Results 1 to 8 of 8

Thread: Another ranking system(!): the enlil strength method

  1. #1
    Join Date
    Oct 2009
    Posts
    333

    Default Another ranking system(!): the enlil strength method

    After reading Balt's thread of I was reminded of my own rating system I tinkered around with a few decades ago when I first got hooked on the Evvv.

    We all know the numbers mean crap. But if we're going to devise a ranking system based on them, we might as well use them all.

    Average Member Prestige

    Total Alliance Prestige/Members = x

    Average Member Honor

    Total Alliance Honor/Members = y

    Average Member Strength

    x + y = z

    (Now here's where it gets tricky)

    City Ratio

    More cities = Less Prestige Gain, Greater Offensive Potential

    Members/Total # of Cities for the Alliance = A

    A x 100 = a

    (A is always going to be a decimal and doesn't factor into the equation properly unless you turn it into a whole number. Plus, z is going to be a number relatively approximate to the typical prestige/honor level found on whatever server you happen to jplay. We're after something different, so turning the city ratio into a larger number we take away the relative similarity to prestige/honor when we reach the end of the equation.)

    Average Member Prestige/Honor when adjusted w/ City Ratio

    z / a = STRENGTH NUMBER








    So...let's take a look at a typical mid-range server: Server 24.

    We'll look at the top three alliances (per prestige) and two outliers.

    All figures were taken around 11:30 PM on Nov. 8th

    #1) NCH - 96 members, 105 mil prestige, 121 mil honor, 454 cities

    x) 105 mil/96 = 1.093 mil
    y) 121 mil/96 = 1.206 mil
    z) x + y = 2.3 mil
    A) 96/454 = .211
    a) .211 x 1000 = 211
    STRENGTH - 2.3 mil / 210 = 10,900

    #2) R~T~H - 94 members, 83.8 mil prestige, 60.9 honor, 440 cities

    x) 83.8 mil/94 = 891.9k
    y) 60.9 mil/94 = 647.9k
    z) x + y = 1.54 mil
    A) 94/440 = .214
    a) .214 x 1000 = 214
    STRENGTH - 1.54 mil / 214 = 7,196

    #3) LIFE - 99 members, 73.6 mil prestige, 118 mil honor, 438 cities

    x) 73.6 mil/99 = 743k
    y) 118 mil/99 = 1.192 mil
    z) x + y = 1.935 mil
    A) 99/438 = .226
    a) .226 x 1000 = 226
    STRENGTH - 1.935 mil / 226 = 8561

    #19) IDK - 40 members, 42.2 mil prestige, 59.9 mil honor, 151 cities

    x) 42.2 mil/40 = 1.055 mil
    y) 59.9 mil/40 = 1.498 mil
    z) x + y = 2.553 mil
    A) 40/151 = .265
    a) .265 x 1000 = 265
    STRENGTH - 2.553 mil / 265 = 9,633

    #21) Templars - 99 members, 41.7 mil prestige, 193 mil honor, 254 cities

    x) 41.7 mil/99 = 421k
    y) 193 mil/99 = 1.949 mil
    z) x + y = 2.370 mil
    A) 99/254 = .389
    a). 389 x 1000 = 389
    STRENGTH - 2.370 mil / 389 = 6092

    Stack them all tentatively together and....

    STRENGTH RANKING

    #1) NCH - 10,900
    #2) IDK - 9,633
    #3) LIFE - 8,561
    #4) R~T~H - 7,196
    #5) Templars - 6,092

    The final number factors in all available numerical variables and gives us the closest approximation to a real number we can use to compare alliances to one another. However...we all are aware that prestige means nothing and honor is the worst thing in the world. This ranking system does do a fairly good job at exposing the prestige poof alliances by adding honor and # of cities into the equation. Notice how far R~T~H fell and how high IDK climbed. NCH still remains at the top as the dominant alliance. When hunting for a new primary target, NCH would be wise to choose R~T~H before IDK.

    When looking at the Templars we see an abnormally high honor average per member and high city ratio. From these number we are able to assume that they are primarily a defensive alliance. Their low city count & high honor leads to the assumption that they are while they are able to defend themselves, they are not strong enough to build and hold additional cities needed to create the size army necessary for successful offensive attacks.

    This ranking method can be used to semi-accurately compare alliances across servers. We would have to create different server classes, however, to differentiate between the older and newer ones. It might be possible to compare server 4 to server 24, but not server 4 to server 83.

    Ok...I'm done for now. Any and all comments appreciated, especially those from Lord Ken. I'd really like to hear your thoughts on this topic. I just wasted an hour's worth of studying time to put this together to show you all because apparently I value my Evony knowledge more so than my grasp of Mandarin . If I made any simple math mistakes, it's 'cause I have a midterm in 8.5 hours and really don't feel like taking the time to go back and give it all a good proper edit. I'll do that tomorrow if I'm still into this idea.
    Last edited by enlil; 11-09-2009 at 01:59 AM.

  2. #2

    Default

    honor is useless
    prestige is useless
    the whole ranking system is useless

  3. #3
    Join Date
    Oct 2009
    Posts
    333

    Default

    .....thanks for adding your sparkling twinkle of brilliance.


    Quote Originally Posted by enlil View Post
    We all know the numbers mean crap.

  4. #4

    Default :)

    well that paints us in a bright light. seeing as we can and have given nch a run for there money in the encounters we have had i can see some validity to your system. I agree in that honor means nothing. I try to keep myne low personally. I am too lazy to run this calculation without it. It would also be cool if you could remove inactives from the equation nch actually has a lot of players with 1 city and 0 pop these don't count.

  5. #5

    Default

    Quote Originally Posted by enlil View Post
    After reading Balt's thread of I was reminded of my own rating system I tinkered around with a few decades ago when I first got hooked on the Evvv.

    We all know the numbers mean crap. But if we're going to devise a ranking system based on them, we might as well use them all.

    Average Member Prestige

    Total Alliance Prestige/Members = x

    Average Member Honor

    Total Alliance Honor/Members = y

    Average Member Strength

    x + y = z

    (Now here's where it gets tricky)

    City Ratio

    More cities = Less Prestige Gain, Greater Offensive Potential

    Members/Total # of Cities for the Alliance = A

    A x 100 = a

    (A is always going to be a decimal and doesn't factor into the equation properly unless you turn it into a whole number. Plus, z is going to be a number relatively approximate to the typical prestige/honor level found on whatever server you happen to jplay. We're after something different, so turning the city ratio into a larger number we take away the relative similarity to prestige/honor when we reach the end of the equation.)

    Average Member Prestige/Honor when adjusted w/ City Ratio

    z / a = STRENGTH NUMBER








    So...let's take a look at a typical mid-range server: Server 24.

    We'll look at the top three alliances (per prestige) and two outliers.

    All figures were taken around 11:30 PM on Nov. 8th

    #1) NCH - 96 members, 105 mil prestige, 121 mil honor, 454 cities

    x) 105 mil/96 = 1.093 mil
    y) 121 mil/96 = 1.206 mil
    z) x + y = 2.3 mil
    A) 96/454 = .211
    a) .211 x 1000 = 211
    STRENGTH - 2.3 mil / 210 = 10,900

    #2) R~T~H - 94 members, 83.8 mil prestige, 60.9 honor, 440 cities

    x) 83.8 mil/94 = 891.9k
    y) 60.9 mil/94 = 647.9k
    z) x + y = 1.54 mil
    A) 94/440 = .214
    a) .214 x 1000 = 214
    STRENGTH - 1.54 mil / 214 = 7,196

    #3) LIFE - 99 members, 73.6 mil prestige, 118 mil honor, 438 cities

    x) 73.6 mil/99 = 743k
    y) 118 mil/99 = 1.192 mil
    z) x + y = 1.935 mil
    A) 99/438 = .226
    a) .226 x 1000 = 226
    STRENGTH - 1.935 mil / 226 = 8561

    #19) IDK - 40 members, 42.2 mil prestige, 59.9 mil honor, 151 cities

    x) 42.2 mil/40 = 1.055 mil
    y) 59.9 mil/40 = 1.498 mil
    z) x + y = 2.553 mil
    A) 40/151 = .265
    a) .265 x 1000 = 265
    STRENGTH - 2.553 mil / 265 = 9,633

    #21) Templars - 99 members, 41.7 mil prestige, 193 mil honor, 254 cities

    x) 41.7 mil/99 = 421k
    y) 193 mil/99 = 1.949 mil
    z) x + y = 2.370 mil
    A) 99/254 = .389
    a). 389 x 1000 = 389
    STRENGTH - 2.370 mil / 389 = 6092

    Stack them all tentatively together and....

    STRENGTH RANKING

    #1) NCH - 10,900
    #2) IDK - 9,633
    #3) LIFE - 8,561
    #4) R~T~H - 7,196
    #5) Templars - 6,092

    The final number factors in all available numerical variables and gives us the closest approximation to a real number we can use to compare alliances to one another. However...we all are aware that prestige means nothing and honor is the worst thing in the world. This ranking system does do a fairly good job at exposing the prestige poof alliances by adding honor and # of cities into the equation. Notice how far R~T~H fell and how high IDK climbed. NCH still remains at the top as the dominant alliance. When hunting for a new primary target, NCH would be wise to choose R~T~H before IDK.

    When looking at the Templars we see an abnormally high honor average per member and high city ratio. From these number we are able to assume that they are primarily a defensive alliance. Their low city count & high honor leads to the assumption that they are while they are able to defend themselves, they are not strong enough to build and hold additional cities needed to create the size army necessary for successful offensive attacks.

    This ranking method can be used to semi-accurately compare alliances across servers. We would have to create different server classes, however, to differentiate between the older and newer ones. It might be possible to compare server 4 to server 24, but not server 4 to server 83.

    Ok...I'm done for now. Any and all comments appreciated, especially those from Lord Ken. I'd really like to hear your thoughts on this topic. I just wasted an hour's worth of studying time to put this together to show you all because apparently I value my Evony knowledge more so than my grasp of Mandarin . If I made any simple math mistakes, it's 'cause I have a midterm in 8.5 hours and really don't feel like taking the time to go back and give it all a good proper edit. I'll do that tomorrow if I'm still into this idea.
    To me, the number of cities/ number of members matters most.
    TriPham | Mutiny | Server 6 | 1.5M + Prestige
    "We make war that we may live in peace." - Aristotle


  6. #6

    Default

    I believe the honor factor is relatively meaningless.

    The prestige/members ratio is a lot more telling to me when I am evaluating targets.

    For example, when the Tweedles alliance first appeared on my server, they were ranked in the mid 50's with 11 players. All the alliances around them were in the 90's.

    Clearly, they were much more formidable opponents. On paper. Naturally, since prestige never goes down, only up, that's not to say they couldn't be easy targets -- nothing to say they didn't have 11 cities total at that particular moment with zero troop strength between the players.

    Thus, I don't really think there is a generic truism you can use to properly evaluate.... it more or less becomes a judgement call based on previous experience and other factors.

    For example, your cities on my server aren't all that far away... but I'd only scout or attack them if I wanted to commit suicide

  7. #7
    Join Date
    Oct 2009
    Posts
    333

    Default

    There really isn't any accurate way to compare based on just the numbers...

    It really all depends on army size and the knowledge of how to use your troops efficiently. But until people start sharing their troop numbers with the world...all we got is what Evony gives us.

  8. #8
    Join Date
    Jul 2009
    Location
    Check your Reports
    Posts
    1,360

    Default

    a bit complicated there, but glad to know I inspired someone.

    Quit Evony (and Forums) 8/18/2010
    Periodically returns, but spends most of time on reddit.

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •