Jump to content
You need to play a total of 5 battles to post in this section.
LowSpeed_US

Chance or Coincidence?

42 comments in this topic

Recommended Posts

203
[NSEW]
Members
1,101 posts

Hello Captains,

 

Has anyone given thoughts to how your ship is calculated into the match making algorithm when hoisting these special signals?

i.e. Scylla, Hydra, Ouroboros, Leviathan, Dragon.

I've noticed a significant chance (or coincidence) that whenever these special signals are used, the match maker goes into hyperdrive mode of prove it.

 

What are your experiences?

 

 

  • Cool 3

Share this post


Link to post
Share on other sites
Members
2,121 posts
9,978 battles

I save alot of those flags for 100%/200% xp bonus times. Last time I used a few was this weekend on my Hipper. It did place me in a t10 match. I finished with 2.7k base XP top score as the lone tier 8. Maybe MM said prove it, and I was like. Sure.

Share this post


Link to post
Share on other sites
533
[KSD]
Members
933 posts
5,798 battles
5 minutes ago, LowSpeed_US said:

Hello Captains,

 

Has anyone given thoughts to how your ship is calculated into the match making algorithm when hoisting these special signals?

i.e. Scylla, Hydra, Ouroboros, Leviathan, Dragon.

I've noticed a significant chance (or coincidence) that whenever these special signals are used, the match maker goes into hyperdrive mode of prove it.

 

What are your experiences?

 

 

yes. every time i take those signals in a random battle, I am on the worst team possible. I have stopped using signals in random battles just for the sake of it and only use signals in clan battles or ranked battles.

WG will never admit it, but i just won 40% of my matches in the past 24 hours whereas my avergage is 53%. true I was playing the missouri, I suck at BBs, but still I got stuck on teams that threw a game aaway. but whatever, winning or losing is part and parcel of the game. come again and play tomorrow.

I am on the last mission of obtaining Halsey, and i needed to get 50k XP as fast as possible. havent been able to do it in 15 games as I am stuck on the worst teams. even had a match where enemy team didnt lose a single ship. sure that means I have a chance of farming damage and crap, but not if literally a flank of my team just suicided and now getting cross shots from every side of the map. that also affected my PR and now with a rating of average instead of good or very good....

Edited by lungiwear

Share this post


Link to post
Share on other sites
1,712
[WOLF1]
Beta Testers
8,239 posts
11,685 battles

It is just the ship class, tier and frequency of uptiering. That is it. MM does not know about you, or if you are stock or elite , or what signals your are flying. 

  • Cool 1

Share this post


Link to post
Share on other sites
879
[N-W-T]
Members
1,545 posts
6,386 battles

You think the matchmaker is deliberately placing you on bad teams/bad MM because of the flags you're flying? Lol the stuff some of you guys come up with.

  • Cool 3
  • Funny 2

Share this post


Link to post
Share on other sites
988
[R-F]
Members
1,382 posts
8,358 battles

For everyone with a wacky conspiracy theory about the matchmaker: These things can be conclusively proven given sufficient data.  If you think you're on to something, the burden of proof lies with you.  Go out, log some data!  Needs to be a lot of data.  If I suddenly only win 8 out of 20 battles matchmaker doesn't have it in for me.  If I win 80 out of the next 200 than maybe something's up.

For this particular theory regarding special signals, people who put down serious money on santa crates should have piles of them at this point.  So it should be doable.

 

One thing to consider: How many combat flags do you normally run?  If I normally run 6 combat flags and 2 economic flags, but now I have all these special flags so I run 1 combat flag and 7 special flags, I am setting sail with a weaker ship than I normally do, and over a large enough sample size I should expect to see decreased performance as a result.

  • Cool 2

Share this post


Link to post
Share on other sites
2,401
[YORHA]
Members
4,201 posts
7,376 battles

So use them on tier X.  Worth it!

ob9tAhd.jpg

 

Although they can be pretty juicy at tier VIII as well (Helena battle)...

 

9IKZmOl.jpg

Edited by JCC45

Share this post


Link to post
Share on other sites
368
[CSALT]
Members
966 posts
54 minutes ago, lungiwear said:

yes. every time i take those signals in a random battle, I am on the worst team possible. I have stopped using signals in random battles just for the sake of it and only use signals in clan battles or ranked battles.

WG will never admit it, but i just won 40% of my matches in the past 24 hours whereas my avergage is 53%. true I was playing the missouri, I suck at BBs, but still I got stuck on teams that threw a game aaway. but whatever, winning or losing is part and parcel of the game. come again and play tomorrow.

I am on the last mission of obtaining Halsey, and i needed to get 50k XP as fast as possible. havent been able to do it in 15 games as I am stuck on the worst teams. even had a match where enemy team didnt lose a single ship. sure that means I have a chance of farming damage and crap, but not if literally a flank of my team just suicided and now getting cross shots from every side of the map. that also affected my PR and now with a rating of average instead of good or very good....

What are you expecting WG to admit?

They already say MM attempts to keep your win rate about 50% and assigns teams to achieve that.  If your above 50%, chances are that MM has placed you with teammates that have a lowwer win rate.  If you're under 50%, your more likely to be placed on a team that will help raise your average.

Share this post


Link to post
Share on other sites
1,075
[OO7]
[OO7]
Members
1,886 posts
6,609 battles

How do people come up with this stuff?  Is Elvis still alive and choosing the matchmaking too?

Share this post


Link to post
Share on other sites
1,796
[WORX]
Members
4,229 posts
15,313 battles

MM, MM, MM, MM, before that was the CV rework. Before that the apocalypses of the Asashio. The problem is when people start believing these fairy tales. Adults, they lie kids...

  • Santa is not real,
  • There is no cookie monster (that is me on a hungry afternoon)
  • There is no boogie man

I feel better, in this case the truth will set you free.

Edited by Navalpride33

Share this post


Link to post
Share on other sites
2,401
[YORHA]
Members
4,201 posts
7,376 battles
20 minutes ago, Gunga_Dinner said:

What are you expecting WG to admit?

They already say MM attempts to keep your win rate about 50% and assigns teams to achieve that.  If your above 50%, chances are that MM has placed you with teammates that have a lowwer win rate.  If you're under 50%, your more likely to be placed on a team that will help raise your average.

 

Going to need a citation on that.

  • Cool 1

Share this post


Link to post
Share on other sites
476
[4HIM]
[4HIM]
Beta Testers
1,524 posts
10,920 battles
1 hour ago, paradat said:

It is just the ship class, tier and frequency of uptiering. That is it. MM does not know about you, or if you are stock or elite , or what signals your are flying. 

I would disagree with your stock statement... I have found that not only in Tanks but in WoWS, a stock tank or ship almost always gets a favorable MM.  I have been playing for seven years and do not think its coincidence, thought the chance is extremely remote that it is.  Just my observations and seems to bear up in talks with other folks.

Share this post


Link to post
Share on other sites
1,796
[WORX]
Members
4,229 posts
15,313 battles
25 minutes ago, Gunga_Dinner said:

What are you expecting WG to admit?

They already say MM attempts to keep your win rate about 50% and assigns teams to achieve that.  If your above 50%, chances are that MM has placed you with teammates that have a lowwer win rate.  If you're under 50%, your more likely to be placed on a team that will help raise your average.

^^^^ I dont see this anywhere a Russian bear from the DEV team has held a Q&A opportunity... This needs a citation or its hearsay/fake news.

Share this post


Link to post
Share on other sites
1,712
[WOLF1]
Beta Testers
8,239 posts
11,685 battles
1 minute ago, Morpheous said:

I would disagree with your stock statement... I have found that not only in Tanks but in WoWS, a stock tank or ship almost always gets a favorable MM.  I have been playing for seven years and do not think its coincidence, thought the chance is extremely remote that it is.  Just my observations and seems to bear up in talks with other folks.

First battle does get favorable MM. I meant in general.

Share this post


Link to post
Share on other sites
368
[CSALT]
Members
966 posts
3 minutes ago, Navalpride33 said:

^^^^ I dont see this anywhere a Russian bear from the DEV team has held a Q&A opportunity... This needs a citation or its hearsay/fake news.

It is detailed in the patent Wargaming has for MM and has been cited numerous times in these threads.

  • Cool 1

Share this post


Link to post
Share on other sites
1,796
[WORX]
Members
4,229 posts
15,313 battles
23 minutes ago, Gunga_Dinner said:

It is detailed in the patent Wargaming has for MM and has been cited numerous times in these threads.

I dont think he was talking about the "patent" it self. That I know.

Up to a certain point, having a very high % can be interpreted as something else like for example: 

  • A Boxer who pads his stats by facing off or facing "soft" competition.

I dont think for a sec MM is restricting people to %50 or close to it. There is a balance affect. In essence, where in order for one group of players to be at a certain percentage than the there is a certain number of the player population that will be lower or higher percentage. What determines this is the win or the loss produced in game. IMO they're two factors in which MM can;t calculate which are 

  1. The Human Factor
  2. The over capable ship factor.

At least with the #2 part of the argument, mirror match making should've resolve that. Its not perfect because factor #3 and #4 comes into play

  • The volume of players in queue,
  • what ships  those players have selected to play.

Now, not every game is going to be a perfect game based on these factors. One thing for sure, many games in the past. Have ceased operation when  implementing skills based MM. But if we HAVE to instill it just to give the community a taste of the harshness of the idea (because IMO I dont think it will not resonate how bad the idea is in the first place). Then its going to be a long ride. :). 

Edited by Navalpride33

Share this post


Link to post
Share on other sites
988
[R-F]
Members
1,382 posts
8,358 battles
24 minutes ago, Gunga_Dinner said:

It is detailed in the patent Wargaming has for MM and has been cited numerous times in these threads.

Just because they hold a patent doesn't mean they're actively applying it.

If they're trying to force me down to a 50% win rate I must be doing better than I thought to stay above it!  Uplifting news, truly. 

  • Funny 1

Share this post


Link to post
Share on other sites
219
[GT99]
Members
596 posts
8,116 battles
53 minutes ago, GE_Capital said:

How do people come up with this stuff?  Is Elvis still alive and choosing the matchmaking too?

What? Everyone knows Elvis is an alien. :Smile_amazed:

Share this post


Link to post
Share on other sites
988
[R-F]
Members
1,382 posts
8,358 battles
14 minutes ago, mohawkdriver said:

I can't prove it, but it seems to me that whenever I hang such signals on my ship, I get wasted almost immediately.

getting wasted almost immediately would be beyond the function of the matchmaker.  Even if the other team has somewhat better players, getting wasted right off the bat is 100% on you - and this is coming from someone who does get wasted immediately sometimes - most recently when I steered my Harugumo right into the enemy Gearing's opening torpedo salvo.

I wonder if there's a psychological aspect on it.  You know you're blowing a lot of special flags so you want to have a monster game, so you push too hard too soon and pay for it.

  • Cool 2

Share this post


Link to post
Share on other sites
Beta Testers
1,611 posts
9,876 battles
1 hour ago, Gunga_Dinner said:

It is detailed in the patent Wargaming has for MM and has been cited numerous times in these threads.

And as has been said in numerous threads both here and in World of Tanks, it is just a patent.  As has been pointed out ad infinitum, companies patent ideas all the time in case they want to implement the idea at some point.  Patenting the idea prevents competitors from using the idea, which would prevent anyone else from doing so.  The fact that there is a patent for something in no way proves it has ever been implemented.  On top of that, the patent isn't even a definitive statement on how they would do it.  It is filled with "we could do this or we could do that," some of which would make no sense if all of it was implemented.  It is just an outline of ways they could do something if they decided to do it.  It is not in any way a declaration of what actually got implemented.

More importantly, there is not one single authoritative statement from WG that it was put in any of their games, and there is no data proving it is.  In fact, there have been people who have published data showing it doesn't work that way at all.  The only "evidence" of it is always anecdotal statements from statistically insignificant (even that is generous), small sample sizes.  These are usually from people with average or below average win rates who cannot accept that they are average to below average players and who are looking for an excuse to explain why they don't have the higher win rates they think they deserve.  More often than not, these same people won't even show their stats so that people can honestly dispute their conclusions (e.g., by respectfully pointing out that their below average damage or below average survival rates might just be the bigger factor).

Even without all that, the patent says absolutely nothing about "trying to keep players at a 50% win rate."  First, that is absolutely unnecessary.  A purely random match maker will push people towards a 50% win rate all by itself.  This is because, over time, randomness will place you on pretty much exactly the same number of winning teams as losing teams, and as a result, everyone is pushed towards a 50% win rate.  Second, if it such a mechanic were in place, it is failing miserably because there many, many people with well below 50% win rates and many who have above 50% win rate.  But even if that were not the case, it very clearly states the purpose and means of doing what it said it would be doing if this was implmented, and nothing in that statement has anything to do with "keeping people at a 50% win rate."

What it very clearly says is that people will get frustrated if they are continually placed in difficult battles or if they are constantly in battles that offer them no challenge.  Thus, the patent proposes ways to combat this.  It suggests the possibility of looking at a player's recent win rate and determine if they are on a significant winning streak or a significant losing streak.  It could then adjust the likelihood of what tier battle they would end up in.  If they are on significantly long winning streak, they end up with a higher chance of being in more battles where they are mid to low tier and less battles where they are at high tier.  Conversely, someone on a significant losing streak would have a higher probability of ending up in more battles where they were mid to high tier and less where they are low tier.  Note that it does not say it would guarantee they would end up in a battle at a specific tier, just that it would change the odds so that they are likely to end up in more of the desired tier battles for a time.

More importantly, no where does it even remotely suggest that it would accomplish this by putting good players on bad teams destined to lose or bad players on good teams destined to win.  That is a lie fabricated by the conspiracy theorists who always assume they are the good player being put on the bad team so that "the man" can keep them down.  Isn't it odd how you never hear an admittedly bad player claim he is getting nothing but winning teams sweeping away teams and feasting on the lamentations of their women despite his own incompetence, driving his win rate undeservedly high?

Please, for the love of all that is good, stop pointing to that patent.  It is meaningless.  Quote an actual WG authority stating it is actually in the Match Maker (not some statement from some random person who is possibly a WG employee working in accounting where he would obviously have the low-down on how match making is programmed).  Either that or show actual data, because the fact that they have some vague statements in a patent on what they could possibly do is hardly any kind of admission or proof said processes ever made it into the game.

Share this post


Link to post
Share on other sites
Beta Testers
1,611 posts
9,876 battles
50 minutes ago, Brhinosaurus said:

Just because they hold a patent doesn't mean they're actively applying it.

If they're trying to force me down to a 50% win rate I must be doing better than I thought to stay above it!  Uplifting news, truly. 

You are truly a gaming god since their puny algorithms can only hold you to the win rate you have.

Share this post


Link to post
Share on other sites
236
[RLGN]
Members
719 posts
5,744 battles
1 hour ago, Gunga_Dinner said:

What are you expecting WG to admit?

They already say MM attempts to keep your win rate about 50% and assigns teams to achieve that.  If your above 50%, chances are that MM has placed you with teammates that have a lowwer win rate.  If you're under 50%, your more likely to be placed on a team that will help raise your average.

 

1 hour ago, JCC45 said:

 

Going to need a citation on that.

Yeah, they have been clear they still have no intention of incorporating skill into the MM algorithm, which is something I wish they would do (like most other competitive multiplayer games), and pretty much the only explanation they ever gave it it would become harder for unicum stats to be as outstanding.

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

×