Forged Alliance Forever

Re: Crumbly Cookie Christmas Tourney

2016-12-19T12:58:07+02:00

Gorton wrote:
The TD is supposed to be an impartial judge. I'm pretty certain of our ability to do that.

indeed

Statistics: Posted by Stups — 19 Dec 2016, 12:58

Re: Crumbly Cookie Christmas Tourney

2016-12-17T18:51:46+02:00

CookieNoob wrote:
lets just make an example:
player A has a ladder rating of 1400 and a global rating of 0
player B has a ladder rating of 2000 and a global rating of 1800
based on what exactly should they be treated differently?

"common sence" is not a valid argument to make exceptions for some players but not for others. Where do you want to draw the line? At 200 pts difference? what happens to the one with 201 pts difference? should we count the higher rating for the one with 250 pts difference?

I would like to keep it fair for everyone and that means no exceptions. If the tourney director arbitrarily sets the rating for the participants based on how he "feels" the players play, hosting tourneys with rating restriction makes no sence. Tourney director could just say I want this player but not this one.

I for example have no idea about the strength of most 1k rated players because I rarely play in that rating range. How am I supposed to know if the 1k ladder and 700 global should count as 700 or 1k rated player?? There is no way to do so.

plz keep this discussion out of the tourney topic from now on

There is no drawing the line - that's what you are doing with this rule. No, what I suggest is thinking about players.
I'd like to think a lot of us know quite a bit about the players of this game, since there's not *that* many in reality.
Perhaps I have just experienced a wider range of play due to watching replays and such, but...

Moving on,
Someone can be very good at 1v1 and terrible at teamgames. This also applies in reverse but isn't important to our discussion.
In the end, while being good at 1v1 generally means you can perform reasonably in teamgames, it doesn't mean you are good at them.
We can't take a rating solely made for 1v1 games and apply it to a 3v3. Global rating isn't that good for this either, as it can comprise many things. It's still better than this.

There is no meaningful correlation between 1v1 and global rating is something that needs saying. If you look at the list of people in 1v1 leaderboard and global, you'll see their ratings , compared to everyone else there, don't follow any sort of pattern.
What we can use a 1v1 rating for : Checking if people are very far below in their global to 1v1.

All this rule does is make a teamgame tourney harder on people who have a higher 1v1 rating (which has no pattern or relation to their global, essentially teamgame rating) while not affecting the "98%" the percentage of people unaffected.

The TD is supposed to be an impartial judge. I'm pretty certain of our ability to do that.

Statistics: Posted by Gorton — 17 Dec 2016, 18:51

Re: Rating limit for tourneys

2016-12-17T14:28:28+02:00

I moved the discussion into a separate thread.

Statistics: Posted by Stups — 17 Dec 2016, 14:28

Re: Crumbly Cookie Christmas Tourney

2016-12-17T13:39:14+02:00

Totally agree with Cookies approach to the rating subject

Sounds like this is going to be a fun tourney. And this bot you have is very interesting. I would like to learn more about it if possible.

Cookie you should advertise this more within FAF ask the moderators to place annoucements in main chat every 2 hours?

Statistics: Posted by SYSTEM_FAILURE — 17 Dec 2016, 13:39

Re: Crumbly Cookie Christmas Tourney

2016-12-17T12:49:48+02:00

lets just make an example:
player A has a ladder rating of 1400 and a global rating of 0
player B has a ladder rating of 2000 and a global rating of 1800
based on what exactly should they be treated differently?

"common sence" is not a valid argument to make exceptions for some players but not for others. Where do you want to draw the line? At 200 pts difference? what happens to the one with 201 pts difference? should we count the higher rating for the one with 250 pts difference?

I would like to keep it fair for everyone and that means no exceptions. If the tourney director arbitrarily sets the rating for the participants based on how he "feels" the players play, hosting tourneys with rating restriction makes no sence. Tourney director could just say I want this player but not this one.

I for example have no idea about the strength of most 1k rated players because I rarely play in that rating range. How am I supposed to know if the 1k ladder and 700 global should count as 700 or 1k rated player?? There is no way to do so.

plz keep this discussion out of the tourney topic from now on

Statistics: Posted by CookieNoob — 17 Dec 2016, 12:49

Re: Crumbly Cookie Christmas Tourney

2016-12-16T23:10:24+02:00

CookieNoob wrote:
In my oppinion the case where someone only plays ladder and has a significantly higher rating there (very few players) should be taken into account and thus the higher of both ratings is used. The tourney bot does that automatically so no additional action involved. The tourney bot will automatically split the teams that are too high rated, so it's not a problem when you signed up and your team is too high rated.

It is the christmas tourney, so I wanted a rating limit that allows as many players that could be interested to play to actually participate. That also means that overpowered teams should be avoided.

Besides: for 98% of the players this is irrelevant cause their ladder rating is lower than the global.

Significantly higher rating is of the case where you have someone like 1200 rated and 1800 1v1. It's very important to note that 1800 rating in 1v1 does not automatically mean you can compete with 1800 global in a teamgame, and this is a 3v3 tourney

Besides: for 98% of the players this is irrelevant cause their ladder rating is lower than the global

So *98%* of the time it's completely pointless, and screws over a couple of people unfairly. Seems like a pointless rule to me.

No issues with rating limit. Issues with using 1v1 rating in an non-1v1 scenario

Statistics: Posted by Gorton — 16 Dec 2016, 23:10

Re: Crumbly Cookie Christmas Tourney

2016-12-16T22:39:52+02:00

In my oppinion the case where someone only plays ladder and has a significantly higher rating there (very few players) should be taken into account and thus the higher of both ratings is used. The tourney bot does that automatically so no additional action involved. The tourney bot will automatically split the teams that are too high rated, so it's not a problem when you signed up and your team is too high rated.

It is the christmas tourney, so I wanted a rating limit that allows as many players that could be interested to play to actually participate. That also means that overpowered teams should be avoided.

Besides: for 98% of the players this is irrelevant cause their ladder rating is lower than the global.

Statistics: Posted by CookieNoob — 16 Dec 2016, 22:39

Re: Crumbly Cookie Christmas Tourney

2016-12-16T22:15:13+02:00

Morax wrote:
Not everyone has the same though-process, though, Gorton, so it is really difficult to know how one will relay information and is good for you and I and others to make these comments.

My post was originally done to help Cookie so he experiences less complaints and stress inflicted the day-of the tourney. I noticed that Creh, AchievedJaguar8, and their 3rd mate were well over the 4200 limit by actual number. They posted "our team is 4200 or less by these numbers," which were the rounded versions.

With this knowledge it will only help people decide who to play with and lead to a smoother experience for all to enjoy the games.

I don't see how this is relevant to my post.

Statistics: Posted by Gorton — 16 Dec 2016, 22:15

Re: Crumbly Cookie Christmas Tourney

2016-12-16T21:56:31+02:00

Not everyone has the same though-process, though, Gorton, so it is really difficult to know how one will relay information and is good for you and I and others to make these comments.

My post was originally done to help Cookie so he experiences less complaints and stress inflicted the day-of the tourney. I noticed that Creh, AchievedJaguar8, and their 3rd mate were well over the 4200 limit by actual number. They posted "our team is 4200 or less by these numbers," which were the rounded versions.

With this knowledge it will only help people decide who to play with and lead to a smoother experience for all to enjoy the games.

Statistics: Posted by Morax — 16 Dec 2016, 21:56

Re: Crumbly Cookie Christmas Tourney

2016-12-16T21:53:44+02:00

Gorton wrote:
think all our TDs have.

Statistics: Posted by Morax — 16 Dec 2016, 21:53

Re: Crumbly Cookie Christmas Tourney

2016-12-16T21:03:37+02:00

I'm not playing, but I am a good example.

I have 1689 global rating currently.
My 1v1 is almost 200 points clear, at 1880 or something.

Anyone who knows me knows for sure that in this 3v3 setting, I am not competing at anything like a 1900 level. However, Because of the 1v1 rating, I can only have teammates with a combined 2300~ or 1150 average rating.
It's a significant difference imo.

It would make more sense to use the teamgames rating for a teamgame, and use the 1v1 to check the authenticity of that rating - A 1200 entering with 1800 1v1 rating for example would be found odd.

In the end, it would be up to the TD to exercise some reasonable judgement and common sense, something that I think all our TDs have.

Statistics: Posted by Gorton — 16 Dec 2016, 21:03

Re: Crumbly Cookie Christmas Tourney

2016-12-16T20:53:22+02:00

I personally don't see an issue with it, since as you say it doesn't really matter for most people. (or what do you mean with pointless?)

But what do you mean with: compare them to see?
If you're going to look at the 1v1 rating anyway, Then you Should just mention it from the start so nobody gets surprised by having his team get disbanded due to too high 1v1 ratings...

Statistics: Posted by KeyBlue — 16 Dec 2016, 20:53

Rating limit for tourneys

2016-12-16T19:58:45+02:00

I would just like to mention that I think the 1v1 rating restriction for a 3v3 tournament is a little silly. I can see that you'd want to prevent someone with a drastically lower global rating than 1v1 rating playing, but for the most part it's a bit pointless. Perhaps better to just use global, and take their 1v1 ratings and compare them to see?

Statistics: Posted by Gorton — 16 Dec 2016, 19:58