Thank you for the nice evening WG - sarcasm

[it3llig3nc3]

I just want to vent a little regarding how today's gameplay went.

No idea what happened, but as you can see the image here

it was not possible to win a game in a row of 10 no matter what I tried both in Ranked and Random.

just to illustrate that I'm not a complete noob, the lost battles included ones like this

 

This is my most pathetic and laughable MM experience ever since WoWs started.

I'll take a break now as this [edited] is nonsense and WG hopefully will analyze the results and figure out where they messed up the player pairings in MM.

this is crap.

 

[Reaper171]

Ive had worse, and ive brushed it off.

 

craphappens

[Vogel]

Your average stats paint a better picture than one great game. Even a broken clock shows the correct time twice a day.

 

So acknowledge that you are a part of the problem, so that you can improve and get better. Saying that WG is the only villain involved will not make you a better player.

[Foehamster419]

You have only yourself to blame for putting up with it =p

 

That said, thanks for having the bad luck so I don't have to have it

[KptStrzyga]

Past 24 hours I played 20 games, won 14, 70% WR. Thank you for the nice evening WG. 

 

[mtm78]

Past 24 hours I played 20 games, won 14, 70% WR. Thank you for the nice evening WG. 

 

Did you have better RNG then OP? ​

 

/ 

[Ecotech]

 

just to illustrate that I'm not a complete noob, the lost battles included ones like this

 

 

Man, your stats are worse than mine and I'm most certainly not the sharpest thing when it comes to sailing ships.

 

Pardon me if I don't really take your word for WG being the only one to blame for your loss streak.

 

In other news, [edited] happens. Wipe it off, move on.

[Crysantos]

Your average stats paint a better picture than one great game. Even a broken clock shows the correct time twice a day.

 

So acknowledge that you are a part of the problem, so that you can improve and get better. Saying that WG is the only villain involved will not make you a better player.

 

Love this statement, so true. And besides that, everybody has crappy days - you can have x kills, y awards if your team sucks, you still might not win. It's part of the game, try to take another closer look at where you lost the game and what you could have done differently to avoid the same outcome. We all keep learning and improving, just try your best and act as a team.

[MrDane]

Well it is not impossible to get the worst teammates any times in a row, I mean, I have had fights where half the team dies within minutes. But part of the issue might also be that you start getting frustraded and then play worse, I know that happens to me sometimes.

[Hedgehog1963]

Played 7, won 6.

[Reyte]

Won all randoms today I played ( around 8) craphappens, somtimes u ve got just the teams. 

[Unintentional_submarine]

craphappens

 

Indeed.

 

I had five 6 Mogami losses in a row last night, before the salvation of a win. One game I deserved a loss, 20k damage, barely had an impact, yet it seems I wasn't the only one having a terrible game as I was ranked third in my team.

 

The thing is, bad streaks happen. They are effectively unavoidable. That just the nature of the beast when you play 21-23 random people. Sometimes the games will suck, and sometimes they will suck for a prolonged period. The only thing you can do it work hard yourself. At some point you will have that game where your contribution will be the deciding factor.

[Kruzenstern]

What's really interesting is that the screenshot shows 8 battles in the Nürnberg and a 9th ongoing, yet the OPs profile only shows 2 battles for him in the Nürnberg...

[Xevious_Red]

What's really interesting is that the screenshot shows 8 battles in the Nürnberg and a 9th ongoing, yet the OPs profile only shows 2 battles for him in the Nürnberg...

 

The other battles are in ranked.

[it3llig3nc3]

Thanks for the encouragement and also the critics. I know that I'm by far not the best player, however this series of loosing battles made me frustrated to put it nicely.

To the last question I indeed played Nurnberg in RANKED most of the time, the only one in progress was a random one into which I entered by mistake (forgot to switch to ranked after playing AMAGI)

 

After calming down from last evening a few points:

 

Yes I indeed blame a bit WG for creating a system that allows such a bad playing experience, however I do understand that it can happen and I'm part of the results most of the time.

What makes me disappointed in this is not the pure fact that I lose so many battles, rather than the fact that my own playing result has absolutely no correlation to the battle results. Taking out 3 ships with a cruiser in RANKED battle practically means almost half the enemy is taken care of. Similarly I had experience in RANDOM with AMAGI where myself and a fellow BB occupied 4-5 enemy ships so the rest of our team was in full numerical and quality advantage yet we lost big time. 

 

This disparity of individual performance and battle result is a "problem" I think, because in WG's scoring system winning the battle creates a HUGE advantage in earning XP and Credits. This is why loosing series becomes a frustration as while you perform OK and time is invested the "randomness" of the system penalizes you (in my case FOR HOURS)

 

My say is that if it is to be accepted that battle result does not depend on individual performance, the scoring system must reflect it in such way that a given player does not get penalized for somebody else's mistakes. 

 

Last observation from me is that just like it was mentioned many times in the World of Tanks forums, the server side setting that WG does day by day absolutely influence game quality. It was proven in many occasions that the "randomness" for example of the aiming zones (hit/miss ratios) can be and are changed from time to time without any announcement. It may happen here as well.

 

Conclusion: yes I admit that I have weaknesses and a lot depends on me, however I would like to suggest that WG has its own share in enabling / allowing such frustrating situations to happen. Randomness in not an excuse for everything even is statistics.

 

 

[bazzarac]

Your sample size isn't big enough to draw any meaningful conclusions, what you have there is what is called a statistical anomaly.

 

To put it into real world terms, if you flip a coin you have a 50/50 chance of getting heads or tails, everytime you flip the coin, so lets say you flip the coin and gets heads 3 times, the 4th time you flip the coin there is still a 50/50 chance of it being heads and tails, this is pure logic, HOWEVER!

If you flip the coin a large number of times say 1000, you would expect to see a 50/50 distribution, say 502 heads and 498 tails.

 

It is very hard to put this on wargaming, if this was something you wanted to test you would have to do what is called a double blind-test.

 

One of the blinds is easy, since wargaming can't see who is playing the game, the other one is trickier, you would have to find another player with the same sort of skills you have, and then make him play say 1000 games on your account, without him knowing that he wasn't on his own account (to eliminate bias) and then compare it to 1000 games you played on your account. if there still was a statistically significant (note these 2 words) difference in win/loss rate then you could argue that wargamings matchmaking  is biased against you.

 

Now if i was in charge of wargamings developing efforts i would work on a way to identify good play (quantifying this would be hard, but at least some headway could be made) and rewarding this with credits, exp maybe a title or something, and then have brackets like say LOL ranked or HOTS ranked have.

 

Anyways i wish you luck with recovering your loosing streak, what i find that if i am in a loosing streak it helps to go watch a movie, or play a diffrent game. something completely diffrent, my fav is currently plague.inc, do a couple of those games and then come back to wows.

 

Bazz

Masta Slacka

 

 

[LilJumpa]

Thanks for the encouragement and also the critics. I know that I'm by far not the best player, however this series of loosing battles made me frustrated to put it nicely.

To the last question I indeed played Nurnberg in RANKED most of the time, the only one in progress was a random one into which I entered by mistake (forgot to switch to ranked after playing AMAGI)

 

After calming down from last evening a few points:

 

Yes I indeed blame a bit WG for creating a system that allows such a bad playing experience, however I do understand that it can happen and I'm part of the results most of the time.

What makes me disappointed in this is not the pure fact that I lose so many battles, rather than the fact that my own playing result has absolutely no correlation to the battle results. Taking out 3 ships with a cruiser in RANKED battle practically means almost half the enemy is taken care of. Similarly I had experience in RANDOM with AMAGI where myself and a fellow BB occupied 4-5 enemy ships so the rest of our team was in full numerical and quality advantage yet we lost big time. 

 

This disparity of individual performance and battle result is a "problem" I think, because in WG's scoring system winning the battle creates a HUGE advantage in earning XP and Credits. This is why loosing series becomes a frustration as while you perform OK and time is invested the "randomness" of the system penalizes you (in my case FOR HOURS)

 

My say is that if it is to be accepted that battle result does not depend on individual performance, the scoring system must reflect it in such way that a given player does not get penalized for somebody else's mistakes. 

 

Last observation from me is that just like it was mentioned many times in the World of Tanks forums, the server side setting that WG does day by day absolutely influence game quality. It was proven in many occasions that the "randomness" for example of the aiming zones (hit/miss ratios) can be and are changed from time to time without any announcement. It may happen here as well.

 

Conclusion: yes I admit that I have weaknesses and a lot depends on me, however I would like to suggest that WG has its own share in enabling / allowing such frustrating situations to happen. Randomness in not an excuse for everything even is statistics.

 

 

 

You should actually change your nickname if you really believe what you wrote.

[Reisen__]

I just want to vent a little regarding how today's gameplay went.

No idea what happened, but as you can see the image here

it was not possible to win a game in a row of 10 no matter what I tried both in Ranked and Random.

just to illustrate that I'm not a complete noob, the lost battles included ones like this

 

This is my most pathetic and laughable MM experience ever since WoWs started.

I'll take a break now as this [edited] is nonsense and WG hopefully will analyze the results and figure out where they messed up the player pairings in MM.

this is crap.

 

SHIPHAPPENS

[it3llig3nc3]

 

You should actually change your nickname if you really believe what you wrote.

 

Your comment is a simple insult, that adds nothing the subject. I wonder why you spent precious 1 minute of your time creating it...? .-)   (you know as Murphy said: "Never argue with a fool — people might forget who's who.")

Have a nice day!

 

[it3llig3nc3]

Your sample size isn't big enough to draw any meaningful conclusions, what you have there is what is called a statistical anomaly.

 

To put it into real world terms, if you flip a coin you have a 50/50 chance of getting heads or tails, everytime you flip the coin, so lets say you flip the coin and gets heads 3 times, the 4th time you flip the coin there is still a 50/50 chance of it being heads and tails, this is pure logic, HOWEVER!

If you flip the coin a large number of times say 1000, you would expect to see a 50/50 distribution, say 502 heads and 498 tails.

 

It is very hard to put this on wargaming, if this was something you wanted to test you would have to do what is called a double blind-test.

 

One of the blinds is easy, since wargaming can't see who is playing the game, the other one is trickier, you would have to find another player with the same sort of skills you have, and then make him play say 1000 games on your account, without him knowing that he wasn't on his own account (to eliminate bias) and then compare it to 1000 games you played on your account. if there still was a statistically significant (note these 2 words) difference in win/loss rate then you could argue that wargamings matchmaking  is biased against you.

 

Now if i was in charge of wargamings developing efforts i would work on a way to identify good play (quantifying this would be hard, but at least some headway could be made) and rewarding this with credits, exp maybe a title or something, and then have brackets like say LOL ranked or HOTS ranked have.

 

Anyways i wish you luck with recovering your loosing streak, what i find that if i am in a loosing streak it helps to go watch a movie, or play a diffrent game. something completely diffrent, my fav is currently plague.inc, do a couple of those games and then come back to wows.

 

Bazz

Masta Slacka

 

 

 

Dear Masta,

 

I like very much your demonstration on statistics. I was thinking about similar things myself, but wasn't brave enough to summarize it in a simple way as you did. Thanks!

 

There are two points that came in my mind:

1.

One is directly related to the statistical chances. I agree that despite the 50% odds even in simple cases such as coin flipping we can experience a series of "heads" andonly the large number of flips are going to show the "true" trend of 50/50.

I may have made the mistake of assuming that a series of 10 battles is a big enough sample to assume that the chance of having another loss must be very very very low and yet it happened...

WG's matchmaking is obviously a very complex system compared to a coin flip so you can argue that sample size should be much much larger - I can agree to that.

In order to get closer to the problem there is a way statistics deals with the extremities -as far as I understand- that is the standard deviation measure. What I would be VERY interested in is to see the standard deviation of the current MM win/loss trends.

In simple terms what I mean is the MIN and MAX win/loss series the system have experienced since release for one given player nick. This would indicate how much fluctuation is "normal" for this complex system.

 

>> my favorite illustration for standard deviation is the following: let's assume you walk out into a bus stop to take the bus. The schedule table displayed at the stop shows that the bus runs every 10 minutes. What would be your expectation regarding how much wait time you will have? 

Without training in statistics one could say it is on average 5 minutes, but no longer than 10. But this is wrong. Why? because what is NEVER mentioned on bus stop schedules is the standard deviation - that is being the intended MIN and MAX fluctuation of the bus arrivals - i.e. traffic jams in the city and you can expect to wait even 20 minutes that is still valid <<

 

Back to our subject: so we know from WG's official communication that the MM has the capability to break player trends (i.e. stop putting the player into relative HIGH TIER battles versus its ship tier even if the tier ruling in general allows that). Obviously the MM is not considering player skills today that may be good to change...

Besides that my comment is still valid for the case if WG does not want to change MM logic and it is proven that the standard deviation of the win/loss ratio can be VERY HIGH: simply reduce the player penalty on loosing by adjusting the multiplier applied against losing team members in favor of the winners.

 

2.

My other point is that looking at the comments it may appear that I was implying that WG or the MM system is putting me personally at disadvantage.

Just for the record: I did not want to say that. I have no conspiracy theory against myself. 

What I was trying to say:  since this is a GAME that supposed to be entertaining, it is a bit harsh that the system in GENERAL allows such a bad experience to happen - especially when there are many ideas as how the discomfort can be eased for everybody's benefit. Ideally I would want WG to acknowledge that.

My apologies if it was not this clear the first time.

 

 

[JapLance]

My personal best is 12 losses in a row, in 2 consecutive sessions. Beat that.

[bazzarac]

 

(shorted down a bit because i can and i am EVIIIIL)

 

Back to our subject: so we know from WG's official communication that the MM has the capability to break player trends (i.e. stop putting the player into relative HIGH TIER battles versus its ship tier even if the tier ruling in general allows that). Obviously the MM is not considering player skills today that may be good to change...

Besides that my comment is still valid for the case if WG does not want to change MM logic and it is proven that the standard deviation of the win/loss ratio can be VERY HIGH: simply reduce the player penalty on loosing by adjusting the multiplier applied against losing team members in favor of the winners.

 

2.

My other point is that looking at the comments it may appear that I was implying that WG or the MM system is putting me personally at disadvantage.

Just for the record: I did not want to say that. I have no conspiracy theory against myself. 

What I was trying to say:  since this is a GAME that supposed to be entertaining, it is a bit harsh that the system in GENERAL allows such a bad experience to happen - especially when there are many ideas as how the discomfort can be eased for everybody's benefit. Ideally I would want WG to acknowledge that.

My apologies if it was not this clear the first time.

 

 

 

 

About the MM breaking trends as i understand this it doesn't give you a higher/lower chance of being on the winning team as the system (AFAIK) doesn't take into account player skill, only tier and class of ships.

This would mean that the game would break a trend where you are always top or bottom tier, sucks to always have to face Kongos in your St. Louis, when your buddy always faces St. Louises in his Kongo.

So the trendbreaker would be to mix it up a bit so that you sometimes are top tier, and sometimes are bottomtier, sometimes in the middle.

 

The game system you are looking for is what we have in LOL and other MOBAs where the game will throw in harder opponents if you are doing good.

so that most people always have a 50% winrate no matter how good or bad they are (unless you are unbelievably good, or complete rubbish).

 

As far as what WOWS MM does, without haveing a look at the actuall code, from the information we have gotten,

It will pair up so that the tiers are evenish,

That both sides have sorta the same number of the various classes,

That people shouldn't wait too long

People should have a mix of toptier/bottomtier battles

Carriers are hardcorded to be same tier, and same number in a match.

Divisions should be evenly matched.

 

Now people will post a ton of pictures with MM fails, but i would wager that if we could get some official thing from WG on this, it would come down to the weighting of the above factors, especially the "That people shouldn't wait too long" and "People should have a mix of toptier/bottomtier battles".

 

But as i wrote earlier this is speculation, and unless WG throws us a bone this is where it will stay.

 

I hope you got over the loosing streak, otherwise come shoot me tonight, i am terrible at this game.

 

Bazz

Masta Slacka

[it3llig3nc3]

My personal best is 12 losses in a row, in 2 consecutive sessions. Beat that.

 

You are quite a good cruiser captain - I might say Admiral :-) looking at your stats vs. mine on the only comparable ship Konigsberg tells a lot :-)

But I'm in no business to beat your personal record of 12 losses in a row if you don't mind :-D

 

[mtm78]

 

Dear Masta,

 

I like very much your demonstration on statistics. I was thinking about similar things myself, but wasn't brave enough to summarize it in a simple way as you did. Thanks!

 

There are two points that came in my mind:

1.

One is directly related to the statistical chances. I agree that despite the 50% odds even in simple cases such as coin flipping we can experience a series of "heads" andonly the large number of flips are going to show the "true" trend of 50/50.

I may have made the mistake of assuming that a series of 10 battles is a big enough sample to assume that the chance of having another loss must be very very very low and yet it happened...

WG's matchmaking is obviously a very complex system compared to a coin flip so you can argue that sample size should be much much larger - I can agree to that.

In order to get closer to the problem there is a way statistics deals with the extremities -as far as I understand- that is the standard deviation measure. What I would be VERY interested in is to see the standard deviation of the current MM win/loss trends.

In simple terms what I mean is the MIN and MAX win/loss series the system have experienced since release for one given player nick. This would indicate how much fluctuation is "normal" for this complex system.

 

>> my favorite illustration for standard deviation is the following: let's assume you walk out into a bus stop to take the bus. The schedule table displayed at the stop shows that the bus runs every 10 minutes. What would be your expectation regarding how much wait time you will have? 

Without training in statistics one could say it is on average 5 minutes, but no longer than 10. But this is wrong. Why? because what is NEVER mentioned on bus stop schedules is the standard deviation - that is being the intended MIN and MAX fluctuation of the bus arrivals - i.e. traffic jams in the city and you can expect to wait even 20 minutes that is still valid <<

 

Back to our subject: so we know from WG's official communication that the MM has the capability to break player trends (i.e. stop putting the player into relative HIGH TIER battles versus its ship tier even if the tier ruling in general allows that). Obviously the MM is not considering player skills today that may be good to change...

Besides that my comment is still valid for the case if WG does not want to change MM logic and it is proven that the standard deviation of the win/loss ratio can be VERY HIGH: simply reduce the player penalty on loosing by adjusting the multiplier applied against losing team members in favor of the winners.

 

2.

My other point is that looking at the comments it may appear that I was implying that WG or the MM system is putting me personally at disadvantage.

Just for the record: I did not want to say that. I have no conspiracy theory against myself. 

What I was trying to say:  since this is a GAME that supposed to be entertaining, it is a bit harsh that the system in GENERAL allows such a bad experience to happen - especially when there are many ideas as how the discomfort can be eased for everybody's benefit. Ideally I would want WG to acknowledge that.

My apologies if it was not this clear the first time.

 

 

 

Source?

 

Or do you mean the many MM patents which are filed and have been discussed to death on WG forums to no avail since having a patent doesn't mean they're using it. Infact the MM patents filed by WG are mutually exclusive, they can't do everything they got patents for at the same time. 

[it3llig3nc3]

 

>shortversion< 

 

But as i wrote earlier this is speculation, and unless WG throws us a bone this is where it will stay.

 

I hope you got over the loosing streak, otherwise come shoot me tonight, i am terrible at this game.

 

Bazz

Masta Slacka

 

Well thanks for the encouragement and I agree with what you say about the MM. I also on the belief that the MM does quite a lot of things, but it can be as good as the actual by the minute player population allows - so the MM itself can be victim of the statistics random behavior Regardless I would really like to see what you mention that eventually MM gets to a state where it watches out for player skills as well :-D

 

Looking forward to meet you on the high seas - but rather a division mate than an opponent based on what I learned here about the state of your mind :-)

 

I'll get over this streak soon do not worry about me. Perhaps both of us should take some lessons from JapLance who seems to be a very fine cruiser player based on his stats...

 

Cheers