Jump to content


6<-15 Games


  • Please log in to reply
35 replies to this topic

Blavod #1 Posted 05 February 2018 - 08:18 AM

    Private

  • Player
  • 18790 battles
  • 38
  • [DUDS] DUDS
  • Member since:
    05-05-2011

Can we have a conversation about this? (I'm aware it has probably been discussed before) It seems so much more prevalent recently, and it isn't so great. A few patches ago I remember having many more games where there was an actual fight involved in the games, whereas more recently most games are over quite quickly, and are quite one sided. I don't use XVM, so I don't know if one team is just miles better than the other, but there just seems there is such a disparity in how teams perform, that I can't really fathom why?

I used to enjoy having longer games where it gets really tense towards the end, and there is perhaps some adrenaline going, making the game really fun and rewarding, but getting stomped/stomping as it is now is really just a bit pants. I miss actual fights :izmena:.

 

Your opinions, fellow tankers?



ZlatanArKung #2 Posted 05 February 2018 - 08:32 AM

    Major General

  • Player
  • 1529 battles
  • 5,112
  • Member since:
    12-20-2014
My guess is that it is an effect of the 3-5-7 matchmaker putting most of the power in few tanks (2-3/team). And if one team have a good top tier and the other don't, the outcome is a steamroll.

HidesHisFace #3 Posted 05 February 2018 - 08:35 AM

    Second Lieutenant

  • Player
  • 17849 battles
  • 1,299
  • Member since:
    05-12-2012

View PostBlavod, on 05 February 2018 - 08:18 AM, said:

Can we have a conversation about this? (I'm aware it has probably been discussed before) It seems so much more prevalent recently, and it isn't so great. A few patches ago I remember having many more games where there was an actual fight involved in the games, whereas more recently most games are over quite quickly, and are quite one sided. I don't use XVM, so I don't know if one team is just miles better than the other, but there just seems there is such a disparity in how teams perform, that I can't really fathom why?

I used to enjoy having longer games where it gets really tense towards the end, and there is perhaps some adrenaline going, making the game really fun and rewarding, but getting stomped/stomping as it is now is really just a bit pants. I miss actual fights :izmena:.

 

Your opinions, fellow tankers?

 

Look, games like this are always prone to steamrolls and it depends on numerous factors. I'm not really going to delve into this, because the issue complex as hell but I personally didn't notice any significant change. Without statistics, personal feeling about this do not matter anyway. 

And I suppose you do not have statistics from your last hundred games or so to prove your point?

 

In any case 6-15, as far as I remember is completely statistically NORMAL. Hell, there was even a mathematical law about this, though I forgot the name of it. Basically - the team who scores the first kill generally gains actually quite significant advantage, statistically speaking, and every new kill makes it much more difficult for the other team to pull from the pit - basically creating the steamroll. As unpleasant as it is, it is absolutely bloody normal, and it is assuming that both teams are balanced.


Edited by HidesHisFace, 05 February 2018 - 08:35 AM.


DurtiArry #4 Posted 05 February 2018 - 08:40 AM

    Sergeant

  • Player
  • 58396 battles
  • 281
  • [ANG3L] ANG3L
  • Member since:
    11-27-2012

I do use XVM and the majority of games are orange and yellows .vs. yellows and greens.

The balance is close but not close enough to have a 50/50 chance.

Despite the cries for "proof" and "where is the evidence" ............. - I know what I see !

 

 



ZorzDePaloma #5 Posted 05 February 2018 - 08:47 AM

    Sergeant

  • Player
  • 38301 battles
  • 222
  • [DBWS] DBWS
  • Member since:
    04-02-2013

How many of you guys remember the battles before "improved matchmake"? I'm not talking about this 3-5-7 thing, I'm talking about 4+ years before.

5 minute battles were not a common thing. You often had 7+ m battles, and also 15m battles. Now I get these battles 1/100 because:

1) My team crushes enemy in 3 minutes

2) My team is crushed by enemy in 3 minutes.

Something is wrong here, very wrong: players, maps, mm, op tanks... Something of this, or all of this combined in some amount.



Jigabachi #6 Posted 05 February 2018 - 08:48 AM

    Field Marshal

  • Player
  • 17927 battles
  • 19,064
  • [T-D-U] T-D-U
  • Member since:
    08-12-2011
- Broken gamebalance with many balancebreaking tanks
- MM that puts lots of responsibility and power in the hands of a few players but doesn't distribute those (or all) players based on skill
- Bad mapdesign that doesn't allow for different tactical approaches
- Lots of other reasons

Considering all that, it's not really surprising that you get steamroll most of the time.

kaneloon #7 Posted 05 February 2018 - 08:53 AM

    Lieutenant

  • Player
  • 28125 battles
  • 1,634
  • [OBLIC] OBLIC
  • Member since:
    11-18-2011

It is pretty simple : most of the time 10 to 13 tanks go one side, camp because whaterver reason : no corones, super heavies, invisible swedish tds, arty hammering, etc.

The other team walks on the other part of tve team, and then take them from behind.

Cross fire is often roflstomp.



Dru_UK #8 Posted 05 February 2018 - 08:53 AM

    Staff Sergeant

  • Player
  • 6562 battles
  • 484
  • [SPIKE] SPIKE
  • Member since:
    01-02-2013

View PostDurtiArry, on 05 February 2018 - 08:40 AM, said:

I do use XVM and the majority of games are orange and yellows .vs. yellows and greens.

The balance is close but not close enough to have a 50/50 chance.

Despite the cries for "proof" and "where is the evidence" ............. - I know what I see !

 

 

 

 

so example, I could be a green player, but be in my new t5 leopard, in a T7 game on a city map, but I'm stock tank, got a 75% crew, remind me again why that would give me and my team a massive advantage, over the orange guy on the other side, in a fully decked out tank, crew etc?



evilchaosmonkey #9 Posted 05 February 2018 - 09:01 AM

    Lieutenant

  • Player
  • 16655 battles
  • 1,756
  • [EIGHT] EIGHT
  • Member since:
    05-04-2013

Putting stats aside for a second, it is the streaks of losses (yeah wins aren't a problem) that make people angry about MM.  Partly caused by un-balanced team setups.

 

What WG should look at is a MM that actively checks for the overly random nature of the MM and after a shortish losing streak (where you've been on un-balanced teams for a while), kick in to make your battles a little fairer for a period by balancing say average WR on both teams, then switching off after a while.  Or something like this.

 

No, this won't force everyone to a 50% win rate - in the same way that the completely random one doesn't now.

 

Stats hat back on.  MM is fair but crap.

 

Lots of data / threads have been made showing that the MM over a large number of games is balanced.

From a skill perspective, picking teams at random does indeed produce long runs (10+ games) of one sided teams.

Platoons of good players screw the fairness.

The MM template 3/5/7 screws the fairness.

As HidesHisFace explores, any tanks lost (AFK/Yolo/Suicide/Kill) weight the win chances against you.  Early loses = early screwed game.

Great players can win matches if <50% win chance more often than not.



DurtiArry #10 Posted 05 February 2018 - 09:06 AM

    Sergeant

  • Player
  • 58396 battles
  • 281
  • [ANG3L] ANG3L
  • Member since:
    11-27-2012

View PostDru_UK, on 05 February 2018 - 07:53 AM, said:

 

 

so example, I could be a green player, but be in my new t5 leopard, in a T7 game on a city map, but I'm stock tank, got a 75% crew, remind me again why that would give me and my team a massive advantage, over the orange guy on the other side, in a fully decked out tank, crew etc?

 

A very common occurance for you - 

Top 5 tanks are green on one team, the top 5 tanks are orange on the other.

How is that not an advantage?



Jigabachi #11 Posted 05 February 2018 - 09:09 AM

    Field Marshal

  • Player
  • 17927 battles
  • 19,064
  • [T-D-U] T-D-U
  • Member since:
    08-12-2011

View Postevilchaosmonkey, on 05 February 2018 - 09:01 AM, said:

What WG should look at is a MM that actively checks for the overly random nature of the MM and after a shortish losing streak (where you've been on un-balanced teams for a while), kick in to make your battles a little fairer for a period by balancing say average WR on both teams, then switching off after a while.  Or something like this.

Ugh... I don't think that kind of selective or dynamic skillbased MM would be a good idea.

 

Why not a skillbased MM in general? The problem with these discussions is that people have no idea what that really is when talking about it. They can only think of "you vs. your clone... hurrdurr... 50%!"



clixor #12 Posted 05 February 2018 - 09:12 AM

    Lieutenant Сolonel

  • Player
  • 51483 battles
  • 3,058
  • Member since:
    08-07-2011

View Postevilchaosmonkey, on 05 February 2018 - 09:01 AM, said:

Putting stats aside for a second, it is the streaks of losses (yeah wins aren't a problem) that make people angry about MM.  Partly caused by un-balanced team setups.

 

What WG should look at is a MM that actively checks for the overly random nature of the MM and after a shortish losing streak (where you've been on un-balanced teams for a while), kick in to make your battles a little fairer for a period by balancing say average WR on both teams, then switching off after a while.  Or something like this.

 

No, this won't force everyone to a 50% win rate - in the same way that the completely random one doesn't now.

 

Stats hat back on.  MM is fair but crap.

 

Lots of data / threads have been made showing that the MM over a large number of games is balanced.

From a skill perspective, picking teams at random does indeed produce long runs (10+ games) of one sided teams.

Platoons of good players screw the fairness.

The MM template 3/5/7 screws the fairness.

As HidesHisFace explores, any tanks lost (AFK/Yolo/Suicide/Kill) weight the win chances against you.  Early loses = early screwed game.

Great players can win matches if <50% win chance more often than not.

 

You might be right from a stats perspective, but like you say, 'balance' only works with a large data-set. And as players play like 10 - 30 battles a session, a lot comes down to luck in what teams you are put in.

 

And mostly i think that in essence is what frustrates the most players, and this becomes worse if you more experienced in this game, as you actually see what your team is doing wrong. But what may be obvious for the experienced player, may not be for the 'learning' one.

 

So, for me atleast, losses are WAY more frustrating, and i'm forgetting about the fact i profit from the mistakes the enemy team make. I'm not exactly sure what needs to be done, but something has to drastically change, as the percentage of 'fun' game is getting less and less.



Dex022 #13 Posted 05 February 2018 - 09:30 AM

    Lieutenant

  • Player
  • 38409 battles
  • 1,622
  • Member since:
    12-26-2011
Never saw so many one sided battles and i play this game for long time. Just last night it was 15:5 or 5:15 almost every second game and it is far from fun being gang raped or having no time to do dmg if i was in T9 French heavy who can't keep up with the rest after flank is won. Games OVER in 2 minutes and cleanup goes to 3-4 minute and that  simply sucks and it wasn't like that 2-3-4 even 5 years ago for sure not this often. Simply terrible games with little or no satisfaction.

Edited by Dex022, 05 February 2018 - 09:31 AM.


Blavod #14 Posted 05 February 2018 - 09:40 AM

    Private

  • Player
  • 18790 battles
  • 38
  • [DUDS] DUDS
  • Member since:
    05-05-2011

View PostHidesHisFace, on 05 February 2018 - 07:35 AM, said:

 

Look, games like this are always prone to steamrolls and it depends on numerous factors. I'm not really going to delve into this, because the issue complex as hell but I personally didn't notice any significant change. Without statistics, personal feeling about this do not matter anyway. 

And I suppose you do not have statistics from your last hundred games or so to prove your point?

 

In any case 6-15, as far as I remember is completely statistically NORMAL. Hell, there was even a mathematical law about this, though I forgot the name of it. Basically - the team who scores the first kill generally gains actually quite significant advantage, statistically speaking, and every new kill makes it much more difficult for the other team to pull from the pit - basically creating the steamroll. As unpleasant as it is, it is absolutely bloody normal, and it is assuming that both teams are balanced.

 

Of course, I'm not going to deny that games like this have steamrolls, but from the session I have just played it was around 70-75% or so. You may not have noticed, I'm sure many others like you haven't. It may be personal feeling, and yes of course I don't have actual "stats" to back up my view, but I do have on my HD my last 7000 replays, if you would like me to upload them for your perusal?

Having studied statistics, no, a regular 6-15 is not "normal", though if you can point me in the direction of your law about this I would be grateful, and intrigued to read of it.

Cannot deny that a team losing a tank early does put them on the back foot, if it is an important one.

 

Further to this though, I was lucky enough to have many one-tier games, where no team should have a particular advantage, as the composition of teams was well balanced. Why was there still a steamroll in 75% of those games?


Edited by Blavod, 05 February 2018 - 09:47 AM.


OneSock #15 Posted 05 February 2018 - 09:49 AM

    Lieutenant

  • Player
  • 34944 battles
  • 1,698
  • [BBMM] BBMM
  • Member since:
    06-06-2011

View PostZlatanArKung, on 05 February 2018 - 08:32 AM, said:

My guess is that it is an effect of the 3-5-7 matchmaker putting most of the power in few tanks (2-3/team). And if one team have a good top tier and the other don't, the outcome is a steamroll.

 

There is truth in this. I had a game yesterday with a defender in the top slot 3-5-7 team. had he gone to the normal spot to fight the enemy tier 8s, we would have won. However, he went to the base and was totally ineffective until rest of the team was decimated and he was swamped.

 

this example of poor game craft isn't reflected well in the stats, because as he went down swinging he did a fair bit of damage. but damage in the wrong place and wrong time is not accounted for in WN8. 



NUKLEAR_SLUG #16 Posted 05 February 2018 - 09:49 AM

    Captain

  • Player
  • 29236 battles
  • 2,280
  • Member since:
    06-13-2015
If you are demanding that WG artificially manipulate the game so that you win more when you're on a losing streak then that would require WG balance it by placing high ratio winners onto the "losing" team as obviously everyone can't win all the time. So when you demand help to win what you're also demanding from WG is that they artificially manipulate the game so that you lose more when you're on a winning streak.

Are you sure that's something you want?

Blavod #17 Posted 05 February 2018 - 10:19 AM

    Private

  • Player
  • 18790 battles
  • 38
  • [DUDS] DUDS
  • Member since:
    05-05-2011

View PostNUKLEAR_SLUG, on 05 February 2018 - 08:49 AM, said:

If you are demanding that WG artificially manipulate the game so that you win more when you're on a losing streak then that would require WG balance it by placing high ratio winners onto the "losing" team as obviously everyone can't win all the time. So when you demand help to win what you're also demanding from WG is that they artificially manipulate the game so that you lose more when you're on a winning streak.

Are you sure that's something you want?

No, nothing of the sort! I'm not really that fussed if I win or lose, obviously I'd prefer to win, as we all do, but that is not my point. I'd much prefer to lose a hard fought game, than to win a steamrolled one, because those games where a fight happens are so much more engaging and fun, and, frankly, is why I play games against other people. I don't want to roll over a team, I want a fight. This is what I believe is missing, currently.



HidesHisFace #18 Posted 05 February 2018 - 10:30 AM

    Second Lieutenant

  • Player
  • 17849 battles
  • 1,299
  • Member since:
    05-12-2012

View PostBlavod, on 05 February 2018 - 09:40 AM, said:

Having studied statistics, no, a regular 6-15 is not "normal", though if you can point me in the direction of your law about this I would be grateful, and intrigued to read of it.

 

I believe it would be Lanchester's square law.

 

Now, bear in mind, it is a simplification here, because there are other factors in WoT than just attrition that this law implies, but in a balanced, properly spread team YOU WILL HAVE this kind of result. You will not have a battle ending with 14-15 results.

That is because a single unit death is already a massive advantage.

For the sake simplicity assume that power of each unit equals 1. In a perfect spread you have effectively 15 1v1 combats.

But say someone dies, and the victor moves to help his ally. now you have 2v1 in this case. According to this square law, the power of these two is not twice as much, but a square as much. 2^2 is 4. It makes sense - they fire twice as fast, deal twice as much damage with each volley, have potentially twice as much health all at the same time. This chain will carry over in ideal situation. 

 

Now, sure - this does not take into consideration factors such as difference between individuals, their equipment, quality of position and so on. Still, it explains the general principles behind such steamrolls. WoT has no respawn mechanics, so each loss is permanent and degrades the team's ability to fight back, unless it can be compensated by team tactics (unlikely), or a good position (moderately likely).

 

 

 



evilchaosmonkey #19 Posted 05 February 2018 - 07:45 PM

    Lieutenant

  • Player
  • 16655 battles
  • 1,756
  • [EIGHT] EIGHT
  • Member since:
    05-04-2013

View PostJigabachi, on 05 February 2018 - 08:09 AM, said:

Why not a skillbased MM in general? The problem with these discussions is that people have no idea what that really is when talking about it. They can only think of "you vs. your clone... hurrdurr... 50%!"

 

I agree whole heartily.

 

Contemplating my dynamic (aka shite) idea for MM in the traffic jam this morning, I wondered if I could write a simulation code for it and base it on a random sampling of real player stats. 

That is, pick some players look at their WR and how they choose which tiers / tanks to play over a small(ish) data set, consider how they perform in that tank and against their overall WR and all that jazz then see how a MM may pick these teams randomly, then dynamically if they lost too much. The aim being to see what happens to their theoretical stats.

 

Whilst 'doable' the slight major problem is simulating whether they won or lost a game with any sense of realism.

 

In the end, the traffic moved on and I realised it was a stupid proposition.



Enforcer1975 #20 Posted 05 February 2018 - 08:11 PM

    Field Marshal

  • Player
  • 20760 battles
  • 10,863
  • [WJDE] WJDE
  • Member since:
    05-04-2014

View PostDurtiArry, on 05 February 2018 - 08:40 AM, said:

I do use XVM and the majority of games are orange and yellows .vs. yellows and greens.

The balance is close but not close enough to have a 50/50 chance.

Despite the cries for "proof" and "where is the evidence" ............. - I know what I see !

 

 

Even with equally skilled teams there will be a losing one and the chances of steamrolls are just as high when one team falters. 






1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users