Page 1 of 5 123 ... LastLast
Results 1 to 10 of 49

Thread: CWL interim results March 2019

  1. #1

    Join Date
    Sep 2014
    Digging another tunnel in the data mine

    CWL interim results March 2019

    I've had a few requests, so here are the interim results I've found for Champ 1 this time. I only show the number of days completed by ALL clans. Early starters will have finished 1 day more than this.

    First up, the rankings:
    (NOTE: the ranking on this page are done by war wins, then stars; however the top 4 qualification for Poland is done by position in group then stars - so clans winning 6 might qualify as long as they won their group)

    It shows both total stars and optionally net stars. Triples is partly obvious - BUT some wars are excluded. Silly wars (45-0) are not counted. So a clan might only be 1/15, ie showing 15 attacks at the end of day 2 if one of their wars was excluded. This field isn't used by SC, but is just included for interest. The Group column is a made up code for the group of 8 clans - this comes from the other file, below.

    The final columns, all under the heading "Daily stars" shows the actual stars in each day. There is a tick box at the top of the page which will also show the net stars each day in brackets after the total. A net (+0) means a win in tie-break whereas (0) is a loss. This is hidden by default to make it more readable.

    The order is sorted pretty much by wins then total stars which is what SC do (they might have another tie-breaker after that though?)

    UPDATE: On Day 5 the list has been upgraded with a shenanigan-detector which highlights all daily scores that might be considered suspect. The meaning of the colours is at the top of the page. The first ones (orange, yellow, 2*blue) indicate weak opponents. Green is just a score much higher than anyone else in the group managed - that could just be good attackers. Brown looks at a clan's expected score based on their score in other wars and their opponent's defence scores in other wars, and highlights if the real score is a LOT higher than this - which MIGHT be the opponent gifting them stars but often isn't. Black is used when BOTH clan scores meet this condition, and there is a high chance of star trading here, especially if the scores are high. Note that most highlights are not foul play - just weak opponents.

    The second file is the complete set of results:

    This is pretty self-explanatory. At the bottom is a 3* rate summary. Lines in red in the results are ignored for the 3* rate (any war which was hugely lopsided probably didn't have proper TH12s on one side). If the stars are tied in a war, a "*" shows the winner.

    And as always: these results are unofficial and could be incomplete or wrong. The data comes from the API, but requires a fair bit of processing to get to this stage.

    SECOND UPDATE: As an experiment I've added an alternative ranking table:

    This calculates a daily score based on comparing a clan's result, both offence and defence, against an expected result based on their opponent's typical results (typical being the average of the middle ranked results). So if the opponent is going for 35* on defence every time, then scoring 35 against them is a score of 0. The target is adjusted up for scores under 30 - so if the opponent had numerous 25* defences, then 28 is needed for a 0 score.
    Silly days are just disregarded and not given a score. If the opponent is not all TH12 the war is still counted because the expected scores will account for that - but taken to extremes the war is omitted.
    The total score is then an average of the daily scores (so it doesn't matter if a clan has fewer than 7 valid wars), plus 5 * win percentage. The table is then ranked by score, without counting wins first.

    You can see this table doesn't have a big cluster of yellow (non-TH12 opponent) results at the top, like the normal one!

    THIRD UPDATE: I've generated the same files for champ 2...
    Full results:
    Alternate rankings:

    The Ch3 versions are also available, but the files are quite large:

    Last edited by OnyxDS; March 12th, 2019 at 04:02 PM.

  2. #2
    You are the man. Thank you so much. Please keep it updated.

  3. #3
    Forum Veteran ChiefTuk's Avatar
    Join Date
    May 2015
    Good work, as always!
    Leader of Bellum Tigers #2CR0QQC
    Expert Goblin Butterknife Attacker

  4. #4

    Join Date
    Sep 2014
    Digging another tunnel in the data mine
    Results including day 3 now posted. BUT group "A" is running over 90 minutes late, so I've just posted without their day 3

  5. #5
    Senior Member
    Join Date
    Feb 2019
    I'm wondering what will SC change this time to rank clans.
    Top4 will go to the WC Oflfine Qualifiers supposedly. But you will never make it there unless you're in a group with tanking clans.
    Like look at that #1 clan - already has a 40* and 45* wars. Good luck to legit clans getting 25 triples to catch up.

  6. #6
    Centennial Club
    Join Date
    Feb 2019
    If I take a look over Champs1 and see that in all seasons there are clans listed in topten which: had banns in their clan or banns in opponents clans during the season, or groups with tanking clans and so on....
    I think: why I'm not facing things like that in our matches ) joking, never want such things for us, but it does not represent that everything is competitive in champ 1.
    What is needed is a kind of punishment for destroying competitiveness in champ 1, if a clan tanks he destroys the fairplay for clans in other groups and should get freezed out of champs 1 forever (if this is promoted, I think No one would do this anymore or maybe less would do)

  7. #7
    Join Date
    Feb 2015
    As usual, awesome work Onyx. Thanks.

    As expected, the 3 star rate has reduced slightly with those balance changes.

  8. #8

    Join Date
    Sep 2014
    Digging another tunnel in the data mine
    Updated for day 4.

    The day 4 data for the clans in a group of 6 (groups AA and BB in my list) is missing; this is an artefact of how I detect complete days. I could fix it, but it's not really worth it since these groups aren't eligible for the top of the rankings anyway

  9. #9
    UNfair Cwl World championship, look at first 8 and last 8 teams )

  10. #10
    Millennial Club
    Join Date
    Dec 2015
    Has any clan other then Dark Looters been in the top 10 every month? Has Dark Looters ever had the "luck" to get in the same group as one of the clans giving up free 45's?
    My attacks are always awesome...just not necessarily in a good way.

    If you disagree with me, you are being toxic. If I disagree with you, I'm just speaking the truth.

Posting Permissions

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