Page 5 of 10
Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 3:05 am
by stalkor
well there is the crux of the problem isnt it. lowering the dropout rate depends on the number of dedicated (aka 1 game per week or more) players in the league. about 70% or 80% of the dropouts occur in the lowest classes (delta) of the league. A lot (and i mean a _LOT_) of players sign up because they like the league and the ideas behind them and then just never come back to KGS ever again. The negative impact this has not only on the league but on the players that actually _do_ want to play league games is really big.
But how does the ASR solve a problem like this, statistics won't tell us how and you can't prospect how a person will play next month because ppl are not numbers.
it is possible to lower the number of games needed to stay but will that make the league more active or less active as a result of it, i think in case we lower the number of games to 3 we will see an increase at first but it will then slowly degenerate into what it was because ppl are pretty lazy:P
Upping the number of minimum games to 5 or higher can make the signup threshold harder for ppl but it will be a cause for more dropouts, so all in all the number wont change much.
i therefor conclude that changing the number of minimum games is not a good option and please correct me if i am wrong on this:)
Motivating players to play more is something i can only do on a limited scale, players should motivate themselves to play because this is a great game and not playing this game (you love, because you want to play in the ASR league which is meant for competitive serious games) is more disgracing to the game then playing a bad game of go. I try to motivate players by giving them prizes if i can, giving them a nice place to play games and learn, try to give them a community which is friendly and helpful and if this is not enough, what else can i do for these players?
Yes i want to lower the dropout rate to 0% if possible (impossible goal) but to keep it real 25% dropout is a goal we've tried to reach the past few months and frankly its damn hard. Also we try to reach the 20% games played each month goal, but it all depends on others to play games. If just one player doesnt play in a class this could cost us 10 or more games, multiply this by 100 dropous this is easily 1000 games thats not played. This impact on the league is huge, and players don't often see this. demotivation by others because players sign up and don't come back online leads to more players not coming anymore resulting in a bigger snowball which is ultimately the downfall of a class. Could i help this? not really...
i guess the answer to how many games is many = 4?
Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 3:27 am
by topazg
My entirely subjective feeling is that 4 is about perfect.
If anything I'm still in favour of even bigger class sizes, but it feels like it's getting much closer now. When I've gone online, there are normally at least one or two people from classes on there.
Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 8:56 am
by ez4u
stalkor wrote:well there is the crux of the problem isnt it. lowering the dropout rate depends on the number of dedicated (aka 1 game per week or more) players in the league. about 70% or 80% of the dropouts occur in the lowest classes (delta) of the league. A lot (and i mean a _LOT_) of players sign up because they like the league and the ideas behind them and then just never come back to KGS ever again. The negative impact this has not only on the league but on the players that actually _do_ want to play league games is really big.
...
i therefor conclude that changing the number of minimum games is not a good option and please correct me if i am wrong on this:)
...
i guess the answer to how many games is many = 4?
topazg wrote:My entirely subjective feeling is that 4 is about perfect.
If we stick with four games and concentrate on the issue of the very high rate of non-performance in delta for the moment, let me ask something. Do you think it would be possible to convert delta into one big bucket, where anyone could play anyone and the goal was to earn your way out of delta and into your first structured class (gamma)? It seems like it would minimize the impact of no/limited shows by greatly expanding the available pool of opponents. It would also create a sort of "probationary" status for deltas that hopefully people would want to qualify their way out of as a distinct goal (motivation). This last might also have an knock-on effect on performance in gamma as people are motivated not to lose their "permanent member" status and drop back into the great "unwashed" in delta. Have you tried anything like that in the past?
Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 9:14 am
by topazg
ez4u wrote:If we stick with four games and concentrate on the issue of the very high rate of non-performance in delta for the moment, let me ask something. Do you think it would be possible to convert delta into one big bucket, where anyone could play anyone and the goal was to earn your way out of delta and into your first structured class (gamma)? It seems like it would minimize the impact of no/limited shows by greatly expanding the available pool of opponents. It would also create a sort of "probationary" status for deltas that hopefully people would want to qualify their way out of as a distinct goal (motivation). This last might also have an knock-on effect on performance in gamma as people are motivated not to lose their "permanent member" status and drop back into the great "unwashed" in delta. Have you tried anything like that in the past?
The big source of complaints the league had at one point with very large classes was the crazy people that managed to fit 60+ games in a month. With only 2 games instead of 3 per opponent now, and I would have said more people complaining about the lack of availability of players as compared to the over abundance of games, I'd say give it a shot, particularly as a "Delta only" trial to begin with.
Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 10:04 am
by Gabalon
If u make one "Deltapot" the few hyperactive ppl won't count much and they'd dominate their small Delta class anyway.
So putting everyone in "the one Delta" and letting the top50 or so (normal promotions + additionals for kicks) promote doesn't sound bad.
I don't think there will be more then 10 very active ppl, so playing a moderate number of games will be ok for promotion.
Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 10:32 am
by oren
topazg wrote:If anything I'm still in favour of even bigger class sizes, but it feels like it's getting much closer now. When I've gone online, there are normally at least one or two people from classes on there.
The problem when I was in ASR was that it was always the same one person.

Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 10:34 am
by Jash
I find the suggestion to have a huge Delta where everyone can play everyone very interesting. The hyper active players as Gab said aren't really a problem since around 40 places are available approximately to promote (so the motivated people will promote anyway). I think it's a great idea to push the competition and motivate people to play more in Delta. At least there will always be someone logged to play with so no one will complain that he didn't find a classmate to play against =)! The only problem that might happen is some players who are highly ranked might not play much after since they'd think they're safe, but if we consider Delta more like a gamma purgatory that shouldn't be an issue =)!
Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 11:46 am
by topazg
Remember also that the top 40 or so would promote

Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 11:53 am
by jts
topazg wrote:ez4u wrote:If we stick with four games and concentrate on the issue of the very high rate of non-performance in delta for the moment, let me ask something. Do you think it would be possible to convert delta into one big bucket, where anyone could play anyone and the goal was to earn your way out of delta and into your first structured class (gamma)? It seems like it would minimize the impact of no/limited shows by greatly expanding the available pool of opponents. It would also create a sort of "probationary" status for deltas that hopefully people would want to qualify their way out of as a distinct goal (motivation). This last might also have an knock-on effect on performance in gamma as people are motivated not to lose their "permanent member" status and drop back into the great "unwashed" in delta. Have you tried anything like that in the past?
The big source of complaints the league had at one point with very large classes was the crazy people that managed to fit 60+ games in a month. With only 2 games instead of 3 per opponent now, and I would have said more people complaining about the lack of availability of players as compared to the over abundance of games, I'd say give it a shot, particularly as a "Delta only" trial to begin with.
Yeah, I've always been a fan of this idea.
Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 3:28 pm
by stalkor
posted a poll on the
ASR site on the right hand side, please vote for one big group in delta or keep as is
edit: posted it
here
Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 4:12 pm
by jdl
One more data point for you. If delta moves to a tub-o-players, I would strongly consider joining again. I dropped out because it was next to impossible to find games. Seemed to be due mainly to time zone differences within the classes.
Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 4:17 pm
by oren
jdl wrote:One more data point for you. If delta moves to a tub-o-players, I would strongly consider joining again. I dropped out because it was next to impossible to find games. Seemed to be due mainly to time zone differences within the classes.
Same here.
Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 4:42 pm
by stalkor
keep in mind that if this happens and you promote to gamma you are still in a 20 player group, but hopefully with more motivated players
Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 4:50 pm
by daal
stalkor wrote:posted a poll on the
ASR site on the right hand side, please vote for one big group in delta or keep as is
edit: posted it
here
I tried voting for yes, but got a javascript error: "Please choose a valid poll answer" Do I have to vote no to get it counted?

edit: works now.
Re: what is the ASR doing wrong?
Posted: Mon Oct 17, 2011 11:29 pm
by usagi
stalkor wrote:well there is the crux of the problem isnt it. lowering the dropout rate depends on the number of dedicated (aka 1 game per week or more) players in the league. about 70% or 80% of the dropouts occur in the lowest classes (delta) of the league. A lot (and i mean a _LOT_) of players sign up because they like the league and the ideas behind them and then just never come back to KGS ever again. The negative impact this has not only on the league but on the players that actually _do_ want to play league games is really big.
But how does the ASR solve a problem like this, statistics won't tell us how and you can't prospect how a person will play next month because ppl are not numbers.
But what's the problem? You let anyone in, and the result is that the population approximately represents "everyone"; people join and just try it out, but they make it to gamma or beta before they quit. That's just how the system works. Now that you're finally considering the bigroom idea from a year or two ago though, please allow me to reiterate how was originally intended to solve this problem;
the "bigroom" idea is intended to only allow the most active players into the "real" league.Let's think very carefully about what the bigroom really means. You have alpha, two betas, and four gammas. If you have 14 people in each class that is around 100 people. Say put 16 in alpha, 100 people exactly. That is the "real league". Then you have everyone else, who fight for their right to enter the real league. Ok, so if you include delta classes then you have 212 people (16 in alpha, 14 in the rest). Same point. What you are doing is placing an entry restriction on the league: "You must be at least this active, and no less active, to play in the league".
So there, you've placed an entry restriction on the league.
it is possible to lower the number of games needed to stay but will that make the league more active or less active as a result of it, i think in case we lower the number of games to 3 we will see an increase at first but it will then slowly degenerate into what it was because ppl are pretty lazy:P
Upping the number of minimum games to 5 or higher can make the signup threshold harder for ppl but it will be a cause for more dropouts, so all in all the number wont change much.
i therefor conclude that changing the number of minimum games is not a good option and please correct me if i am wrong on this:)
There are two ways out of this box; change the entry requirements or change the requirements for people while they are in the league. Again, if you let anyone in, then the league will be filled with anyone. That is both a plus and a minus. If you are willing to live with it, then it's not a minus though.
If I'm right and you are no longer willing to live with this, please consider placing real entry restructions on the league; a bigroom is a really great way to do this, as long as it is meaningful (I say make the bigroom at delta level to promote 28 into gamma)...
4 game rule is fine imo. Lets just try bigroom and see how it works out? please? ^^