83 replies to this topic

[acrisis]

TDS are the worst thing to run right now, with these teams, equipment, etc.

 

[__Crusader6__]

Enzo_Gorlami, on 25 June 2017 - 02:01 AM, said:

 

No can do. The heavies in 1-5 scuk. That is the issue, i really don't like medium or light tanks. Trying out TDs now but I guess I may not have the patience to sit back and hope the team does it's part. The teams MM gives me, no way that will happen. 

 

 

Uh, tier 4 has the DW2, tier 5 the  KV-1, Excelsior.  Those are some of the most op tanks in tier. 

 

You just played your Tiger 2, E75 and E100 again and are getting decent tier 5 damage with them. 

 

If you don't take advice, stop moaning about it.   

[Enzo_Gorlami]

__Crusader6__, on 25 June 2017 - 12:39 PM, said:

 

Uh, tier 4 has the DW2, tier 5 the  KV-1, Excelsior.  Those are some of the most op tanks in tier. 

 

You just played your Tiger 2, E75 and E100 again and are getting decent tier 5 damage with them. 

 

If you don't take advice, stop moaning about it.   

 

Only to get these 3 back up to where I left off Friday night. Mission Accomplished, 0 Spare Parts awarded. 

[Enzo_Gorlami]

acrisis, on 25 June 2017 - 12:34 PM, said:

TDS are the worst thing to run right now, with these teams, equipment, etc.

 

 

Hmmm Thinkathat. I see some good TD players and talk with them after the game. They give good advice and I am trying, but it is an alien mentality lol  I forget his name but one of them did a Raisin and Mastery on a Valentine, taking out my SU or was it AT TD last. I swear he had only a notch off his HP when he came for me. 

Edited by Enzo_Gorlami, 25 June 2017 - 01:06 PM.

[CHAOS_ZONE]

you have no way of knowing

[Enzo_Gorlami]

ZoneZ72, on 25 June 2017 - 02:08 PM, said:

you have no way of knowing

 

it is black box software. Proprietary, and all that. My point was that if it were really random for a tier, we would not have lopsided games where one side is stacked and one side is a joke, as often as we see now. I see these lopsided games every other game or two. Making it random would not have this happen that often. Any statistician can say that with confidence. 

[CptCheez]

Enzo_Gorlami, on 25 June 2017 - 11:25 AM, said:

it is black box software. Proprietary, and all that. My point was that if it were really random for a tier, we would not have lopsided games where one side is stacked and one side is a joke, as often as we see now. I see these lopsided games every other game or two. Making it random would not have this happen that often. Any statistician can say that with confidence. 

 

Actually, that's completely wrong.  Any statistician will tell you that a random matchmaker will MORE often result in unbalanced games than balanced games.  Look at this post of mine from 2 years ago, where I explain exactly why math backs up the fact of a random matchmaker in this game:

 

CptCheez, on 19 March 2015 - 02:26 PM, said:

Unfortunately, that's rarely the result of a truly random matchmaker.  Truly random matches, which these are, are much more likely to result in a vast mismatch in skill between teams than they are to result in evenly matched teams.

 

Let's say any 1 person's skill in the game is represented by a value from 1-10.  Now roll 7 * 10-sided dice.  That's the "skill level" for that team.  Roll them again, and that's the "skill level" for the other team.  You're going to see many more rolls with a very wide variance than you will rolls that are close in value.  Here's some results from a random dice rolling app:

 

7d10 → [8,10,8,9,5,2,10] = (52)
7d10 → [4,3,10,5,8,1,1] = (32)

 

7d10 → [3,1,10,1,3,5,4] = (27)
7d10 → [1,4,8,2,9,8,10] = (42)

 

7d10 → [9,10,2,1,6,6,10] = (44)
7d10 → [10,4,4,6,3,8,10] = (45)

 

7d10 → [3,10,10,5,10,2,8] = (48)
7d10 → [6,4,3,5,7,7,7] = (39)

 

7d10 → [3,8,7,10,10,10,1] = (49)
7d10 → [1,6,5,5,1,6,10] = (34)

 

7d10 → [1,4,6,2,5,4,7] = (29)
7d10 → [7,10,1,9,7,5,2] = (41)

 

7d10 → [6,7,6,5,3,1,5] = (33)
7d10 → [2,10,4,9,6,3,3] = (37)

 

7d10 → [1,4,5,7,2,9,10] = (38)
7d10 → [5,6,3,2,9,10,6] = (41)

 

7d10 → [10,4,1,1,2,9,4] = (31)
7d10 → [8,8,8,4,9,1,9] = (47)

 

7d10 → [2,8,6,3,8,7,4] = (38)
7d10 → [7,4,2,4,5,3,4] = (29)

 

Assuming that a "close game" would only be the result of the team values being within, say, 5 of each other, then only 3 of the above 10 examples would be a close game.  The other 7 would be steamrolls one way or the other.

 

[cheasesteak]

CptCheez, on 25 June 2017 - 04:51 PM, said:

 

Actually, that's completely wrong.  Any statistician will tell you that a random matchmaker will MORE often result in unbalanced games than balanced games.  Look at this post of mine from 2 years ago, where I explain exactly why math backs up the fact of a random matchmaker in this game:

 

 

Waiting for someone to tell you that your dice aren't random. WG please fix. 

[madogthefirst]

__Crusader6__, on 25 June 2017 - 04:39 AM, said:

Uh, tier 4 has the DW2, tier 5 the  KV-1, Excelsior.  Those are some of the most op tanks in tier. 

 

You just played your Tiger 2, E75 and E100 again and are getting decent tier 5 damage with them. 

 

If you don't take advice, stop moaning about it.   

Now here is the a phrase I thought I would never see. Yeah the DW2 is sh!t loads better here thanks to that longer 75.

[CptCheez]

cheasesteak, on 25 June 2017 - 01:06 PM, said:

Waiting for someone to tell you that your dice aren't random. WG please fix. 

 

Well, it's a somewhat flawed example for a few reasons.  1) because I only posted a small sample size of only 10 games.  I did a simulation of hundreds of games with the same results, but that's entirely too much to post here on the forums.  But more importantly 2) because it assumes that the pool of skill to draw from is an even distribution from 1-10, where we all know it's not.  There are far fewer people in the "8-10" skill range than there are in the 1-4 and 5-7.  That's simply the result of a closed system with a set number of wins and losses.  For every 1 person with a 60% WR, there are 199 other people with an average 49% WR (42-55%) across them (see here for why the math works that way).

 

But the math is solid and perfectly explains why we see so many more lopsided games.

Edited by CptCheez, 25 June 2017 - 06:07 PM.

[SpartacusDiablo]

No. This is a ridiculous joke.

 

Want to hear a joke about paper?

Spoiler

 it's tearable.

 

[Ericmopar]

There is more to the MM than meets the eye, and I'll tell you guys why.

The MM frequently puts together teams that play well together against wandering noobs/bozos.

 

I've had bad mornings, where nearly 100% of the teams I'm on for a couple of hours straight, split up and literally wander aimlessly without a clue, but nearly 100% of the red teams do exactly the opposite, they stay together and wolf pack.

That isn't chance, even if the way the tanks are matched up are.

I strongly suspect the MM is making up "teams" and "targets" based on some kind of stats.

 

Don't even get me started on the fact that a lot of teams are like 4 or 5 heavies against 1 heavy and some mediums, lights and TDs. The MM would have to know to move some heavies to the team with only one heavy, and some of the lights and mediums to the opposite team...  The programming for that wouldn't be difficult at all, any programmer could tell you that.

[Gojira5475]

Ericmopar, on 25 June 2017 - 03:33 PM, said:

There is more to the MM than meets the eye, and I'll tell you guys why.

The MM frequently puts together teams that play well together against wandering noobs/bozos.

 

I've had bad mornings, where nearly 100% of the teams I'm on for a couple of hours straight, split up and literally wander aimlessly without a clue, but nearly 100% of the red teams do exactly the opposite, they stay together and wolf pack.

That isn't chance, even if the way the tanks are matched up are.

I strongly suspect the MM is making up "teams" and "targets" based on some kind of stats.

 

Don't even get me started on the fact that a lot of teams are like 4 or 5 heavies against 1 heavy and some mediums, lights and TDs. The MM would have to know to move some heavies to the team with only one heavy, and some of the lights and mediums to the opposite team...  The programming for that wouldn't be difficult at all, any programmer could tell you that.

 

You've noticed that too? Where your team (green) has no clue yet red seems way too coordinated to be a random thing? I honestly don't know what to say about that....

[SpartacusDiablo]

I've noticed that when I play really bad my win rate tends to decline. I've also noticed that when I play well my win rate tends to increase.  

Strange...

[Ericmopar]

Gojira5475, on 25 June 2017 - 02:07 PM, said:

 

You've noticed that too? Where your team (green) has no clue yet red seems way too coordinated to be a random thing? I honestly don't know what to say about that....

 

Yes, but I don't mean occasionally, I mean like up to 4 hours of straight playing, yet the same thing happens.
I mean really bad teams, that quite literally go all over and/or stay confused at the spawn for too long, over, and over, and over for hours. Yet, the teams you go up against, are coordinated more than 90% of the time and are frequently loaded with heavies, while your team has one or two heavies. It can happen over and over for hours on end. I suppose it could be random, but that kind of consistent "randomness" seems awfully improbable.

[__Crusader6__]

SpartacusDiablo, on 25 June 2017 - 09:22 PM, said:

I've noticed that when I play really bad my win rate tends to decline. I've also noticed that when I play well my win rate tends to increase.

Strange...

 

same -  I haven't figure that one one out yet

[Tackmoney]

Just get a t95 and camp in the corner

[__Crusader6__]

Tackmoney, on 25 June 2017 - 10:13 PM, said:

Just get a t95 and camp in the corner

 

He did it earlier today. On red...

 

credit where credit is due - it was a good spot for him and he did the most damage by far on red - and he killed my T49 as I was about to Derp him

Edited by __Crusader6__, 26 June 2017 - 03:19 AM.

[Enzo_Gorlami]

__Crusader6__, on 26 June 2017 - 03:17 AM, said:

 

He did it earlier today. On red...

 

credit where credit is due - it was a good spot for him and he did the most damage by far on red - and he killed my T49 as I was about to Derp him

 

It can't have been him in town then, a couple nights ago. There was a T95 in the middle of a pack of tanks in the streets, and since I was lower tier, all I could see was a glowing orange/red hitskin. 

[4sfield]

Tackmoney, on 25 June 2017 - 10:13 PM, said:

Just get a t95 and camp in the corner

 

  Blue_rock has to be lovin it now.