Page 1 of 3 123 LastLast
Results 1 to 10 of 26

  Click here to go to the first staff post in this thread.   Thread: War search time reset effect match?

  1. #1
    Senior Member DerpSmash's Avatar
    Join Date
    Jan 2018
    Posts
    205

    War search time reset effect match?

    A long time ago in a clan far, far away... the leader insisted that the longer the search ran, the wider the potential match parameters become, thus the chance of a mismatch increase, therefore the search should be reset every 30 minutes or so to increase the likelyhood of an even match. I've been doing it out of habit now, but... is it true? Is there an ideal time to wait before resetting it?

    Thanks.

  2. #2
    Forum Veteran Thegreatpuma's Avatar
    Join Date
    Mar 2016
    Location
    Druidia
    Posts
    1,675
    Quote Originally Posted by DerpSmash View Post
    A long time ago in a clan far, far away... the leader insisted that the longer the search ran, the wider the potential match parameters become, thus the chance of a mismatch increase, therefore the search should be reset every 30 minutes or so to increase the likelyhood of an even match. I've been doing it out of habit now, but... is it true? Is there an ideal time to wait before resetting it?

    Thanks.
    it's common belief around here that that's true.

    I'm doing 5v5 th6 wars right now and it would seem that the longer the search runs the more likely it is that I'll be facing a 7 at the top. at the same time in my main clan, once or twice we've let the search run into multi hour and have never had a bad mismatch as a result. we usually stop it simply because if it matches too late it makes it a pain to manage the war.

    it's probably hard to say for sure.

  3. #3
    Forum Elder
    Join Date
    Jul 2017
    Location
    NE England, UK.
    Posts
    2,727
    I had time to think about this, so changing my post..
    It is popular for people to accept that the search parameters widen the longer a search runs, but..

    We know that a war match is made using the full roster..
    A common break down will match quickly.. But does an uncommon one widen its scope in order to get a hit?
    For that to be the case, then your opposition would also need to have been searching for a similar length of time, otherwise it would simply dismiss matching you, since those skewed parameters would not be available for their match.

    A quick match for one side against a clan with widened parameters would lead to a mis match.

    There is more going on than simply adding the offence and defence of two rosters, then looking for a similar number.

    Each individual account will have some bearing on what is being sought..
    For instance, I was running 8s on top, and decided to jump the halls on each account.
    Nothing else was upgraded, and the hall carries no weight, so the halls simply changed colour.
    But, I started to draw 9s at the top.
    My conclusion from this, which was pre suggested by some prominent engineers, is that if you carry a "colour" at the top, you are likely to draw that colour hall, regardless of the weight, or lack of.

    So, we can be confident that the war weight of the full roster is used, and check boxes contained within that roster break down choice will adjust the search to encompass parameters that are triggered by that break down.
    Recent War Performance is another factor used, according to SC, but I have never managed to find any link or pattern that is governed by this factor.

    A long winded way of saying that parameters may not widen I guess, but rather a combination of matching criteria coming together to mean that you are either matching in a deep pool with a common roster, or a shallower search pool with a more uncommon one.
    It is highly possible that the search parameters widen, but other factors would have to align in order for it to be of any value.

    Balanced bases in a balanced roster is always best for a good match.. That way it doesnt matter.
    Last edited by joshsgrandad; August 13th, 2019 at 11:26 AM.

  4.   This is the last staff post in this thread.   #4
    SharkyFinn's Avatar
    Join Date
    Mar 2014
    Location
    Everywhere. Always watching.
    Posts
    10,384
    There's no need to stop the search. Players siezed on a benign comment made a long time ago about war matchmaking, attributed far too much to it, and created the myth that you should stop your search regularly. Matchmaking and the playerbase has changed so much from those days, even if there was ever a kernel of truth to the idea that stopping the search caused better results.

    The progress of the myth is very much like the progress of other known myths (you ned to drink 6-8 glasses of water a day, you'll kill someone if you drop a penny off the Empire State Building, don't touch a baby bird or its mother will abandon it, and knuckle cracking causes arthritis). The reasons myths last is because each takes a common need, greatly magnifies a tiny bit of fact, treats shaky anecdotal information as actual proof, and makes conclusions that most can't (or won't) verify.

    You won't find any gain in stopping and starting. All you're doing is delaying your wars.
    SharkyFinn.com Official Web Site

    SharkyFinn.com Official Web Site -+- SharkBite YouTube Channel
    USAR (#2GR9YCGR) - Level 19 - 673 Clan War wins
    Private Message - + - Forum Rules - + - Fan Content Policy

  5. #5
    Forum Elder
    Join Date
    Jul 2017
    Location
    NE England, UK.
    Posts
    2,727
    Quote Originally Posted by joshsgrandad View Post
    I had time to think about this, so changing my post..
    It is popular for people to accept that the search parameters widen the longer a search runs, but..

    We know that a war match is made using the full roster..
    A common break down will match quickly.. But does an uncommon one widen its scope in order to get a hit?
    For that to be the case, then your opposition would also need to have been searching for a similar length of time, otherwise it would simply dismiss matching you, since those skewed parameters would not be available for their match.

    A quick match for one side against a clan with widened parameters would lead to a mis match.

    There is more going on than simply adding the offence and defence of two rosters, then looking for a similar number.

    Each individual account will have some bearing on what is being sought..
    For instance, I was running 8s on top, and decided to jump the halls on each account.
    Nothing else was upgraded, and the hall carries no weight, so the halls simply changed colour.
    But, I started to draw 9s at the top.
    My conclusion from this, which was pre suggested by some prominent engineers, is that if you carry a "colour" at the top, you are likely to draw that colour hall, regardless of the weight, or lack of.

    So, we can be confident that the war weight of the full roster is used, and check boxes contained within that roster break down choice will adjust the search to encompass parameters that are triggered by that break down.
    Recent War Performance is another factor used, according to SC, but I have never managed to find any link or pattern that is governed by this factor.

    A long winded way of saying that parameters may not widen I guess, but rather a combination of matching criteria coming together to mean that you are either matching in a deep pool with a common roster, or a shallower search pool with a more uncommon one.
    It is highly possible that the search parameters widen, but other factors would have to align in order for it to be of any value.

    Balanced bases in a balanced roster is always best for a good match.. That way it doesnt matter.

    As an edit to this.. I matched today in 25 minutes.. Same roster, same result.
    Maybe further proof that the roster breakdown is simply activating triggers/check boxes, rather than waiting for the search to widen.
    The more common the bases and roster, the quicker and better the match.. The less common the bases and roster, the wider the "apparent" match will be.. although in reality it will be "fair" in the eyes of the MMA, according to the parameters/checks contained within the search engine.. otherwise it will likely just throw out the search and a "match not found" message will be displayed after 24 hours.
    Still cant make heads or tails of recent war performance though.. Done 11 wars, scored 11 maximums, and nothing looks any different.

  6. #6
    Forum Veteran Thegreatpuma's Avatar
    Join Date
    Mar 2016
    Location
    Druidia
    Posts
    1,675
    Quote Originally Posted by joshsgrandad View Post
    As an edit to this.. I matched today in 25 minutes.. Same roster, same result.
    Maybe further proof that the roster breakdown is simply activating triggers/check boxes, rather than waiting for the search to widen.
    The more common the bases and roster, the quicker and better the match.. The less common the bases and roster, the wider the "apparent" match will be.. although in reality it will be "fair" in the eyes of the MMA, according to the parameters/checks contained within the search engine.. otherwise it will likely just throw out the search and a "match not found" message will be displayed after 24 hours.
    Still cant make heads or tails of recent war performance though.. Done 11 wars, scored 11 maximums, and nothing looks any different.
    I dunno. For the third war in a row we have matched up against a TH7 at the top. Search this time was the longest it's been - took 13 hours. Average search time over these three wars was 7 hours per war. Last war it took 7 hours and we matched a mid TH7 with maxed king and dragons.

    We roll with this:
    3x fully maxed TH6
    2x TH4 not maxed

    Now, don't get me wrong, I have been able to 3* the TH7 bases that have been presented to me but this is mostly because
    (1) They are almost exclusively bad layouts
    (2) I have access to max hogs
    (3) The disco wizard and lack of defending CC troops

    We also still win these wars because we're facing mostly inexperienced opponents. So I'm not complaining at all just pointing out that if everything was otherwise equal, the last 3 wars we were at a decent disadvantage, and should have lost, probably by multiple stars (it's quite difficult for a TH6 to 2* a properly built TH7 with defending CC troops). The only real reason I can't point to the length of search is that it could be something as mundane as our win streak (12 and counting), having 3 maxed bases at the top, or any other number of known and unknown factors.
    Last edited by Thegreatpuma; August 13th, 2019 at 01:15 PM.
    TMBG War Clan always recruiting - Wars with TH8-12
    TMBG Discord server:
    https://discord.gg/aXdgW6w
    TH10 - Doctor Worm permamax (for now)
    TH11 - doug 50/50/20 heroes

  7. #7
    Forum Elder
    Join Date
    Jul 2017
    Location
    NE England, UK.
    Posts
    2,727
    Quote Originally Posted by Thegreatpuma View Post
    I dunno. For the third war in a row we have matched up against a TH7 at the top. Search this time was the longest it's been - took 13 hours. Average search time over these three wars was 7 hours per war. Last war it took 7 hours and we matched a mid TH7 with maxed king and dragons.

    We roll with this:
    3x fully maxed TH6
    2x TH4 not maxed

    Now, don't get me wrong, I have been able to 3* the TH7 bases that have been presented to me but this is mostly because
    (1) They are almost exclusively bad layouts
    (2) I have access to max hogs
    (3) The disco wizard and lack of defending CC troops

    We also still win these wars because we're facing mostly inexperienced opponents. So I'm not complaining at all just pointing out that if everything was otherwise equal, the last 3 wars we were at a decent disadvantage, and should have lost, probably by multiple stars (it's quite difficult for a TH6 to 2* a properly built TH7 with defending CC troops). The only real reason I can't point to the length of search is that it could be something as mundane as our win streak (12 and counting), having 3 maxed bases at the top, or any other number of known and unknown factors.
    yeah mate.. Bit of a spanner in my logic there.
    I have read 5v is sketchy, but shouldnt really matter if my theory held.
    Do those max 6s not offset that 7?
    Even tho the difficulty can be huge, the weight differential may not be so significant, if the rest of their roster were "paying" for that 7.

    certainly food for thought.
    I use 2 anchor 9s, with a third 9 sitting below playing cover.
    However, only 1 of those anchors carries bows, other is just big 9.
    rest of the roster simply tapers down the weight distribution, to allow those top 3 breathing room.
    I have drawn similar in the 5 spins since I set up that roster.

    i fully agree with the reasons for winning though.. Noob players, and carrying max cc troops.
    Although am simply looking to max for now, dont really care what they score or building up a streak.
    Will likely push from 10v to 15v once the perfects are broken, as the bigger wars may show up any variables better.

  8. #8
    Forum Veteran Thegreatpuma's Avatar
    Join Date
    Mar 2016
    Location
    Druidia
    Posts
    1,675
    Do the 6s offset? i dunno. Generally the 2nd and third bases are equal or stronger than their mirrors, and the th4s are usually outmatched by their mirrors. In all cases, the opponent base 2 would be able to 3* any of the top three bases, assuming proper attacks and cc troops.

    there have been cases where my third base won the war. i intentionally keep the most difficult base in the third spot in the lineup as any th7 usually just attacks the top 2 bases right away, rather than going bottom up like they should.

    But really all this is moot. the reason i say this is that if I was working both sides on these wars, my current clan would have lost all three wars by at least one star, in some cases 2. This same should hold true for two inexperienced opponents.
    Last edited by Thegreatpuma; August 13th, 2019 at 02:36 PM.

  9. #9
    Forum Elder
    Join Date
    Jul 2017
    Location
    NE England, UK.
    Posts
    2,727
    Quote Originally Posted by Thegreatpuma View Post
    Do the 6s offset? i dunno. Generally the 2nd and third bases are equal or stronger than their mirrors, and the th4s are usually outmatched by their mirrors. In all cases, the opponent base 2 would be able to 3* any of the top three bases, assuming proper attacks and cc troops.

    there have been cases where my third base won the war. i intentionally keep the most difficult base in the third spot in the lineup as any th7 usually just attacks the top 2 bases right away, rather than going bottom up like they should.

    But really all this is moot. the reason i say this is that if I was working both sides on these wars, my current clan would have lost all three wars by at least one star, in some cases 2. This same should hold true for two inexperienced opponents.
    Yeah, I know.. But matching can only be numbers plus modifiers, since its basically maths.
    Its those modifiers I am trying to find.
    If one could break the code, then losing wars would be a thing of the past.

    Of course, I am finding that simply by playing mini's losing is a thing of the past anyway lol.

  10. #10
    Forum Veteran Thegreatpuma's Avatar
    Join Date
    Mar 2016
    Location
    Druidia
    Posts
    1,675
    Quote Originally Posted by joshsgrandad View Post
    Yeah, I know.. But matching can only be numbers plus modifiers, since its basically maths.
    Its those modifiers I am trying to find.
    If one could break the code, then losing wars would be a thing of the past.

    Of course, I am finding that simply by playing mini's losing is a thing of the past anyway lol.
    understood

    would be nice if we could figure out how much leeway the mm allows at the top based on corresponding offset lower. certainly we've never matched an 8 at the top, but would it be possible? seems to me it should be (maybe a heavily rushed 8), but if not, why? any chance army camps are considered?
    Last edited by Thegreatpuma; August 13th, 2019 at 03:10 PM.

Posting Permissions

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