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

The Match Making

22 comments in this topic

Recommended Posts

16
[LRM3]
Members
281 posts
16,064 battles

Good evening at all.
As you see the title of the post concerns the MM.
There are things I do not understand.

1) 2 Tier too.
You've seen that, almost always, there are clashes with tier far above the tier you entered into the game.
I can understand that if you are basing the ship in the middle (for example: a Tier 6 is between Tier 5 and Tier 7) the difference is a Tier only, but for ships with a lower Tier the clashes become just a frustration.
Now, could not you enter the code of the program that in the presence of low tiers the MM should favor these and thus reduce Tier upper input? I am not saying that you eliminate the 2 Tier tolerance, but only reduce the number of high-tier ships.
From what I read in chat every time that there are div divs with Tier significantly lower than the amount of ships with Tier higher the defeat is almost certain.

2) What is the criterion for sorting players between the teams?
I mean, by the way, it's certainly happened to get into a game, lose it, and eventually see that the "enemy" has, for example, high-level players. Could it not be possible even here to enter a code that at least adds the PR of the deployments? Yes, maybe it will take a few seconds to get into the game but at least it will be a fun match and will not win or lose with 9/10 different ships. What a cabbage !!

3) Why is there a difference between ships?
This too is incomprehensible.

How many times have you got 3 or 4 ships in the clearing (opponent or in your team) with, for example, radar. Then 4 radar ships and your only one clearing. In short, I mean, it is clear that most DDs can be easily tested.

Now, imagine all is 3 combinations together. Ah, no, you do not have to imagine them, these things always happen in MM. But does anyone think of solving these problems that I have, ever since I remember?
Thank you

 

Versione Italiana.

Buona sera a tutti.
Come vedete il titolo del post riguarda il MM.
Ci sono delle cose che non comprendo.

1) 2 Tier di troppo.
Avrete visto che, quasi sempre, vi sono scontri con tier nettamente supeirori rispetto al tier con cui siete entrati nella partita.
Posso capire che se si prende di base la nave che sta nel mezzo (ad esempio: un Tier 6 sta fra il Tier 5 e il Tier 7) la differenza è di un Tier solamente, ma per le navi con un Tier inferiore gli scontri diventano solo una frustrazione.
Ora, non si potrebbe inserire nel codice del programma che in presenza di Tier bassi il MM deve favorire questi ultimi e quindi ridurre l'ingresso di Tier superiori? Non sto dicendo di eliminare la tolleranza di 2 Tier, ma solo di ridurre il numero di navi con tier elevato.
Da quello che leggo in chat tutte le volte che in squadra ci sono div con Tier nettamente inferiore rispetto alla quantità di navi con Tier superiore la sconfitta è quasi sicura.

2) Qual è il criterio di smistamento giocatori fra le squadre?
Voglio dire, dai, vi è capitato di sicuro di entrare in una partita, perderla, e alla fine vedere che il "nemico" ha, ad esempio, giocatori con livelli alti. Possibile che anche qui non si possa inserire un codice che sommi quantomeno i PR degli schieramenti? Si, magari ci vorrà qualche secondo in più per entrare in game ma almeno sarà una partita divertente e non vincere o perdere con 9/10 navi di differenza. Che cavolo!!

3) Perchè vi è una differenza di potenziale fra navi?
Anche questa è una cosa incomprensibile.
Quante volte vi è capitato di avere 3 o 4 navi nello schiaramento (avversario o nel tuo team) con, ad esempio, radar. Quindi 4 navi con radar e nel vostro schiaramento una soltanto. Insomma, voglio dire, è chiaro che la maggior parte dei DD può essere detettata con estrema facilità.

Ora, immaginate tutte è 3 le combinazioni insieme. Ah, no, non dovete immaginarle, queste cose accadono sempre nel MM. Ma qualcuno pensa di risolvere questi problemi che ci sono, che io ricordi, da sempre?
Grazie 

Share this post


Link to post
Share on other sites
623
[NMKJT]
Members
2,728 posts
1 hour ago, Ombromondo said:

Good evening at all.
As you see the title of the post concerns the MM.
There are things I do not understand.

1) 2 Tier too.
You've seen that, almost always, there are clashes with tier far above the tier you entered into the game.
I can understand that if you are basing the ship in the middle (for example: a Tier 6 is between Tier 5 and Tier 7) the difference is a Tier only, but for ships with a lower Tier the clashes become just a frustration.
Now, could not you enter the code of the program that in the presence of low tiers the MM should favor these and thus reduce Tier upper input? I am not saying that you eliminate the 2 Tier tolerance, but only reduce the number of high-tier ships.
From what I read in chat every time that there are div divs with Tier significantly lower than the amount of ships with Tier higher the defeat is almost certain.

2) What is the criterion for sorting players between the teams?
I mean, by the way, it's certainly happened to get into a game, lose it, and eventually see that the "enemy" has, for example, high-level players. Could it not be possible even here to enter a code that at least adds the PR of the deployments? Yes, maybe it will take a few seconds to get into the game but at least it will be a fun match and will not win or lose with 9/10 different ships. What a cabbage !!

3) Why is there a difference between ships?
This too is incomprehensible.

How many times have you got 3 or 4 ships in the clearing (opponent or in your team) with, for example, radar. Then 4 radar ships and your only one clearing. In short, I mean, it is clear that most DDs can be easily tested.

Now, imagine all is 3 combinations together. Ah, no, you do not have to imagine them, these things always happen in MM. But does anyone think of solving these problems that I have, ever since I remember?
Thank you

 

 

1: There are benefits to playing bottom tier. You earn more XP against higher tier ships than equal or lower tier.  There is more freedom to play your game when you are bottom tier. As top tier, you have to contribute for the team to win. Bottom tier, there is no pressure. You have more freedom. And if you are in a DD bottom tier matters less, DDs punch up very well

2: Your team has just as many high level players as the other team. If you compare the two teams WOWS is good at fielding comparable teams. Try and find a similar game that does this better. 

3: There are different ships because MM works with what is in queue. If you are questioning ship balance that is another topic. WG is always working to balance ships and lines better. As far as Radar ships, the MM does not look at the consumables loaded, that is not a criteria, and even if a ship can carry radar how do you know it is carrying radar? Not all cruisers load it. And radar is something you can play around anyway. I say this as a DD player

  • Cool 1

Share this post


Link to post
Share on other sites
1,267
[SEP]
Members
3,723 posts
14,979 battles
6 minutes ago, Destroyer_Suzukaze said:

2: Your team has just as many high level players as the other team. If you compare the two teams WOWS is good at fielding comparable teams. Try and find a similar game that does this better. 

 

 

  I think OP is talking about skill based MM.

Share this post


Link to post
Share on other sites
623
[NMKJT]
Members
2,728 posts
Just now, m373x said:

 

  I think OP is talking about skill based MM.

Skill based MM was Team Battles, which was not popular. In WG products, Randoms is just that. Same with any WG product, different modes may match on skills, never random battles. 

Share this post


Link to post
Share on other sites
2,549
[PSP]
Members
6,251 posts
8,888 battles
Quote

Matchmaker, Matchmaker,
Make me a match,
Find me a find,
catch me a catch
Matchmaker, Matchmaker
Look through your book,
And make me a perfect match

 

 

Share this post


Link to post
Share on other sites
Members
3,470 posts
6,170 battles

2) What is the criterion for sorting players between the teams?


I mean, by the way, it's certainly happened to get into a game, lose it, and eventually see that the "enemy" has, for example, high-level players. Could it not be possible even here to enter a code that at least adds the PR of the deployments? Yes, maybe it will take a few seconds to get into the game but at least it will be a fun match and will not win or lose with 9/10 different ships. What a cabbage !!

Id this is skill based MM - no one can define what skill means.  Numbers from websites are just one persons calculation of a skill that really doe snot have a huge level of confidence.  I have been unicum several times - but only for a day or 2.  I can play great for a game or 2 or even a day then play like crap.  MM can't determine who will play great and then make up a team.

Share this post


Link to post
Share on other sites
699
[OO7]
Members
2,092 posts
8,186 battles
7 hours ago, Ombromondo said:

Good evening at all.
As you see the title of the post concerns the MM.
There are things I do not understand.

1) 2 Tier too.
You've seen that, almost always, there are clashes with tier far above the tier you entered into the game.
I can understand that if you are basing the ship in the middle (for example: a Tier 6 is between Tier 5 and Tier 7) the difference is a Tier only, but for ships with a lower Tier the clashes become just a frustration.
Now, could not you enter the code of the program that in the presence of low tiers the MM should favor these and thus reduce Tier upper input? I am not saying that you eliminate the 2 Tier tolerance, but only reduce the number of high-tier ships.
From what I read in chat every time that there are div divs with Tier significantly lower than the amount of ships with Tier higher the defeat is almost certain.

2) What is the criterion for sorting players between the teams?
I mean, by the way, it's certainly happened to get into a game, lose it, and eventually see that the "enemy" has, for example, high-level players. Could it not be possible even here to enter a code that at least adds the PR of the deployments? Yes, maybe it will take a few seconds to get into the game but at least it will be a fun match and will not win or lose with 9/10 different ships. What a cabbage !!

3) Why is there a difference between ships?
This too is incomprehensible.

How many times have you got 3 or 4 ships in the clearing (opponent or in your team) with, for example, radar. Then 4 radar ships and your only one clearing. In short, I mean, it is clear that most DDs can be easily tested.

Now, imagine all is 3 combinations together. Ah, no, you do not have to imagine them, these things always happen in MM. But does anyone think of solving these problems that I have, ever since I remember?
Thank you

 

Versione Italiana.

Buona sera a tutti.
Come vedete il titolo del post riguarda il MM.
Ci sono delle cose che non comprendo.

1) 2 Tier di troppo.
Avrete visto che, quasi sempre, vi sono scontri con tier nettamente supeirori rispetto al tier con cui siete entrati nella partita.
Posso capire che se si prende di base la nave che sta nel mezzo (ad esempio: un Tier 6 sta fra il Tier 5 e il Tier 7) la differenza è di un Tier solamente, ma per le navi con un Tier inferiore gli scontri diventano solo una frustrazione.
Ora, non si potrebbe inserire nel codice del programma che in presenza di Tier bassi il MM deve favorire questi ultimi e quindi ridurre l'ingresso di Tier superiori? Non sto dicendo di eliminare la tolleranza di 2 Tier, ma solo di ridurre il numero di navi con tier elevato.
Da quello che leggo in chat tutte le volte che in squadra ci sono div con Tier nettamente inferiore rispetto alla quantità di navi con Tier superiore la sconfitta è quasi sicura.

2) Qual è il criterio di smistamento giocatori fra le squadre?
Voglio dire, dai, vi è capitato di sicuro di entrare in una partita, perderla, e alla fine vedere che il "nemico" ha, ad esempio, giocatori con livelli alti. Possibile che anche qui non si possa inserire un codice che sommi quantomeno i PR degli schieramenti? Si, magari ci vorrà qualche secondo in più per entrare in game ma almeno sarà una partita divertente e non vincere o perdere con 9/10 navi di differenza. Che cavolo!!

3) Perchè vi è una differenza di potenziale fra navi?
Anche questa è una cosa incomprensibile.
Quante volte vi è capitato di avere 3 o 4 navi nello schiaramento (avversario o nel tuo team) con, ad esempio, radar. Quindi 4 navi con radar e nel vostro schiaramento una soltanto. Insomma, voglio dire, è chiaro che la maggior parte dei DD può essere detettata con estrema facilità.

Ora, immaginate tutte è 3 le combinazioni insieme. Ah, no, non dovete immaginarle, queste cose accadono sempre nel MM. Ma qualcuno pensa di risolvere questi problemi che ci sono, che io ricordi, da sempre?
Grazie 

1.) You cannot reduce the number of high tier ships/players in the queue. Its just your rotten luck you hit “battle” the same time they did.

2.) MM takes into account ONLY your ships type and tier. Nothing else

3.) Wat? 

 

And why did you post it in italian? 

Share this post


Link to post
Share on other sites
16
[LRM3]
Members
281 posts
16,064 battles
6 hours ago, FireAndHEspam said:

1.) You cannot reduce the number of high tier ships/players in the queue. Its just your rotten luck you hit “battle” the same time they did.

2.) MM takes into account ONLY your ships type and tier. Nothing else

3.) Wat? 

 

And why did you post it in italian? 

Because many do not understand how to translate google and for security I put it in Italian, since I'm Italian.

Share this post


Link to post
Share on other sites
16
[LRM3]
Members
281 posts
16,064 battles
7 hours ago, CylonRed said:

 

 

Id this is skill based MM - no one can define what skill means.  Numbers from websites are just one persons calculation of a skill that really doe snot have a huge level of confidence.  I have been unicum several times - but only for a day or 2.  I can play great for a game or 2 or even a day then play like crap.  MM can't determine who will play great and then make up a team.

this is true but could be taken into account in order to avoid huge differences in clashes, which is always repeated

Share this post


Link to post
Share on other sites
16
[LRM3]
Members
281 posts
16,064 battles
12 hours ago, Destroyer_Suzukaze said:

 

1: There are benefits to playing bottom tier. You earn more XP against higher tier ships than equal or lower tier.  There is more freedom to play your game when you are bottom tier. As top tier, you have to contribute for the team to win. Bottom tier, there is no pressure. You have more freedom. And if you are in a DD bottom tier matters less, DDs punch up very well

2: Your team has just as many high level players as the other team. If you compare the two teams WOWS is good at fielding comparable teams. Try and find a similar game that does this better. 

3: There are different ships because MM works with what is in queue. If you are questioning ship balance that is another topic. WG is always working to balance ships and lines better. As far as Radar ships, the MM does not look at the consumables loaded, that is not a criteria, and even if a ship can carry radar how do you know it is carrying radar? Not all cruisers load it. And radar is something you can play around anyway. I say this as a DD player

1) uhm, it might be an interesting consideration.
2) False. Who tells you that my or the other team has high level players
3) I thought WOW was a program, so based on computer-readable data very quickly, I did not think I should go asking for ship info

Share this post


Link to post
Share on other sites
Members
3,470 posts
6,170 battles
3 hours ago, Ombromondo said:

this is true but could be taken into account in order to avoid huge differences in clashes, which is always repeated

What can be taken into consideration?  What is "it"?  Define the skill and how it would be defined as it would have to be coded in some way.  Without a definition on what 'skill' is - nothing can be done.  Understand win rates can be manipulated in several ways.  So by going on win rate alone is not a good idea.

Share this post


Link to post
Share on other sites
16
[LRM3]
Members
281 posts
16,064 battles
1 hour ago, CylonRed said:

What can be taken into consideration?  What is "it"?  Define the skill and how it would be defined as it would have to be coded in some way.  Without a definition on what 'skill' is - nothing can be done.  Understand win rates can be manipulated in several ways.  So by going on win rate alone is not a good idea.

in fact I'm not talking about win rate but PR

Share this post


Link to post
Share on other sites
Members
3,470 posts
6,170 battles
1 hour ago, Ombromondo said:

in fact I'm not talking about win rate but PR

What is PR?  My guess it is another version of WTR - that Warships Today uses.

Rating calculations:

This is a made up calculation based on criteria that someone created.  It is - at best - a complete guess to try and define some sort of 'skill' rating.  Problems is still - none of the ratings out there knows how one person, much less 2 teams of people, will play in a game.  And that is the problem.

Share this post


Link to post
Share on other sites
Members
92 posts
1,880 battles
19 hours ago, Destroyer_Suzukaze said:

 

1: There are benefits to playing bottom tier. You earn more XP against higher tier ships than equal or lower tier.  There is more freedom to play your game when you are bottom tier. As top tier, you have to contribute for the team to win. Bottom tier, there is no pressure. You have more freedom. And if you are in a DD bottom tier matters less, DDs punch up very well

2: Your team has just as many high level players as the other team. If you compare the two teams WOWS is good at fielding comparable teams. Try and find a similar game that does this better. 

3: There are different ships because MM works with what is in queue. If you are questioning ship balance that is another topic. WG is always working to balance ships and lines better. As far as Radar ships, the MM does not look at the consumables loaded, that is not a criteria, and even if a ship can carry radar how do you know it is carrying radar? Not all cruisers load it. And radar is something you can play around anyway. I say this as a DD player

Well said. Also, there is no reason why a ship on the bottom of a 2 tier spread can't get first on a well played battle.

Share this post


Link to post
Share on other sites
16
[LRM3]
Members
281 posts
16,064 battles
59 minutes ago, CylonRed said:

What is PR?  My guess it is another version of WTR - that Warships Today uses.

Rating calculations:

This is a made up calculation based on criteria that someone created.  It is - at best - a complete guess to try and define some sort of 'skill' rating.  Problems is still - none of the ratings out there knows how one person, much less 2 teams of people, will play in a game.  And that is the problem.

I'm talking about PR. If you do not know him, take a look here?
put your name and see how it calculates it..

 

https://na.wows-numbers.com/

Share this post


Link to post
Share on other sites
16
[LRM3]
Members
281 posts
16,064 battles
19 minutes ago, SkeletonDenial said:

Well said. Also, there is no reason why a ship on the bottom of a 2 tier spread can't get first on a well played battle.

Of course, everyone sees things as he wants to see them, even for me. but I do not quite agree as I have specified above.

Share this post


Link to post
Share on other sites
Members
3,470 posts
6,170 battles
3 minutes ago, Ombromondo said:

I'm talking about PR. If you do not know him, take a look here?
put your name and see how it calculates it..

 

https://na.wows-numbers.com/

As I said - it is a made up calculation just like WTR.

 

This is a made up calculation based on criteria that someone created.  It is - at best - a complete guess to try and define some sort of 'skill' rating.  Problems is still - none of the ratings out there knows how one person, much less 2 teams of people, will play in a game.  And that is the problem.

The above is STILL valid and true.

Edited by CylonRed

Share this post


Link to post
Share on other sites
16
[LRM3]
Members
281 posts
16,064 battles
2 minutes ago, CylonRed said:

As I said - it is a made up calculation just like WTR.

Ineffects in PR calculation are added to that of battles, not meaning to me, but so is it. In any case, a value is to be able to compare something.

Share this post


Link to post
Share on other sites
Members
3,470 posts
6,170 battles

In any case, a value is to be able to compare something.

Something????  This is the problem - compare to WHAT and is that comparison even valid?  Again - and for the last time - it is a made up calculation just like WTR.  Completely and totally made up.  It does not convey any amount of factual info on how any one person will play in any one game and as such - is not valid for use in MM currently.  Again - MM can't know the future and that is the problem.

Share this post


Link to post
Share on other sites
16
[LRM3]
Members
281 posts
16,064 battles
18 minutes ago, CylonRed said:

 

 

Something????  This is the problem - compare to WHAT and is that comparison even valid?  Again - and for the last time - it is a made up calculation just like WTR.  Completely and totally made up.  It does not convey any amount of factual info on how any one person will play in any one game and as such - is not valid for use in MM currently.  Again - MM can't know the future and that is the problem.

But are you or do you do it?
I have not said that is perfect. I'm saying that since there are no parameters you could use a parameter like PR or Wr, choose what you think of yourself or even another. But as I think it is not good, because it only evaluates ships and not the abilities (in short, some capacity) of the opponent. Obviously, this leads to imbalances in the final game.

Share this post


Link to post
Share on other sites
764
[ARS]
Beta Testers
2,244 posts
1,826 battles
59 minutes ago, Ombromondo said:

Ineffects in PR calculation are added to that of battles, not meaning to me, but so is it. In any case, a value is to be able to compare something.

The things that contribute to victory are often not the things tracked, or not the things given most weight in those rating systems.

 

Currently WoWS does not try to do any skill based matchmaking in Random Battles.  It looks at type of ship and tier of ship, making sure each team is as close on those as possible given what is in the queue and how long they've waited. 

 

(If you ever get in a battle with hardly any ships, but an aircraft carrier on both sides it means one of those carriers waited in queue for more than 5 minutes and the other just queued.  Matchmaker just grabbed what was available because queue times are not supposed to exceed 5 minutes, so it had to do something for the waiting CV.)

Share this post


Link to post
Share on other sites
16
[LRM3]
Members
281 posts
16,064 battles
22 hours ago, Helstrem said:

The things that contribute to victory are often not the things tracked, or not the things given most weight in those rating systems.

 

Currently WoWS does not try to do any skill based matchmaking in Random Battles.  It looks at type of ship and tier of ship, making sure each team is as close on those as possible given what is in the queue and how long they've waited. 

 

(If you ever get in a battle with hardly any ships, but an aircraft carrier on both sides it means one of those carriers waited in queue for more than 5 minutes and the other just queued.  Matchmaker just grabbed what was available because queue times are not supposed to exceed 5 minutes, so it had to do something for the waiting CV.)

I understand, but you make that without a parameter other than that of the ships, in the end, the quality of the teams is rarely balanced. This is a problem that I think is frustrating, I say in time.
I happen to lose badly, in the sense we stay in 1 or 2 ships and on the other side there are 9. You know that there is a problem of quality players. This naturally reflects on both PR and WR. The PR would be a parameter that could help to avoid these problems.
I also note that the waiting queue should not exceed 5 minutes, but I think it would be desirable to do extra effort on the part of the program. Of course, only programmers can know if it is feasible or not.
Balancing would also lead to the acceptance of missed battles. Why in the end if you lose 3 vs. 2 you accept it, but if you lose 9 against 2 you know well that the problem is the team.

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

×