Percentage Won Statistic Suggestion

Ask questions, answer questions

Moderators: Global Moderators, Discussions Team

Forum rules
Please read the Community Guidelines before posting.

Percentage Won Statistic Suggestion

Postby General Quigley on Sun Sep 01, 2013 2:43 pm

In my opinion the games won percentage statistic as it is currently computed is a useless statistic. If all you play are two player games, a winning percentage of 40% is awful. If you play only 8 or more players per game (as I mostly do), then 20% would be an outstanding statistic. The usefulness of this statistic will decline even more as Conquer Club moves to allowing 12-player games.

The games won percentage statistic could easily be changed so that it provides a more worthwhile, meaningful, and competetive statistic. The new game won percentage statistic I would like to propose would be calculated as follows:

(Games won / games played) X number of players in game on average

Thus, let's say a person plays two 2-player games and wins one and loses the other. The result is (1/2) X 2 = 1.00. Thus, 1.00 is what an average person will have as a games won percentage statistic under this computation. Now let's say you play in eight 8-player games and actually win two - an outstanding result. Using my proposed formula (2/8) X 8 = 2.0. The player with the 2.0 statistic thus deservedly has a higher games won percentage statistic even though he only won 1/4 of his games instead of half, as the first person did.

Conclusion: My proposed more meaningful games won percentage statistic would tell me a whole lot more about my opponent's prowess than the current one does. Frankly, it may even be a better more objective measure of CC strength than the current rating system.
User avatar
Sergeant 1st Class General Quigley
 
Posts: 87
Joined: Tue Apr 29, 2008 5:00 pm
Location: South Carolina
Medals: 38
Standard Achievement (4) Doubles Achievement (2) Triples Achievement (1) Quadruples Achievement (2) Terminator Achievement (1)
Manual Troops Achievement (3) Freestyle Achievement (4) Nuclear Spoils Achievement (4) Fog of War Achievement (4) Trench Warfare Achievement (2)
Speed Achievement (1) Teammate Achievement (2) Cross-Map Achievement (3) Beta Map Achievement (1) Ratings Achievement (4)

Re: Percentage Won Statistic Suggestion

Postby betiko on Sun Sep 01, 2013 5:59 pm

Completely agree with you but you should post it in suggestions
Image
User avatar
Brigadier betiko
SoC Training Adviser
 
Posts: 4554
Joined: Fri Feb 25, 2011 3:05 pm
Location: Paris
Medals: 149
Conquer Cup Gold Achievement (1) Standard Achievement (4) Doubles Achievement (3) Triples Achievement (3) Quadruples Achievement (3)
Terminator Achievement (2) Assassin Achievement (3) Manual Troops Achievement (3) Freestyle Achievement (3) Polymorphic Achievement (2)
Nuclear Spoils Achievement (3) Fog of War Achievement (4) Trench Warfare Achievement (3) Speed Achievement (4) Teammate Achievement (2)
Random Map Achievement (3) Cross-Map Achievement (4) Beta Map Achievement (2) Battle Royale Achievement (2) Bot Achievement (3)
Ratings Achievement (4) Tournament Achievement (14) General Achievement (30) Clan Achievement (7) Training Achievement (2)

Re: Percentage Won Statistic Suggestion

Postby macbone on Mon Sep 02, 2013 3:34 am

General, you'd also have to factor in team games. I'd suggest computing each team as a single player, so an 8-player doubs game would count as 4 players.
User avatar
Major macbone
SoC Training Instructor
SoC Training Instructor
 
Posts: 5270
Joined: Wed Jun 03, 2009 7:12 pm
Location: Looking out over Victoria Harbor, thinking about Middle Earth and how good my wife looks
Medals: 75
Standard Achievement (3) Doubles Achievement (3) Triples Achievement (3) Quadruples Achievement (3) Terminator Achievement (3)
Assassin Achievement (1) Manual Troops Achievement (1) Freestyle Achievement (1) Nuclear Spoils Achievement (2) Fog of War Achievement (3)
Trench Warfare Achievement (2) Speed Achievement (1) Teammate Achievement (2) Random Map Achievement (3) Cross-Map Achievement (3)
Beta Map Achievement (1) Ratings Achievement (4) Tournament Achievement (8) General Achievement (8) Clan Achievement (9)
Training Achievement (2) Tournament Contribution (2) General Contribution (7)

Re: Percentage Won Statistic Suggestion

Postby Donelladan on Mon Sep 02, 2013 6:58 am

I agree with you that the current winning percentage doesnt give a real idea of how good a player is.
But I am afraid your formula wont be really clear and many people may not understand the meaning of this calculation.

How about, for each number of opponent team we calculate the % of victory, then substract the "expected percentage" - dunno how to say that exactly. I'll explain

Let's say you play 1vs1 game. so expected is 50% winning. If you won 56% of your game, we do 56% - 50% = +6%. That one will be shown.

Then if you played several type of game, as everyone, we sum for each case :
56% win in 2 players game -> +6%
24% win in 3 players game -> 24% - 33% = -9%
22% win in 4 players game -> 22% - 25% = -3%
30% win in 8 players game -> 30% - 12.5% = +17,5%

This player would have a +11.5% winning ratio.

With this method you know directly wether it is negative or positive if the player is really skiled, and each type of game give a fair contribution.

Of course I agree with macbone, so rather than number of players, we should consider number of teams in the game.
Click image to enlarge.
image
User avatar
Brigadier Donelladan
 
Posts: 565
Joined: Tue Mar 03, 2009 8:48 am
Medals: 73
Standard Achievement (4) Doubles Achievement (4) Triples Achievement (2) Quadruples Achievement (3) Terminator Achievement (2)
Assassin Achievement (2) Manual Troops Achievement (3) Freestyle Achievement (3) Polymorphic Achievement (1) Nuclear Spoils Achievement (2)
Fog of War Achievement (4) Trench Warfare Achievement (3) Speed Achievement (3) Teammate Achievement (2) Random Map Achievement (2)
Cross-Map Achievement (3) Beta Map Achievement (2) Ratings Achievement (3) Tournament Achievement (14) General Achievement (2)
Clan Achievement (9)

Re: Percentage Won Statistic Suggestion

Postby betiko on Mon Sep 02, 2013 4:49 pm

Donelladan wrote:I agree with you that the current winning percentage doesnt give a real idea of how good a player is.
But I am afraid your formula wont be really clear and many people may not understand the meaning of this calculation.

How about, for each number of opponent team we calculate the % of victory, then substract the "expected percentage" - dunno how to say that exactly. I'll explain

Let's say you play 1vs1 game. so expected is 50% winning. If you won 56% of your game, we do 56% - 50% = +6%. That one will be shown.

Then if you played several type of game, as everyone, we sum for each case :
56% win in 2 players game -> +6%
24% win in 3 players game -> 24% - 33% = -9%
22% win in 4 players game -> 22% - 25% = -3%
30% win in 8 players game -> 30% - 12.5% = +17,5%

This player would have a +11.5% winning ratio.

With this method you know directly wether it is negative or positive if the player is really skiled, and each type of game give a fair contribution.

Of course I agree with macbone, so rather than number of players, we should consider number of teams in the game.


Assuming that in your example the player has played the same amount of games of each of those 4 cathegories. But yeah, I ve been telling this for years, this win ratio on our walls is a joke, a mix of pears, apples, tomatoes and artichokes.
Personally, when i look at someone s win ratio it just tells me if he plays mostly 1v1 (team or singles) or multi singles.
Image
User avatar
Brigadier betiko
SoC Training Adviser
 
Posts: 4554
Joined: Fri Feb 25, 2011 3:05 pm
Location: Paris
Medals: 149
Conquer Cup Gold Achievement (1) Standard Achievement (4) Doubles Achievement (3) Triples Achievement (3) Quadruples Achievement (3)
Terminator Achievement (2) Assassin Achievement (3) Manual Troops Achievement (3) Freestyle Achievement (3) Polymorphic Achievement (2)
Nuclear Spoils Achievement (3) Fog of War Achievement (4) Trench Warfare Achievement (3) Speed Achievement (4) Teammate Achievement (2)
Random Map Achievement (3) Cross-Map Achievement (4) Beta Map Achievement (2) Battle Royale Achievement (2) Bot Achievement (3)
Ratings Achievement (4) Tournament Achievement (14) General Achievement (30) Clan Achievement (7) Training Achievement (2)

Re: Percentage Won Statistic Suggestion

Postby agentcom on Fri Sep 06, 2013 4:42 am

There is a suggestion to this effect. I agree that win % is pretty meaningless as is and wouldn't be that hard to change. Basically, it's somewhat of a low priority fix right now (have you seen the cool stuff coming out of CC?!). But if you want to track it down and lend your support to the suggestion, you could bump it back up onto main suggestions page :)
Image
User avatar
Colonel agentcom
 
Posts: 3710
Joined: Tue Nov 09, 2010 8:50 pm
Medals: 69
Standard Achievement (4) Doubles Achievement (3) Triples Achievement (2) Quadruples Achievement (3) Terminator Achievement (1)
Assassin Achievement (2) Manual Troops Achievement (2) Freestyle Achievement (2) Nuclear Spoils Achievement (2) Fog of War Achievement (4)
Trench Warfare Achievement (2) Speed Achievement (3) Teammate Achievement (2) Random Map Achievement (2) Cross-Map Achievement (3)
Battle Royale Achievement (1) Ratings Achievement (2) Tournament Achievement (13) General Achievement (6) Clan Achievement (8)
General Contribution (2)

Re: Percentage Won Statistic Suggestion

Postby Mr Changsha on Fri Sep 06, 2013 5:03 am

General Quigley wrote:In my opinion the games won percentage statistic as it is currently computed is a useless statistic. If all you play are two player games, a winning percentage of 40% is awful. If you play only 8 or more players per game (as I mostly do), then 20% would be an outstanding statistic. The usefulness of this statistic will decline even more as Conquer Club moves to allowing 12-player games.

The games won percentage statistic could easily be changed so that it provides a more worthwhile, meaningful, and competetive statistic. The new game won percentage statistic I would like to propose would be calculated as follows:

(Games won / games played) X number of players in game on average

Thus, let's say a person plays two 2-player games and wins one and loses the other. The result is (1/2) X 2 = 1.00. Thus, 1.00 is what an average person will have as a games won percentage statistic under this computation. Now let's say you play in eight 8-player games and actually win two - an outstanding result. Using my proposed formula (2/8) X 8 = 2.0. The player with the 2.0 statistic thus deservedly has a higher games won percentage statistic even though he only won 1/4 of his games instead of half, as the first person did.

Conclusion: My proposed more meaningful games won percentage statistic would tell me a whole lot more about my opponent's prowess than the current one does. Frankly, it may even be a better more objective measure of CC strength than the current rating system.


20% is not 'outstanding' for 8 player standard games...I would suggest it is 'slightly above average' and would typically result in a lieutenant to captain rank if played with flat or no cards against random opposition. Played against extremely good opposition on escalating, 20% might just about be very good, but would still not qualify as outstanding by any means.

I would hazard that for 8 man public flats 45-50% would be 'outstanding'.
For 8 man private high rank escalating 30-35% would be 'outstanding'.

Always here to educate,

Mr C
Image
User avatar
Colonel Mr Changsha
 
Posts: 1615
Joined: Sat Mar 03, 2007 1:42 am
Medals: 14
Standard Achievement (2) Doubles Achievement (2) Triples Achievement (2) Quadruples Achievement (1) Fog of War Achievement (1)
Speed Achievement (1) Teammate Achievement (1) Ratings Achievement (1) Clan Achievement (3)

Re: Percentage Won Statistic Suggestion

Postby macbone on Fri Sep 06, 2013 6:27 am

MrChangsha, you and I live in different worlds, man. 50% on 8-player games isn't simply outstanding. I'd dare say it would be god-like. =)

If one is playing 8-player escalating games versus extremely good competition, then winning 25% of those games would be very impressive indeed.
User avatar
Major macbone
SoC Training Instructor
SoC Training Instructor
 
Posts: 5270
Joined: Wed Jun 03, 2009 7:12 pm
Location: Looking out over Victoria Harbor, thinking about Middle Earth and how good my wife looks
Medals: 75
Standard Achievement (3) Doubles Achievement (3) Triples Achievement (3) Quadruples Achievement (3) Terminator Achievement (3)
Assassin Achievement (1) Manual Troops Achievement (1) Freestyle Achievement (1) Nuclear Spoils Achievement (2) Fog of War Achievement (3)
Trench Warfare Achievement (2) Speed Achievement (1) Teammate Achievement (2) Random Map Achievement (3) Cross-Map Achievement (3)
Beta Map Achievement (1) Ratings Achievement (4) Tournament Achievement (8) General Achievement (8) Clan Achievement (9)
Training Achievement (2) Tournament Contribution (2) General Contribution (7)

Re: Percentage Won Statistic Suggestion

Postby Mr Changsha on Fri Sep 06, 2013 7:07 am

macbone wrote:MrChangsha, you and I live in different worlds, man. 50% on 8-player games isn't simply outstanding. I'd dare say it would be god-like. =)

If one is playing 8-player escalating games versus extremely good competition, then winning 25% of those games would be very impressive indeed.


That's my point...playing your typical public 8 man striper-fest either flat or no cards then 50% would be outstanding.

Playing those settings against proper organised opposition then 35% would be equally outstanding.

Escalating against high ranks (a game I am useless at) then 25%-30% would also be outstanding.

On my standard settings (flat no card/ 6-8 man) colonels (as no one really gets higher without general farming antics) would be 40-45%, majors 30-40%, captains 25-30%. The reason for this is that the general quality of opposition will be lower and it is easier to win those games. Now I'm not saying everyone could do it - the level of patience required is frankly to great for me to bother with it these days - but it IS possible to have a much higher win rate on those settings (flat no cards) than escalating. However, escalating can build a higher score in the end even with that 25% win rate as one can continue to play high quality opposition. Once one is much over 2500 my standard settings become ball-breaking annoying with 50 point wins and 50 point losses for 6 to 8 man games.

On the settings described (though I included some smaller games) +1384 42\99 42% (sunny). However, I believe my 8 man percentage is higher so I would suggest my point holds. Over the years I have been aware of quite a number of players with similar or better records so, I maintain for my settings that 50% of 8 man wins would be 'outstanding' while for escalating it would be somewhere around 35%. But in neither case is it 20% which would be either 'decidedly average' (my settings) or 'decent to good' escalating.
Image
User avatar
Colonel Mr Changsha
 
Posts: 1615
Joined: Sat Mar 03, 2007 1:42 am
Medals: 14
Standard Achievement (2) Doubles Achievement (2) Triples Achievement (2) Quadruples Achievement (1) Fog of War Achievement (1)
Speed Achievement (1) Teammate Achievement (1) Ratings Achievement (1) Clan Achievement (3)

Re: Percentage Won Statistic Suggestion

Postby betiko on Fri Sep 06, 2013 7:19 am

Mr Changsha wrote:
macbone wrote:MrChangsha, you and I live in different worlds, man. 50% on 8-player games isn't simply outstanding. I'd dare say it would be god-like. =)

If one is playing 8-player escalating games versus extremely good competition, then winning 25% of those games would be very impressive indeed.


That's my point...playing your typical public 8 man striper-fest either flat or no cards then 50% would be outstanding.

Playing those settings against proper organised opposition then 35% would be equally outstanding.

Escalating against high ranks (a game I am useless at) then 25%-30% would also be outstanding.

On my standard settings (flat no card/ 6-8 man) colonels (as no one really gets higher without general farming antics) would be 40-45%, majors 30-40%, captains 25-30%. The reason for this is that the general quality of opposition will be lower and it is easier to win those games. Now I'm not saying everyone could do it - the level of patience required is frankly to great for me to bother with it these days - but it IS possible to have a much higher win rate on those settings (flat no cards) than escalating. However, escalating can build a higher score in the end even with that 25% win rate as one can continue to play high quality opposition. Once one is much over 2500 my standard settings become ball-breaking annoying with 50 point wins and 50 point losses for 6 to 8 man games.

On the settings described (though I included some smaller games) +1384 42\99 42% (sunny). However, I believe my 8 man percentage is higher so I would suggest my point holds. Over the years I have been aware of quite a number of players with similar or better records so, I maintain for my settings that 50% of 8 man wins would be 'outstanding' while for escalating it would be somewhere around 35%. But in neither case is it 20% which would be either 'decidedly average' (my settings) or 'decent to good' escalating.


You are not making much sense here. On a given 8 player game, everyone has 12,5% chances. if captains won 25-30% thery wouldn't be captains. If you're around 20% you're definitely making nice profits out of these settings. Of course it depends on the competition, but playing vs random yahoos is equally difficult; you got to anticipate irrational moves and adapt your gamestyle to who you're facing.
Image
User avatar
Brigadier betiko
SoC Training Adviser
 
Posts: 4554
Joined: Fri Feb 25, 2011 3:05 pm
Location: Paris
Medals: 149
Conquer Cup Gold Achievement (1) Standard Achievement (4) Doubles Achievement (3) Triples Achievement (3) Quadruples Achievement (3)
Terminator Achievement (2) Assassin Achievement (3) Manual Troops Achievement (3) Freestyle Achievement (3) Polymorphic Achievement (2)
Nuclear Spoils Achievement (3) Fog of War Achievement (4) Trench Warfare Achievement (3) Speed Achievement (4) Teammate Achievement (2)
Random Map Achievement (3) Cross-Map Achievement (4) Beta Map Achievement (2) Battle Royale Achievement (2) Bot Achievement (3)
Ratings Achievement (4) Tournament Achievement (14) General Achievement (30) Clan Achievement (7) Training Achievement (2)

Re: Percentage Won Statistic Suggestion

Postby Mr Changsha on Fri Sep 06, 2013 12:45 pm

betiko wrote:
Mr Changsha wrote:
macbone wrote:MrChangsha, you and I live in different worlds, man. 50% on 8-player games isn't simply outstanding. I'd dare say it would be god-like. =)

If one is playing 8-player escalating games versus extremely good competition, then winning 25% of those games would be very impressive indeed.


That's my point...playing your typical public 8 man striper-fest either flat or no cards then 50% would be outstanding.

Playing those settings against proper organised opposition then 35% would be equally outstanding.

Escalating against high ranks (a game I am useless at) then 25%-30% would also be outstanding.

On my standard settings (flat no card/ 6-8 man) colonels (as no one really gets higher without general farming antics) would be 40-45%, majors 30-40%, captains 25-30%. The reason for this is that the general quality of opposition will be lower and it is easier to win those games. Now I'm not saying everyone could do it - the level of patience required is frankly to great for me to bother with it these days - but it IS possible to have a much higher win rate on those settings (flat no cards) than escalating. However, escalating can build a higher score in the end even with that 25% win rate as one can continue to play high quality opposition. Once one is much over 2500 my standard settings become ball-breaking annoying with 50 point wins and 50 point losses for 6 to 8 man games.

On the settings described (though I included some smaller games) +1384 42\99 42% (sunny). However, I believe my 8 man percentage is higher so I would suggest my point holds. Over the years I have been aware of quite a number of players with similar or better records so, I maintain for my settings that 50% of 8 man wins would be 'outstanding' while for escalating it would be somewhere around 35%. But in neither case is it 20% which would be either 'decidedly average' (my settings) or 'decent to good' escalating.


You are not making much sense here. On a given 8 player game, everyone has 12,5% chances. if captains won 25-30% thery wouldn't be captains. If you're around 20% you're definitely making nice profits out of these settings. Of course it depends on the competition, but playing vs random yahoos is equally difficult; you got to anticipate irrational moves and adapt your gamestyle to who you're facing.


Generally on the settings under discussion, captains have around a 25% win rate around the 6-8 player mark. Also while winning 1 in 5 games (8 man) would make points +60 every 5 games, 120 for 10.. 100 games = 1100 ( drop a bit as differentials change) say 1000.

I see your point. What I didn't take into account was losing games on other settings which is why I often see captains with 25% win rates on the settings in question.

My only excuse is the hash.. ;)
Image
User avatar
Colonel Mr Changsha
 
Posts: 1615
Joined: Sat Mar 03, 2007 1:42 am
Medals: 14
Standard Achievement (2) Doubles Achievement (2) Triples Achievement (2) Quadruples Achievement (1) Fog of War Achievement (1)
Speed Achievement (1) Teammate Achievement (1) Ratings Achievement (1) Clan Achievement (3)

Re: Percentage Won Statistic Suggestion

Postby Shannon Apple on Fri Sep 06, 2013 2:25 pm

I can agree with this.

like my 1v1, is something like 51%
Then my single player esc is somewhere like 18% (6-8 player)
team games are somewhat higher. I still have an okay win percentage overall at 41%, so not disgruntled about that.
However, I've seen decent players with a 30% win percentage and it was mainly because they play tons of escalators and such.
Image

18:19:01 ‹#1_stunna› Pixar is the only noob that would leave Kort
18:20:39 ‹*Pixar*› thats a new sig!
18:20:51 ‹*Pixar*› "only player to leave kort"
User avatar
Captain Shannon Apple
 
Posts: 724
Joined: Wed Jan 14, 2009 8:40 pm
Location: Ireland
Medals: 38
Standard Achievement (3) Doubles Achievement (2) Triples Achievement (2) Quadruples Achievement (2) Terminator Achievement (2)
Manual Troops Achievement (1) Polymorphic Achievement (1) Nuclear Spoils Achievement (1) Fog of War Achievement (2) Trench Warfare Achievement (1)
Speed Achievement (1) Teammate Achievement (1) Random Map Achievement (1) Cross-Map Achievement (3) Beta Map Achievement (1)
Ratings Achievement (3) Tournament Achievement (1) General Achievement (5) Clan Achievement (3) Training Achievement (1)
General Contribution (1)


Return to Questions & Answers

Who is online

Users browsing this forum: No registered users and 1 guest

Login