DeletedUser4507
Guest
GRhin deserted to -X-, so we can expect his propaganda will be working for them now. Obviously his hate towards NOH is still there. 
-
Hello, Guest!
Are you passionate about Tribal Wars 2 and like to help your fellow players?
We currently have open positions for Forum Moderators!
>> Join the Tribal Wars 2 Team now! <<
We would love to hear from you!
DeletedUser2369
Guest
Dude, grow up. I gave a personal prediction, i have nothing to back it up except the impression i got from fighting them on this server. That is not propaganda. It is not hate, merely a personal observation, i could well be wrong.
How about instead of getting antsy every time someone forgets to say "walkingrock is the best player to ever grace the game with his precence, and NOH is full of such holy players as he, we should all bow to their glory" how about you sit down, give your own predictions, and have a constructive conversation.
For the rest of the sentence i totally agree, i like debate, just keep it constructive and consistent (i.e. if you are gonna complain about "lack of evidence" then do it for posts that say you are good as well as the ones that say you are bad) this is not a propaganda post, this is not to toot your own horn or to shutdown anyone who disagrees with you.
How about instead of getting antsy every time someone forgets to say "walkingrock is the best player to ever grace the game with his precence, and NOH is full of such holy players as he, we should all bow to their glory" how about you sit down, give your own predictions, and have a constructive conversation.
Gotta say I love how you are happy with syphax not backing up his statement that NOH is doing good, but you arent happy when its against NOH. I said personal prediction to separate it from the updates, this is me saying what i thing, as i said to walkingrock, its based on my experience this world, and is just a general impression of those i have been fighting, and heard of tribemates fighting.
For the rest of the sentence i totally agree, i like debate, just keep it constructive and consistent (i.e. if you are gonna complain about "lack of evidence" then do it for posts that say you are good as well as the ones that say you are bad) this is not a propaganda post, this is not to toot your own horn or to shutdown anyone who disagrees with you.
Last edited by a staff member:
DeletedUser4507
Guest
@GRhin You are asking for constructive coversation? You, who place feelings above evidence. Dont make me laugh at you.
How is propaganda defined."Propaganda is information that is not objective and is used primarily to influence an audience". Thats exactly what you were doing last few weeks. I dont hate you as a person. I hate what you are doing. I hate propaganda in general and all those who use it. Propaganda is enemy of both statistics and real analysis.
NOH are far from perfect, too small and too many newbies to make some difference, but denying our merrits is faul against data. We are 3rd in off and 1st in deff.I want you to show me how we will be the one who fall first.
How is propaganda defined."Propaganda is information that is not objective and is used primarily to influence an audience". Thats exactly what you were doing last few weeks. I dont hate you as a person. I hate what you are doing. I hate propaganda in general and all those who use it. Propaganda is enemy of both statistics and real analysis.
NOH are far from perfect, too small and too many newbies to make some difference, but denying our merrits is faul against data. We are 3rd in off and 1st in deff.I want you to show me how we will be the one who fall first.
DeletedUser3495
Guest
Lmao, you're a crackup and a half Walkingrock.
To actually have some real discussion though, I think it will be interesting to see how the top 3 scenario shakes out. Seems like -D- won't work with -X- due to Melkor playing in the sandbox with his crayons up his nose while crying rather than actually talking. And then there are some members of WOW who are pissed off with -D- so it's hard to see them working together (especially with the aforementioned toddler with crayons). Then of course -X- and WOW who seem to be working towards the alliance due to both having disagreements with -D-.
Currently seems like WOW has the decision making power here, however, their council seems to be unable to make choices, since they are unaligned in their goals. Personally, I think -D- either needs a new leader or they need to go down.
To actually have some real discussion though, I think it will be interesting to see how the top 3 scenario shakes out. Seems like -D- won't work with -X- due to Melkor playing in the sandbox with his crayons up his nose while crying rather than actually talking. And then there are some members of WOW who are pissed off with -D- so it's hard to see them working together (especially with the aforementioned toddler with crayons). Then of course -X- and WOW who seem to be working towards the alliance due to both having disagreements with -D-.
Currently seems like WOW has the decision making power here, however, their council seems to be unable to make choices, since they are unaligned in their goals. Personally, I think -D- either needs a new leader or they need to go down.
DeletedUser2369
Guest
Response to walking rock:
Nick, are you able to divulge what the grievances between WOW and -D- are?
because despite only fighting a third of OTS (central cluster), while they were also being attacked by Darkness, you were telling me how well you were doing, and still barely taking a few villages. because ive been fighting your founder, and top member for weeks, and despite being each bigger than me, they never entered my province, barely even hurt me, resorting to cats, which also failed to hurt me more than as an irritant.
I have made a prediction of the future of the world, Alliance to kill OTS, then WOW, then Darkness turns on either -X- or NOH, when i wrote that i was under the impression that all three were allied, i was disabused of that in another thread, but my prediction (which i think may be wrong) was that NOh would be next on the chopping block, and I dont have faith in the players I have been fighting that they would stick around when they are losing villages. first few sure, but lose a couple provinces of your big players (abe70 and dimeth being the two i was fighting, and both fairly close to were i anticipated the front lines to be) players will start to dissappear, those two in particular.
Have I explained my own personal opinion enough? or do you need me to justify every single word i say that doesnt call you the best ever? you are happy if i say the same against anyone else.
You have told me that what propaganda means, yet you did not justify how i did it.
Lets go through taht shall we?
"Propaganda is information that is not objective" - im human, as syphax said, im not sure we will ever get an unbiased view, but i like to think I do pretty good. If you disagree with what I say, make your own predictions, add to the conversation instead of taking away from it.
"and is used primarily to influence an audience" - what influence do you think I am trying to make? Im obviously not trying to get players to join OTS, i have left there myself.
Here, I shall do as you do, as you seem to be so greate at this
Everything You have said has been propoganda. I dont have to say what for, you didnt, I dont have to say why i think that, you didnt (though you tell me off for the same if it doesnt support your own propoganda)
Now I shall follow my own formula, its not perfect, but its less like propoganda that the previous sentence.
The majority of what you post seems to be propaganda to influence people into thinking NOH is the Bees Knees, You attack any stat I use against NOH, and ignore anything I say against any other tribe, You happily take any compliment I make towards NOH, even though they have the same base stats you are attacking. You rip into me for not backing up my prediction that say NOH may run into trouble because i did not back it up, but you are happy with Syphaxs statement that NOH are doing good, even though he did not back that up.
Now please, instead of attacking me and my attempts to bring a good friendly discussion to the table, either give alternative views, and add to the conversation as Syphax did, and as Nick has just done, or just dont post. If you like, I can put a disclaimer everytime i post saying "Walkingrock disagrees" that way you dont have to spout your negativity all the time.
I have made a prediction of the future of the world, Alliance to kill OTS, then WOW, then Darkness turns on either -X- or NOH, when i wrote that i was under the impression that all three were allied, i was disabused of that in another thread, but my prediction (which i think may be wrong) was that NOh would be next on the chopping block, and I dont have faith in the players I have been fighting that they would stick around when they are losing villages. first few sure, but lose a couple provinces of your big players (abe70 and dimeth being the two i was fighting, and both fairly close to were i anticipated the front lines to be) players will start to dissappear, those two in particular.
Have I explained my own personal opinion enough? or do you need me to justify every single word i say that doesnt call you the best ever? you are happy if i say the same against anyone else.
You have told me that what propaganda means, yet you did not justify how i did it.
Lets go through taht shall we?
"Propaganda is information that is not objective" - im human, as syphax said, im not sure we will ever get an unbiased view, but i like to think I do pretty good. If you disagree with what I say, make your own predictions, add to the conversation instead of taking away from it.
"and is used primarily to influence an audience" - what influence do you think I am trying to make? Im obviously not trying to get players to join OTS, i have left there myself.
Here, I shall do as you do, as you seem to be so greate at this
Everything You have said has been propoganda. I dont have to say what for, you didnt, I dont have to say why i think that, you didnt (though you tell me off for the same if it doesnt support your own propoganda)
Now I shall follow my own formula, its not perfect, but its less like propoganda that the previous sentence.
The majority of what you post seems to be propaganda to influence people into thinking NOH is the Bees Knees, You attack any stat I use against NOH, and ignore anything I say against any other tribe, You happily take any compliment I make towards NOH, even though they have the same base stats you are attacking. You rip into me for not backing up my prediction that say NOH may run into trouble because i did not back it up, but you are happy with Syphaxs statement that NOH are doing good, even though he did not back that up.
Now please, instead of attacking me and my attempts to bring a good friendly discussion to the table, either give alternative views, and add to the conversation as Syphax did, and as Nick has just done, or just dont post. If you like, I can put a disclaimer everytime i post saying "Walkingrock disagrees" that way you dont have to spout your negativity all the time.
DeletedUser3495
Guest
Echoes of rumours I'm afraid. Just generally dissatisfaction brought from congestion I believe is the root of it though.
It's not to say that there isn't the same thing occurring with WOW, -D- and -X-, I just think that its more prevalent currently between the first two since more of the -X- members are on the outside of the core. Most of us aren't pressured yet by congestion. Though if WOW keeps slingshotting over the top of me I'm gonna be a sad munchkin.
It's not to say that there isn't the same thing occurring with WOW, -D- and -X-, I just think that its more prevalent currently between the first two since more of the -X- members are on the outside of the core. Most of us aren't pressured yet by congestion. Though if WOW keeps slingshotting over the top of me I'm gonna be a sad munchkin.
DeletedUser4507
Guest
First, goal of this post is only to open discussion of which metrics, weights and form of result we should use, not to make proper analysis or rate tribes. Some of the datas are not from same time, so what you see below is not 100% relevant and it is shown only as an example of result. Used metrics are the first one that came to my mind and later I need to think about them.
Metrics and data origin:
TOP 10 conquers - TOP 10 losts. - I took that from TW2Analytics.com with data stamp 5th March to 11th March, which should show villages that were conquered between 4th and 10th March. Data are actually not correct, because I took only tribes are remembered being relevant.
Villages: in-game data from mid-day 14th of March. This can do inconsistancy. In real analysis it would be taken from same time like other datas.
Off Bash increase: Difference in total tribe offensive bashpoints in last week. Data taken from GRhin posts from 4th and 11th March.
Deff Bash increase: Difference in total tribe defensive bashpoints in last week. Data taken from GRhin posts from 4th and 11th March.
Evaluation: This is another part we can discus about. Below I used ranking of each metrics and then used weighted average of metrics ranking which was ranked again to final result. Another option is to set best tribe in each category as 1 (100%) and then divide all result to be in relative number/percents against best tribe from each category. Advantage is higher accuracy. Disadvantage is less clarity. In adition we can talk about what should be result of whole analysis. If it is tribe ranking, weighted average rankings (which I believe is more precise) or something completely else.
I hope I posted everything I wanted to post correctly.
Every constructive comment or suggestion is welcome. I would like to see as many ideas and opinions as possible so dont be afraid to speak up. If we later below agree on metrics, weights and evaluations, rest is Math. So result will be 100% unbiased unless you want to say that Math is biased.
P.S. If somebody has need of correcting my grammer, feel free to write it too. English is not my native language so I accept all chances for improving.
Metrics and data origin:
TOP 10 conquers - TOP 10 losts. - I took that from TW2Analytics.com with data stamp 5th March to 11th March, which should show villages that were conquered between 4th and 10th March. Data are actually not correct, because I took only tribes are remembered being relevant.
Villages: in-game data from mid-day 14th of March. This can do inconsistancy. In real analysis it would be taken from same time like other datas.
Off Bash increase: Difference in total tribe offensive bashpoints in last week. Data taken from GRhin posts from 4th and 11th March.
Deff Bash increase: Difference in total tribe defensive bashpoints in last week. Data taken from GRhin posts from 4th and 11th March.
Evaluation: This is another part we can discus about. Below I used ranking of each metrics and then used weighted average of metrics ranking which was ranked again to final result. Another option is to set best tribe in each category as 1 (100%) and then divide all result to be in relative number/percents against best tribe from each category. Advantage is higher accuracy. Disadvantage is less clarity. In adition we can talk about what should be result of whole analysis. If it is tribe ranking, weighted average rankings (which I believe is more precise) or something completely else.
I hope I posted everything I wanted to post correctly.
Every constructive comment or suggestion is welcome. I would like to see as many ideas and opinions as possible so dont be afraid to speak up. If we later below agree on metrics, weights and evaluations, rest is Math. So result will be 100% unbiased unless you want to say that Math is biased.
P.S. If somebody has need of correcting my grammer, feel free to write it too. English is not my native language so I accept all chances for improving.
DeletedUser4523
Guest
now they only thing you need to add is the number of players in each tribe and find what tribe is doing better ratio being the determining factor for example 35 players conquered these many villages vs 80 something players that conquered this many.
DeletedUser486
Guest
I havent checked the available stats sites so I don't know how it is tracked, which is why I am asking; how are conquers tracked in order to give an unbiased overview of war progress? Iirc recently a player was talked about growing solely off gifted villages from an 80k account, which means, presumably, these conquers being tracked could weight the analysis and skew any overview.
DeletedUser2369
Guest
@Walkingrock , i like the direction you are going here.
First off i will make a brief "defense" for my updates, not to take away from you, I really like what you are doing, I just want to justify why I will continue doing things my way. My updates are based around the maps, your stats better ranks tribes than mine (mostly cos i dont rank tribes), however where tribes are in relation to each other, and how they interact has a much bigger impact on the future of the tribes in my opinion (thats what I try to focus on, maybe thats why you and I keep at each others throats, you are expecting a solely numbers based view, I am trying to predict the future on much less tangible player interactions using the numbers soley to help me figure out the nature of the interactions?).
Now if you want to add your rankings below my post each week, the combined overview of the world might allow us to predict the future better than either method no their own, i think that could be really beneficial to everyone
The question I have for you is, what are you trying to quantify? as you cant rank tribes without defining what that rank means, Which is why i never tried doing more than listing the games rankings (which is quantifying who is closer to victory) and trying to predict the future from it.
Whatever your aim is, I wonder if using the weights on the ranking is the best approach? maybe use it on the base numbers? I just think that a tribe with barely lower values in three categories might be better than a tribe with much lower values in the 4th
e.g. -X- in the rankings you put above is one conquer and 30 villages off WOW,Let us pretend both values were the lower values to -X- for the purpose of example. I think these two numbers are almost the same ( <5% differince), now imagine offensive bash remained the same as your post, your ranking would put WOW ahead of -X- based off one extra conquer (remember we swapped those numbers for the example), and 30 villages total. I believe that in that situation the bash should put -X- ahead, i am of course ignoring def bash which is where -X- falls down, but you get the idea im trying to illustrate i hope.
If you are looking for war capacity, i think using only one weeks data is a mistake - if i enter into a massive fight, ill often need a week or two to rebuild, maybe three weeks would be better?
Also you need to look at size of villages, alot of certain tribes are taking more than their share of 2-300 point villages, thes villages start off as being a liability, rather than an asset, maybe you could do a total villages above a certain threshhold?
IF you are looking for war ability, then i think you need to remove barb conquers from your weighted village gain (maybe include them with a lower weighting?)
Also I beleive you need to usea bash-increase vs village increase comparison in order to weight the village increase. if a tribe got 100 villages, for 50k bash, vs a tribe getting 80 villages for 100k bash (unrealistic numbers I know, but they illustrate a realistic scenario) which one should be higher in your rankings?
Regardless, I still maintain you shouldnt use defensive bash, and from our last week long discussion on that i doubt we will agree. I will restate my reasoning in case youve forgotten in the heat of that convo - defensive bash can only be gained through the actions of another tribe, not your own actions. If you are not attacked you do not get defensive bash, no matter what you are looking for with your rankings, I cant imagine that will get the 100% unbiased result you are looking for.
And finally, while a lofty goal, 100% unbiased is a pipe dream, simply because the stats we get from tribalwars is imperfect. Math is not biased, but the way it is used can be, and the numbers used can be, I have said before, as have you, numbers dont tell the whole story.
I mentioned about defensive bash earlier, maybe if there were a way to know who caused the def bash, we could use it to decrease their weighted score, and it would be more accurate, but there isnt, and even that would be somewhat inacurrate because it also depends on the defensive capability of the defender, thus the defensive stats dont give you enough information to get a 100% perfect ranking no matter how good your math is, and all other numbers have the same issue to varying degrees.
In saying that, dont stop trying, you may not get to your perfect numbers, but the more you try, the closer you can get, and thats all anyone can ask.
First off i will make a brief "defense" for my updates, not to take away from you, I really like what you are doing, I just want to justify why I will continue doing things my way. My updates are based around the maps, your stats better ranks tribes than mine (mostly cos i dont rank tribes), however where tribes are in relation to each other, and how they interact has a much bigger impact on the future of the tribes in my opinion (thats what I try to focus on, maybe thats why you and I keep at each others throats, you are expecting a solely numbers based view, I am trying to predict the future on much less tangible player interactions using the numbers soley to help me figure out the nature of the interactions?).
Now if you want to add your rankings below my post each week, the combined overview of the world might allow us to predict the future better than either method no their own, i think that could be really beneficial to everyone
The question I have for you is, what are you trying to quantify? as you cant rank tribes without defining what that rank means, Which is why i never tried doing more than listing the games rankings (which is quantifying who is closer to victory) and trying to predict the future from it.
Whatever your aim is, I wonder if using the weights on the ranking is the best approach? maybe use it on the base numbers? I just think that a tribe with barely lower values in three categories might be better than a tribe with much lower values in the 4th
e.g. -X- in the rankings you put above is one conquer and 30 villages off WOW,Let us pretend both values were the lower values to -X- for the purpose of example. I think these two numbers are almost the same ( <5% differince), now imagine offensive bash remained the same as your post, your ranking would put WOW ahead of -X- based off one extra conquer (remember we swapped those numbers for the example), and 30 villages total. I believe that in that situation the bash should put -X- ahead, i am of course ignoring def bash which is where -X- falls down, but you get the idea im trying to illustrate i hope.
If you are looking for war capacity, i think using only one weeks data is a mistake - if i enter into a massive fight, ill often need a week or two to rebuild, maybe three weeks would be better?
Also you need to look at size of villages, alot of certain tribes are taking more than their share of 2-300 point villages, thes villages start off as being a liability, rather than an asset, maybe you could do a total villages above a certain threshhold?
IF you are looking for war ability, then i think you need to remove barb conquers from your weighted village gain (maybe include them with a lower weighting?)
Also I beleive you need to usea bash-increase vs village increase comparison in order to weight the village increase. if a tribe got 100 villages, for 50k bash, vs a tribe getting 80 villages for 100k bash (unrealistic numbers I know, but they illustrate a realistic scenario) which one should be higher in your rankings?
Regardless, I still maintain you shouldnt use defensive bash, and from our last week long discussion on that i doubt we will agree. I will restate my reasoning in case youve forgotten in the heat of that convo - defensive bash can only be gained through the actions of another tribe, not your own actions. If you are not attacked you do not get defensive bash, no matter what you are looking for with your rankings, I cant imagine that will get the 100% unbiased result you are looking for.
And finally, while a lofty goal, 100% unbiased is a pipe dream, simply because the stats we get from tribalwars is imperfect. Math is not biased, but the way it is used can be, and the numbers used can be, I have said before, as have you, numbers dont tell the whole story.
I mentioned about defensive bash earlier, maybe if there were a way to know who caused the def bash, we could use it to decrease their weighted score, and it would be more accurate, but there isnt, and even that would be somewhat inacurrate because it also depends on the defensive capability of the defender, thus the defensive stats dont give you enough information to get a 100% perfect ranking no matter how good your math is, and all other numbers have the same issue to varying degrees.
In saying that, dont stop trying, you may not get to your perfect numbers, but the more you try, the closer you can get, and thats all anyone can ask.
Last edited by a staff member:
DeletedUser4818
Guest
Well, I didn't respond to Syphax cause I knew what he threw around about the tribes weren't a comment to be taken serious. I do believe we do good, sure but he threw in the comment about -x- being barb nobling noobs and will be gone at the first sign of resistance so I just ignored that discussion about tribes, cause I didn't find it serious.
DeletedUser4507
Guest
I took it some time to think about it more, but below are reactions, example of "another" option structure I talked about above and my preferences of "settings" of analysis. Sadly, it is long, so I put separate parts under spoilers to not disrupt flow of this forum thread.
@Yatogami Thanks to you, I have noticed that stats page mentioned above is not showing internal nobling, which is little sad, because it would show interesting information about integrity of tribe. I would never considered internal nobling as conquer to analysis.
About ratio metrics: This will be reaction to both @miguel8819643 and @GRhin, who both propose them. Generaly ratio metrics has that danger, that are unstable. They have part of divident and divisor. if divisor is stable and change only slowly, ratios are showing some realy good informations. But things goes wild when divisor is changed by a lot. Way how we would probably compute them, for example some form of bash/member, is to take tribe bash and divide it by members. Tribe during week kick 5-10 inactive players (those with low bash) bash/member will skyrocket up, because they add more bash and decrease member count. Does this represent how well they are doing. I say that not. Same goes with something/village ratios. You start loosing villages and despite you are doing bad, your ratios will go up. Using ratio metrics needs to be done after long thinking about all posibilities what could happen and how it will effect the result.
Back to @GRhin. He had most ideas so far. Interaction between tribes is combat and it is rated as off bash and deff bash by devs. Low bash gain is sign of peace. If done correctly, this form of analysis will show how tribes are doing and if they are developing the right way to win the server. It is showing, if nothing will change they will ends up it this order. Of course higher is time period, bigger is accuracy. Thats why I think this is even shouldnt be done weekly. More like every second week. 2 weeks period seems to me like enough time to represent changes that happened.
Few next paragraph to show how rankings not work. I wrote about "another" option and if it wasnt clear, I prefer it more that casual ranks. I will post how it would be looked like, so all of you can see it. All of "minor" differences will represented correctly.
About size of the villages. In longer run it doesnt matter. Those villages will be eventually build up and difference between training capacities is ridiculously low. Plus, dominations worlds have winning condition of 70-80% (not sure) of top10 tribe villages. Number of villages. Not village points. (representation of size) Is this right? I dont know, but winning condition is set and we have to play with it.
About deff bash relevancy. I am open to discuss the weight. I remember you points, but ignoring them completely seems to be huge mistake. We could see what happens when whole tribe (POT) orientate only for off. They loose very quickly. Deff is important part of game. If we watch on the scale of off-bash increase and deff-bash increase, off is more of less double the size and thus it look relevant to say that it should have bigger weight. But what is the correct weight? If we get something like 2:1, it will be similiar (not same) to just doing overall bash increase.
Unbias is possible. 100% relevancy not. Biased are opinions of people. Math is 100% relevant and I think that game where everything (except names) is number is ideal to have pure number analysis. Only question is, what are the numbers that relevantly represents what we are looking for?
As for the final. This is analysis that will have different results through time. Once you can be 2nd, then 4th, week after 1st, but tribes who rates high and will be rated high more often will be major candidates for winning server. All the inaccurities, all of the flaws it has, will in future shrink to war between tribes and because we agreed that conquers are what matter the most (we need to find settings to represent it correctly) those who will capture most will eventualy win the server.
About ratio metrics: This will be reaction to both @miguel8819643 and @GRhin, who both propose them. Generaly ratio metrics has that danger, that are unstable. They have part of divident and divisor. if divisor is stable and change only slowly, ratios are showing some realy good informations. But things goes wild when divisor is changed by a lot. Way how we would probably compute them, for example some form of bash/member, is to take tribe bash and divide it by members. Tribe during week kick 5-10 inactive players (those with low bash) bash/member will skyrocket up, because they add more bash and decrease member count. Does this represent how well they are doing. I say that not. Same goes with something/village ratios. You start loosing villages and despite you are doing bad, your ratios will go up. Using ratio metrics needs to be done after long thinking about all posibilities what could happen and how it will effect the result.
Back to @GRhin. He had most ideas so far. Interaction between tribes is combat and it is rated as off bash and deff bash by devs. Low bash gain is sign of peace. If done correctly, this form of analysis will show how tribes are doing and if they are developing the right way to win the server. It is showing, if nothing will change they will ends up it this order. Of course higher is time period, bigger is accuracy. Thats why I think this is even shouldnt be done weekly. More like every second week. 2 weeks period seems to me like enough time to represent changes that happened.
Few next paragraph to show how rankings not work. I wrote about "another" option and if it wasnt clear, I prefer it more that casual ranks. I will post how it would be looked like, so all of you can see it. All of "minor" differences will represented correctly.
About size of the villages. In longer run it doesnt matter. Those villages will be eventually build up and difference between training capacities is ridiculously low. Plus, dominations worlds have winning condition of 70-80% (not sure) of top10 tribe villages. Number of villages. Not village points. (representation of size) Is this right? I dont know, but winning condition is set and we have to play with it.
About deff bash relevancy. I am open to discuss the weight. I remember you points, but ignoring them completely seems to be huge mistake. We could see what happens when whole tribe (POT) orientate only for off. They loose very quickly. Deff is important part of game. If we watch on the scale of off-bash increase and deff-bash increase, off is more of less double the size and thus it look relevant to say that it should have bigger weight. But what is the correct weight? If we get something like 2:1, it will be similiar (not same) to just doing overall bash increase.
Unbias is possible. 100% relevancy not. Biased are opinions of people. Math is 100% relevant and I think that game where everything (except names) is number is ideal to have pure number analysis. Only question is, what are the numbers that relevantly represents what we are looking for?
As for the final. This is analysis that will have different results through time. Once you can be 2nd, then 4th, week after 1st, but tribes who rates high and will be rated high more often will be major candidates for winning server. All the inaccurities, all of the flaws it has, will in future shrink to war between tribes and because we agreed that conquers are what matter the most (we need to find settings to represent it correctly) those who will capture most will eventualy win the server.
Dont looked on numbers. As I mentioned in first post about this, numbers are from both different sources and different times. Look on structure, if you think that this structure could represent reality better.
Despite I took about 1 minute to make first 4 metrics, that should serve only as an example, it surprised me that was very close to representation of reality and what settings of metrics I would prefer to use.
Metrics:
Relevant tribes conquers-losts (weight 2): As relevant I would now consider 6 above minus OTS. Thus we can call it top5.
All conquers-all losts (weight 1): This is questionable to use, but conquers are what matters the most and despite barb nobling or nobling of inactive player is not great deed. Village will produce troops. So ignore it completely seems to me like hiding before truth. Relevant tribes conquers are included here again and thus they have total weight of 3.
Amount of villages (weight 1): This is how far are tribes to gaining world dominance, but says nothing about how good they are doing, thus weight 1.
Off bash increase (weight 2): Based on both size against deff bash and my agreement with GRhin about off bash being more important than deff bash. I still believe that bash increase is better representaion of reality and total bash are more history that actual deeds.
Deff bash increase (weight 1): Yea, deffending matters.
Before you have need of reaction, please sit to it, think about what metrics you think matters and rather send your own set of metrics and weights (or skip them if you think weights dont matter) than just tear any suggestions apart. Yea I know I did something like that and no, I dont think it was right approach. Thats why I am trying to restarted debate in more constructive form.
Metrics:
Relevant tribes conquers-losts (weight 2): As relevant I would now consider 6 above minus OTS. Thus we can call it top5.
All conquers-all losts (weight 1): This is questionable to use, but conquers are what matters the most and despite barb nobling or nobling of inactive player is not great deed. Village will produce troops. So ignore it completely seems to me like hiding before truth. Relevant tribes conquers are included here again and thus they have total weight of 3.
Amount of villages (weight 1): This is how far are tribes to gaining world dominance, but says nothing about how good they are doing, thus weight 1.
Off bash increase (weight 2): Based on both size against deff bash and my agreement with GRhin about off bash being more important than deff bash. I still believe that bash increase is better representaion of reality and total bash are more history that actual deeds.
Deff bash increase (weight 1): Yea, deffending matters.
Before you have need of reaction, please sit to it, think about what metrics you think matters and rather send your own set of metrics and weights (or skip them if you think weights dont matter) than just tear any suggestions apart. Yea I know I did something like that and no, I dont think it was right approach. Thats why I am trying to restarted debate in more constructive form.
seriously ?! -.- you should treat other people opinion with more respect ...
Last edited:
DeletedUser2369
Guest
Good post once again walkingrock, I like this new approach to the debate, I will attempt to stick to it myself.
Now please dont misinterpret me, im not tearing your suggestions apart, or at least thats not my intention. I dont have the answers, but I figure if i can highlight some problem areas, and suggest directions to fix it, we can get a better solution together.
First off I disagree with one statement - this part isnt actually important to the current discussion so feel free to ignore the spoiler;
"Interaction between tribes is combat and it is rated as off bash and deff bash by devs. Low bash gain is sign of peace. If done correctly, this form of analysis will show how tribes are doing and if they are developing the right way to win the server."
but thats okay, as the updates I have been posting provides what I want, I am just justifying keeping my updates as is, not trying to change your rankings.
Or better said, I agree with everything you said, however i disagree with your implied conclusion that the numbers can predict whats going to happen best. For example, NOH is beside WOW, with the rim and gum on their other side, while WOW has NOH on one side and Darkness on the other, therefore they have different difficulties to get past to progress. I promise this is not a dig on NOH, but it would be easier for NOH to fight both WOW and gum, than it would be for WOW to fight both NOH and Darkness. On the other had, an alliance in the right place can hinder or encourage growth. I personally believe these sorts of scenarios have a major impact on the futre of the tribe, as much as, if not more than, the stats you are using, however we cannot quantify that sort of thing.
Basically my updates give background context to the stats and rankings you are trying to set up, and context is important to understand the results of the math.
Or better said, I agree with everything you said, however i disagree with your implied conclusion that the numbers can predict whats going to happen best. For example, NOH is beside WOW, with the rim and gum on their other side, while WOW has NOH on one side and Darkness on the other, therefore they have different difficulties to get past to progress. I promise this is not a dig on NOH, but it would be easier for NOH to fight both WOW and gum, than it would be for WOW to fight both NOH and Darkness. On the other had, an alliance in the right place can hinder or encourage growth. I personally believe these sorts of scenarios have a major impact on the futre of the tribe, as much as, if not more than, the stats you are using, however we cannot quantify that sort of thing.
Basically my updates give background context to the stats and rankings you are trying to set up, and context is important to understand the results of the math.
I assume you have changed the numbers to be, in laymans terms, a percentage of the top score in that category? i.e. tribe score divided by top score.?
And finally your weightings, tbh i dont want to weigh in here (pun not intended) untill you clarify what you are trying to see with these rankings.
As mentioned earlier war capacity (troop production capacity, max number of nukes etc) =/= war ability (the ability to use what nukes they have), and so the weightings would be different for those two cases, if you are talking of progress towards winning, then number of villages is all you need (nobody cares about this untill everyone already knows who the winner is going to be).
Also bash increase (depending on the period) could also indicate troops lost, so a lessened war capacity. But a total bash more accurately describes war ability as you can assume that past exploits indicate potential future exploits. e.g. high bash so far means they probably know how to use their troops well.
If you are interested only in relative strength (i.e. you want to compare a new rim tribe such as gum to a core tribe such as WOW to see how they are going for their situation, rather than specific current abilities) you could use bash per village, or bash per points (this is partially covered in bash per member, but WOW has much larger members with more villages and more troops than gum has). If you dont do this gum will be at a lower rank right now even if they are fighting better than WOW (i use "better" as a relative term, a player with 5 villages conquering 10 villages is "better" than a player with 30 villages conquering 15).
And finally your weightings, tbh i dont want to weigh in here (pun not intended) untill you clarify what you are trying to see with these rankings.
As mentioned earlier war capacity (troop production capacity, max number of nukes etc) =/= war ability (the ability to use what nukes they have), and so the weightings would be different for those two cases, if you are talking of progress towards winning, then number of villages is all you need (nobody cares about this untill everyone already knows who the winner is going to be).
Also bash increase (depending on the period) could also indicate troops lost, so a lessened war capacity. But a total bash more accurately describes war ability as you can assume that past exploits indicate potential future exploits. e.g. high bash so far means they probably know how to use their troops well.
If you are interested only in relative strength (i.e. you want to compare a new rim tribe such as gum to a core tribe such as WOW to see how they are going for their situation, rather than specific current abilities) you could use bash per village, or bash per points (this is partially covered in bash per member, but WOW has much larger members with more villages and more troops than gum has). If you dont do this gum will be at a lower rank right now even if they are fighting better than WOW (i use "better" as a relative term, a player with 5 villages conquering 10 villages is "better" than a player with 30 villages conquering 15).
DeletedUser4507
Guest
@GRhin Your last post was very helpful. I have been thinking about it and lots of what you mentioned is closely tied to major question: What is goal of my analysis? I admit I had no answer so far. Of course there are many things we disagree upon and I dont think it will change.
To answer I went to start of our discussion and main reason I joined here. You, using metrics that were telling only part of the story or metrics that I didnt even see how are related to your predictions. I hoped I get either explanation that will be logical and make sense or I will be able to pursuade you to change your metrics to something more relevant to your predictions.
So, I am putting out number of villages and keeping rest metrics I suggested in my last post with weights I used and call it measurement of tribe performance. I dont think that the second part table with tribe score divided by top score is not necesary, even though it provide interesting data. In addition, despite I want to post this table/2 tables weekly, it doesnt necessarily need to use 1 week period. 3 weeks I personaly consider too old to reflect current situation, but 2 weeks could be OK. That would mean that anything happens will be included in 2 consecutive "weekly" measurements. Accuracy of 2 week period is questionable, but it should teoreticaly works.
What I described above is not full analysis, but rather data base for your "predictions" or whatever you want call it. (no offense meant) You mentioned somewhere that you didnt want to rank tribes. I dont think it is necessary, but I believe that you should either use no data or all data. (at least those who we, agree upon that are relevant)
When you add diplomacy to that it should work very well.
As posts before, used metrics, weights or other things are open for general discussion that should lead to some kind of consensus about data/metrics.
To answer I went to start of our discussion and main reason I joined here. You, using metrics that were telling only part of the story or metrics that I didnt even see how are related to your predictions. I hoped I get either explanation that will be logical and make sense or I will be able to pursuade you to change your metrics to something more relevant to your predictions.
So, I am putting out number of villages and keeping rest metrics I suggested in my last post with weights I used and call it measurement of tribe performance. I dont think that the second part table with tribe score divided by top score is not necesary, even though it provide interesting data. In addition, despite I want to post this table/2 tables weekly, it doesnt necessarily need to use 1 week period. 3 weeks I personaly consider too old to reflect current situation, but 2 weeks could be OK. That would mean that anything happens will be included in 2 consecutive "weekly" measurements. Accuracy of 2 week period is questionable, but it should teoreticaly works.
What I described above is not full analysis, but rather data base for your "predictions" or whatever you want call it. (no offense meant) You mentioned somewhere that you didnt want to rank tribes. I dont think it is necessary, but I believe that you should either use no data or all data. (at least those who we, agree upon that are relevant)
When you add diplomacy to that it should work very well.
As posts before, used metrics, weights or other things are open for general discussion that should lead to some kind of consensus about data/metrics.
Well, we got bored.
DeletedUser4507
Guest
Anything constuctive or you just have necesity to puke here?
This is constructive if you paid attention to what just happened
DeletedUser4507
Guest
What is constructive on "We got bored."?
Pay attention to the development of the world and you'll understand. Why are you so rude towards your ally anyway?
Please stop with the offtopic stuff, if you don't get the message then don't comment about it
I don't wanna annoy my bestie coolnite (<3)