=27035

I have a feeling that all but the most serial of deadbeats and newest of noobs will have anything less that 98-99%.

I did some number crunching, and amongst active players on the scoreboard we'll see this breakdown:

attendance record (% turns taken) -> distribution
100 -> 9%
99 -> 4.4%
95-98 -> 10%
90-94 -> 5.5%
80-89 -> 6%
51-79 -> 7.5%
4-50 -> 17.5%
0 -> 19%

However this doesn't add up to 100% so maybe I screwed up somewhere

[ParadiceCity9]

Will this account for every turn you've ever taken, ever?

[The Fuzzy Pengui]

However this doesn't add up to 100% so maybe I screwed up somewhere

Can we have someone else right the code for it please?

[wicked]

gimme the numbers lack, let the engineer do it.

[Incandenza]

I did some number crunching, and amongst active players on the scoreboard we'll see this breakdown:

attendance record (% turns taken) -> distribution
100 -> 9%
99 -> 4.4%
95-98 -> 10%
90-94 -> 5.5%
80-89 -> 6%
51-79 -> 7.5%
4-50 -> 17.5%
0 -> 19%

However this doesn't add up to 100% so maybe I screwed up somewhere

Well, assuming your numbers are in the ballpark, that's still a much wider dispersal than I would've thought. And really, the whole point seems to be avoiding serial deadbeats, as opposed to people for whom real life occasionally intervenes. So I'd say this proposed system will work quite nicely.

I'd be curious to know who has the most turns played and still has 100% attendance.

[wicked]

So how do these freestyle situations affect the proposed attendance calculation? Neither should count towards attendance.

1. Person killed off before their first turn. They had the oppty to click "begin turn", but were killed off before they did.

2. In the old freestyle, where you could force someone to miss a turn?

[chipv]

1. Person killed off before their first turn. They had the oppty to click "begin turn", but were killed off before they did.

2. In the old freestyle, where you could force someone to miss a turn?

Do these actually count as missed turns in the game log, out of interest?

[wicked]

1. no
2. yes

[chipv]

1. no
2. yes

2 is relevant, then. If lack is going through the game logs/querying database for data linked to game logs then the missing logs in Feb also will skew the data.

[vikprez]

I love the idea...I played two games with one player and missed NO turns and was given a 2! But then again he gave me a 2 for everything so who knows what he was thinking

An automated system seems like a more truthful representation.

[eye84free]

wonderfull idea..

[cicero]

I would also like to see some space -- perhaps in the member's profile, perhaps adjoining this rating -- where someone can list, BRIEFLY mitigating circumstances. (self- proclaimed)
example: computer issues in past ... now resolved
cannot play on Saturdays
etc.

Players could use their Wall to make such announcements ?

[yeti_c]

It would also be useful to see how many people "ran out of time" on their turn...

i.e. how many people use the cheap freestyle double turn tactic.

Could that be included in the stat too?

C.

[Visaoni]

I would also like to see some space -- perhaps in the member's profile, perhaps adjoining this rating -- where someone can list, BRIEFLY mitigating circumstances. (self- proclaimed)
example: computer issues in past ... now resolved
cannot play on Saturdays
etc.

Players could use their Wall to make such announcements ?

While that would be a good spot, unless it is listed on a specific page hardly anybody is going to bother checking it. It really needs to be on a page dedicated to the purpose of attendance.

On that note, is there a way to make the attendance rating (or whatever it's called in the end) displayable alongside a players name? Or some way to incorporate it into the rating system, without actually assigning different percentage ranges a star amount? Clicking through up to 7 profiles before deciding to join a game or not somewhat inconvenient. However having |medal|name|star rating|attendance rating| might be cluttered and look strange. Not to mention potentially confusing for new recruits without some sort of explanation, which would only add to the clutter.

[cicero]

Almost Perfect!
I think that a simple percentage of turns taken statistic is almost perfect ...

Reflection ...
Consider what I believe are the categories of turn missers:
(Are there other significant ones I've missed?)

(i) Chronic short term deadbeats.
New players who sign up, don't appreciate the casual nature of CC and deadbeat as a result. And never return.

(ii) Reformed chronic short term deadbeats.
New players who sign up, don't appreciate the casual nature of CC and initially deadbeat as a result. However they subsequently return, get their act together and continue to return quickly becoming regular players.

(iii) Chronic long term deadbeats.
Players who continue to return, but continue to deadbeat to a high degree.

(iv) Chronic long term turn missers.
Players who continue to return, but continue to miss turns to a high degree though not actually deadbeating.
This includes those who miss turns as strategy and those who are chronically unreliable.

(v) Regular players.
Players who continue to return and despite the occasional mishap reliably take their turns.

Enhancements/Tweaks Required ... ?

[A] distinction between (iii) and (iv) may be difficult on percentages alone ...
Chronic turn missers are annoying in a quite different way to chronic deadbeats. My personal preference is for a deadbeat over a turn misser (but I might change my mind), but I think it is important to be able to make this distinction ...

[B] distinction between (i) and (ii) may be difficult on percentages alone ...
Perhaps it would be useful (and this could equally apply to existing ratings) if the data was not shown for new recruits ... so that a user has the opportunity to acquire a body of attendance data/ratings before that average is displayed as representative (eg at present one player has been rated only once and received all ones - this is unhelpful if presented as "representative").
Again I think it is important to be able to make this distinction.

[richardgarr]

It would also be useful to see how many people "ran out of time" on their turn...

i.e. how many people use the cheap freestyle double turn tactic.

Could that be included in the stat too?

C.

I have ran out of time in a few speed games trying to finish a board or make an elimination. This is something that happens to
many speed players. Please do NOT lump us in, with those chronic freestyle abusers.

[yeti_c]

It would also be useful to see how many people "ran out of time" on their turn...

i.e. how many people use the cheap freestyle double turn tactic.

Could that be included in the stat too?

C.

I have ran out of time in a few speed games trying to finish a board or make an elimination. This is something that happens to
many speed players. Please do NOT lump us in, with those chronic freestyle abusers.

A few times will not make much difference to your stats... if you're an abuser - then it will be obvious...

C.

[Pedronicus]

I still firmly believe in my idea that attendance should take into consideration the amount of time it takes you to take your turn.
People who log in 2-3 times a day and are amongst the quickest takers of their goes should be noted. If a freemium wants his 4 games to be played as quickly as possible - he/she needs this sort of information.

attendance

noun
1. the act of being present
2. the frequency with which a person is present