Amidon37 wrote:

*Lots of caveats to that also since earning CP's is much easier on boards that have larger numbers of player per game.  

Is this true?  Is it just because playing 1 12-player game is equivalent to playing 3 4-player games, or is there more to it?

> Having rank based on CP's is encouraging players to branch out and try all the user generated content.

I completely agree with your argument ratsy, but I think some of that is my self-interest as a map maker. 

I think if we were to update the CP calculation, that might assuage some of the ranking concerns too.  To take toto as an example (since he offered himself up ;) ).  #1 on WGWF is getting him 20 CP right now, with proposal #2 (my preferred one) on the wiki page, he'd instead get 60 CP.  For antastic, he'd probably get 35-40 instead of 12, with similar effects on War and Europe 1560 and several others.  Overall, I bet he'd gain about 150 CP, going from 101 to 251 CP.  My guess is once everyone else 's scores get adjusted, he'd be about #10 in the CP rankings.   That seems reasonableish to me.

There are three boards on this site that are near-clones of traditional Risk, so players that specialize in traditional risk also have 3x the chance to earn CP.  If toto wanted, I'm sure he could move up in Global Warfare ranking.

Toto wrote:

Babbalouie wrote:

It should be based on global ranking. Championship points mean little to me. Players build up points after playing only 2 games on certain boards. You can become a General by playing boards that are not popular and are seldom played. So in other words, ranks don't mean a thing to me. I came up with a 4 rank system (5 as per Cona Chris), and it seemed popular with most players, but I am still waiting for a reply from Tom. 

+1

Despite I will appear to be self-centred or narcissistic, I will give my case as an example : 

I am number 1 at the tournament ranking, number 1 at Wargear Warfare, number 2 at the global ranking, and I am only a major general. Is that fair ?

+1

Toto, with the 4 rank system you would be a Tourney 5 Star General, which you deserve.

Personally, I would rather be a private with no CP's than play several uninteresting boards just to try to move up in rank doing something that simply does not interest me.

With the 4 rank system everyone can pursue whatever area interests them personally. It would lead to much more competition. There would be 4 5 Star Generals. The majority of the players on this site care very little about CP's and those that do, they would still have a CP 5 Star General. This would be good for everyone and not just the CP boys who feel that everyone should play everything.

There does seem to be some consensus where the multiple general system is concerned - but the underling point system still needs an update/overhaul.

Ozyman wrote:

SquintGnome wrote:

-1, -1, -1

I do not have time now to put my thoughts together coherently on this topic.  But I can say for now that I think the popular boards are harder to be top ranked on because there are so many good players competing.  It will probably be very useful to separate the concepts of 'gaining points' from 'being top ranked on'.

Popular games will be initially easier to get points on because your games fill quickly.  You are gaining points faster because you have opponents, not because you opponents are less skilled necessarily.  But after you climb a bit on these boards it becomes comparatively more difficult to get into the top ten.

I agree with all that.    It's definitely harder to get to #10 or #1 on WGWF then any other board.  But I found it much easier to get to 1500 on WGWF than any other board.  

I'd love to hear your thoughts on this proposal:

http://www.wargear.net/wiki/doku.php?id=general_features:scaling_championship_points_based_on_games_played

which I think is a good compromise.  As you are reasonably highly ranked in both GR & CP, I think you have valuable insight into how these things compare.  I'd especially like your comments on the last table on that page & what you think of the current entries in the table.

Hi Ozy,

I agree with the concept you lay out.  My gut feeling is that you can use a board with about 100 plays as the baseline 20 CPs.  The toughest board should be worth 3-4 times more than that I think which is in line with what one of your tables suggest. 

Another option is to set a comparative ratio base on number of players instead of number of games.  I thought of this when I noticed on our 'Board Ranking' individual tab, that the rankning is given relative to the total number of people who have played.  For example, for WGWF I am ranked 3 out 8995 players, and for Battle for New York 2 out of 937.  I am guessing that many boards have less that 500 players....not sure.  But this might give more precision than games played.

SquintGnome wrote:

Another option is to set a comparative ratio base on number of players instead of number of games.  I thought of this when I noticed on our 'Board Ranking' individual tab, that the rankning is given relative to the total number of people who have played.  For example, for WGWF I am ranked 3 out 8995 players, and for Battle for New York 2 out of 937.  I am guessing that many boards have less that 500 players....not sure.  But this might give more precision than games played.

This is an excellent point; at first glance and at the risk of downgrading boards that get a lot of plays - but with smaller cult followings (some of mine definitely fall into that category), I'm tempted to support # of Players as a better metric than # of Games.

Can the math types combine the two and have some metric that factors in both number of players/number of plays equally? 

Ozyman wrote:

Amidon37 wrote:

*Lots of caveats to that also since earning CP's is much easier on boards that have larger numbers of player per game.  

Is this true?  Is it just because playing 1 12-player game is equivalent to playing 3 4-player games, or is there more to it?

First of all, got to count yourself in the mix: 1 12-player game means 11 opponents.  It would be more like 2 5-player games and a 4-player game (11 opponents).  The points won't work out the same, but neither will your experience: by the time you play that third game, you might actually have learned the board and be better at it.

I specifically went and started games on large, rarely-played boards with 16 players, on the theory that this would be easier to gain CP's than by playing lots of games.  My experience is that this plan works great when it works - but mostly this strategy doesn't work, and it doesn't argue for either CPs or global ranking.  Here's why:

1) To succeed, I have to beat 15 other players without getting eliminated even once - on a board many of them have played before and I have never tried. (Another way of thinking about it: 16 people will try for "easy" points.  One will succeed, 15 will fail.)

2) I do get a big jump in CP's when I win.  But I get an equally large jump in global ranking.

For example, I played and won exactly one 16-player game on Pipeline. I went from 1000 to 1289 in a single game, and now have 6 CP's on that board.  At the same time, my global ranking went up by nearly as much: from 1906 to 2077. 

My whole plan with playing this kind of game was to increase my CP's rapidly - but I ended up moving up in the global points even more quickly.  Go figure.

BTdubs wrote:

Ozyman wrote:

Amidon37 wrote:

*Lots of caveats to that also since earning CP's is much easier on boards that have larger numbers of player per game.  

Is this true?  Is it just because playing 1 12-player game is equivalent to playing 3 4-player games, or is there more to it?

First of all, got to count yourself in the mix: 1 12-player game means 11 opponents.  It would be more like 2 5-player games and a 4-player game (11 opponents).  The points won't work out the same, but neither will your experience: by the time you play that third game, you might actually have learned the board and be better at it.

I specifically went and started games on large, rarely-played boards with 16 players, on the theory that this would be easier to gain CP's than by playing lots of games.  My experience is that this plan works great when it works - but mostly this strategy doesn't work, and it doesn't argue for either CPs or global ranking.  Here's why:

1) To succeed, I have to beat 15 other players without getting eliminated even once - on a board many of them have played before and I have never tried. (Another way of thinking about it: 16 people will try for "easy" points.  One will succeed, 15 will fail.)

2) I do get a big jump in CP's when I win.  But I get an equally large jump in global ranking.

For example, I played and won exactly one 16-player game on Pipeline. I went from 1000 to 1289 in a single game, and now have 6 CP's on that board.  At the same time, my global ranking went up by nearly as much: from 1906 to 2077. 

My whole plan with playing this kind of game was to increase my CP's rapidly - but I ended up moving up in the global points even more quickly.  Go figure.

Except if you do that say 12 times and win 3 you might get CP's for three boards whereas your GR would go up 150 for those three games and down 50 for the other 9 and you actually are just breaking even with your GR.  Your one game sample size is skewing your results to give you a rosier picture of your GR then the actual result using that strategy over a larger sample size. Go figure ;-)

Just playing with the math here..

BTdubs is expected to win 13 out of 100 16-player games with his 70 H-Rating. 

I don't know what the "average" GR is for active players, but let's assume the average opponent has a GR of 1300 and give BTdubs his current ~2200 rating.  Let's further assume that when he loses, he tends to lose to the better opponents at ~1600.

That's 87 losses * (2200/1600) * 20 for 2400 points lost, and (13 * 15) wins * (1300/2200) * 20 = 2300 points gained.  Net 100 points lost.    Looks like he's close to equilibrium.

Of course, I pulled those numbers out of my butt.

Your numbers were much less butt pulled then mine.  But, your math and my gut both produced pretty similar results!

Wow.  I certainly stepped into something here!   Didn't mean to cause a ruckus!  :)  I personally think global ranking is a better score to use for rank than championship points, but I sense we have a hung jury on that matter.

....and for those of you who think I'm a creature of habit on WGWF, note that the vast majority of my games are real-time 2 players (a bit bored at work), thus they add up quickly!  Also, I do have >300 games spread across 53 other boards.  I'm always trying to to spread my good cheer, as I slowly attempt to hone my skills on other boards.