338 replies to this topic

[UnofficialOperator]

Zergling, on 26 October 2016 - 06:17 PM, said:

That's because you don't have a point. It is completely moronic to judge a mech based on the performance of random players seen in a single battle.

You are selecting a single data point to support your argument, a fallacy termed 'cherry picking'. It is not worth refutation as a result.

If you want people to positively respond to your posts, you need to make intelligent arguments.

OMG have you read & understood my previous posts???? That is exactly what I'm saying and what you have written proves my argument again for me!!! You cannot cherry pick stats from a single battle because it is anecdotal! I keep saying, take all the data as a whole, data we do not have. And do you not see how ridiculous it is to show data which contradicts itself in the same screenshot as per OP??? And as previously posted, I can bring up absurd stats like the 50 WLR light pilot to show how you should nerf light mechs instead!

And your opinion that playing 1 mech improves your skills across all mechs. How do you know that to be true across the board? And you can't prove it. The way you play lights and mediums are vastly different. And this is easily disproved by raising WLR and KDR once I play exclusively in light mechs (accounting for weight difference). Perhaps I will do it in November when I got the time (btw my light mech matches are 9% of my assault, med mechs are 3%, i.e. a total of 65 matches in meds, heavies are also at 9%. Do you realise how ridiculous low that med number is and doesn't prove anything about anyone's competence in any chassis).

You claim that we shouldn't cherry pick stats and yet you cherry pick stats from certain players to "prove" your argument. Again contradicting your own stance.

Edited by UnofficialOperator, 27 October 2016 - 05:58 PM.

[Mcgral18]

UnofficialOperator, on 27 October 2016 - 05:42 PM, said:

OMG have you read & understood my previous posts???? That is exactly what I'm saying and what you have written proves my argument again for me!!! You cannot cherry pick stats from a single battle because it is anecdotal! I keep saying, take all the data as a whole, data we do not have. And do you not see how ridiculous it is to show data which contradicts itself in the same screenshot as per OP??? And as previously posted, I can bring up absurd stats like the 50 WLR light pilot to show how you should nerf light mechs instead!

And your opinion that playing 1 mech improves your skills across all mechs. How do you know that to be true across the board? And you can't prove it. The way you play lights and mediums are vastly different. And this is easily disproved by raising WLR and KDR once I play exclusively in light mechs (accounting for weight difference). Perhaps I will do it in November when I got the time (btw my light mech matches are 9% of my assault, med mechs are 3%, i.e. a total of 65 matches in meds, heavies are also at 9%. Do you realise how ridiculous low that med number is and doesn't prove anything about anyone's competence in any chassis).

You claim that we shouldn't cherry pick stats and yet you cherry pick stats from certain players to "prove" your argument. Again contradicting your own stance.

No, because 400+ matches is a decent sample size for Shooty Stompy Robots.
How many matches does that 50 W/l of yours have?


PUG LIFE or Groups?


Because we have correlation in our Bear data, while Lights are pretty much universally poor(er) performers.

[Gen82]

You don't want bias when balancing. The two are mutually exclusive.
If you play the mech/class/character (preferecially), and you don't want to do worse as a result of balance then you have an obvious bias.

This is all I have to say regarding the more recent posts, and probably 3/4 of this thread.

[UnofficialOperator]

Zergling, on 27 October 2016 - 06:28 PM, said:

The data provided by El Bandito and Mcgral18, in addition to the comprehensive assault mech leaderboard events data, is overwhelming statistical evidence of a balancing anomaly with the KDK-3.

That you think that evidence is 'cherry-picking' indicates you have absolutely no knowledge of statistical methodology, so I recommend you shut up before you make yourself look like even more of a fool.

Lol so much fail. So if there are 98 KDK 3 doing 300 average damage and 2 KDK 3 doing exceptionally well consistently over 55 matches, we obviously balance it with the 2 KDK-3 data right? And yet you don't have data for the other KDK 3 users right?

@Zergling, just answer me this very simply. Do we use the data for all KDK 3 users or all Assault users or what? Do we have the data? Why are you conflating Assault MS with KDK 3 MS? Do you have all the KDK MS?

@Zergling, is the data posted by El Bandito representative of the average KDK 3 user? Do you have the MS of the average KDK 3 user?

@Mcgral18, so El Bandito's sample size of 55 matches is good to use but not that 50 WLR of that particular light mech pilot? Hmm OK, and I'm the one that doesn't understand cherry picking... LOL do you guys even read the whole thread and the intent of OP's post?

Keep on dodging guys.

Edited by UnofficialOperator, 27 October 2016 - 07:55 PM.

[-Vompo-]

A way to truly test what is op in the game is remove all restrictions from the mwowc finals and see what the teams use.

[Mcgral18]

UnofficialOperator, on 27 October 2016 - 07:46 PM, said:

@Mcgral18, so El Bandito's sample size of 55 matches is good to use but not that 50 WLR of that particular light mech pilot? Hmm OK, and I'm the one that doesn't understand cherry picking... LOL do you guys even read the whole thread and the intent of OP's post?

Keep on dodging guys.

Show me the data, bro
Where were those matches played?

50 to 1? Group Queue, or PUG god
My 400 matches were spread everywhere

Now, it's about time to start reporting your worthless posts.

[UnofficialOperator]

Mcgral18, on 27 October 2016 - 08:12 PM, said:

Show me the data, bro
Where were those matches played?

50 to 1? Group Queue, or PUG god
My 400 matches were spread everywhere

Now, it's about time to start reporting your worthless posts.

Lol see, when I point out your inconsistency/hypocrisy in your logic, you get angry. I am of course following Zergling's iffy standard of using the leaderboard. The numbers he used does not state origin of whether it is PUG or Group drops either. Go filter out the leaderboards, in one of the seasons, there is a player with 50 WLR. Like I said, if you are comfortable with taking OP's 55 match stats, then it is also fine to take that 50 WLR pilot stats. Again you do not acknowledge this

Again you prove my point. You still do not understand we can't use anecdotal evidence of outlier players. We need data for everyone. Data you don't have. Do you acknowledge this?

Edited by UnofficialOperator, 27 October 2016 - 08:28 PM.

[Ex Atlas Overlord]

El Bandito, on 24 October 2016 - 07:23 PM, said:

Not as good as yours, but way better than it should be considering I suck with ballistics weapons...

Disclaimer: Solo Q, T1



I agree, that KDK3 is easy mode OP shite mech.

Edited by Ex Atlas Overlord, 27 October 2016 - 08:25 PM.

[Dee Eight]

Zergling, on 27 October 2016 - 06:28 PM, said:

SNIP table image

wow...talk about cherry picking stats to make your argument seem valid and assuming my own heavy stats are based on only playing one archer (which I didn't even own until the heavy mech sale happened mid september). If that's how you win an argument amongst tiertards...fine...all the following have already been mastered.

ARCTIC CHEETAH ACH-A 33 22 10 2.20 20 23 0.87 9,386 55,674 04:01:55
ARCTIC CHEETAH ACH-PRIME 25 17 8 2.13 19 19 1.00 6,361 54,466 02:52:00
ARCHER ARC-5W 167 89 76 1.17 134 110 1.22 56,258 190,804 19:14:18
CYCLOPS CP-11-P 36 25 11 2.27 24 25 0.96 11,611 45,074 04:10:55
KODIAK KDK-2 45 27 16 1.69 39 26 1.50 15,274 69,037 05:19:45
WARHAWK WHK-B 43 25 18 1.39 36 25 1.44 21,361 66,801 05:55:16

Quote

Your heavy leaderboard stats for October: 135 wins, 158 losses (0.85 W/L), 188 kills, 213 losses (0.88 K/D), 183 average match score.
Your heavy leaderboard stats for September: 191 wins, 191 losses (1.00 W/L), 185 kills, 300 deaths (0.62 K/D), 166 average match score.
Your heavy leaderboard stats for August: 51 wins, 55 losses (0.93 W/L), 74 kills, 89 deaths (0.83 K/D), 184 average match score.
Your heavy leaderboard stats for July: 59 wins, 77 losses (0.77 W/L), 104 kills, 106 deaths (0.98 K/D), 199 average match score.

My heavy mechs for the past four months have included archers, riflemen, quick draws, warhammers, marauders, night gyrs, summoners, timber wolves, dragons, mad dogs, and a cataphract.

Quote

Stop making **** up: you do not have high win or kill ratios in your Archers. You are not a good player, so stop pretending to be what you are not.

Just like unofficial operator has pointed out... you do not have access to all the data needed to prove anything. You have access to at best, summaries with no actual breakdowns of mech performances or player performances.

My ARC-5S which I have only finished the basic skill level with...

ARCHER ARC-5S 15 11 4 2.75 13 6 2.17 4,677 24,995 01:48:45

Quote

And FYI, there are players talking here that want the KDK-3 nerfed to remove their own advantage.
There are in fact, a bunch of players arguing against nerfs because they don't want to lose their advantage.

Again you're wrong...they're not advocating for white knight status to take away their own advantage. They're advocating to take away advantages from OTHER players who they are threatened by having the same "easymode" mech. How dare people PAY to fast track their results.

Quote

The data provided by El Bandito and Mcgral18, in addition to the comprehensive assault mech leaderboard events data, is overwhelming statistical evidence of a balancing anomaly with the KDK-3.

You mean the leaderboard events that were cheated upon by alt accounting to game the matchmaker and wholesale coordinated group farming ? Those leaderboards?

Quote

so I recommend you shut up before you make yourself look like even more of a fool.

I recommend you take your own advice.

Edited by Dee Eight, 27 October 2016 - 08:32 PM.

[Zergling]

UnofficialOperator, on 27 October 2016 - 07:46 PM, said:

Lol so much fail. So if there are 98 KDK 3 doing 300 average damage and 2 KDK 3 doing exceptionally well consistently over 55 matches, we obviously balance it with the 2 KDK-3 data right? And yet you don't have data for the other KDK 3 users right?

@Zergling, just answer me this very simply. Do we use the data for all KDK 3 users or all Assault users or what? Do we have the data? Why are you conflating Assault MS with KDK 3 MS? Do you have all the KDK MS?

@Zergling, is the data posted by El Bandito representative of the average KDK 3 user? Do you have the MS of the average KDK 3 user?

@Mcgral18, so El Bandito's sample size of 55 matches is good to use but not that 50 WLR of that particular light mech pilot? Hmm OK, and I'm the one that doesn't understand cherry picking... LOL do you guys even read the whole thread and the intent of OP's post?

Keep on dodging guys.

You are continuously misrepresenting the statistical evidence available. This is either a deliberate effort of bias, or simple stupidity.


Here's Mcgral18's stats that he has posted in this very thread:


415 battles of data in the KDK-3, showing the exact same pattern as El Bandito's 55 battles in the KDK-3.


And did you forget or just decide to not mention the event leaderboards data?




Mcgral18 + El Bandito's stats are significant samples, showing that the mech is overpowered in the hands of players that know what they are doing.
The leaderboards show that over a large number of players over a large number of battles, the Kodiak is overpowered.


Those sets of data are statistically significant, and not 'cherry-picking' in any way.


Further, there is in fact, data representative of players closer to average than El Bandito and Mcgral18 in the KDK-3; your own leaderboard stats:
Overall = 1.18 W/L, 1.56 K/D, 270 average MS
Lights = 1.27 W/L, 1.10 K/D, 185 average MS
Mediums = 0.68 W/L, 0.74 K/D, 168 average MS
Heavies = 1.00 W/L, 1.43 K/D, 248 average MS
Assaults = 1.54 W/L, 2.28 K/D, 369 average MS



The same goes for MOBAjobq:

Overall = 1.16 W/L, 1.32 K/D, 255 average MS
Lights = 1.11 W/L, 1.01 K/D, 184 average MS
Mediums = 1.06 W/L, 0.83 K/D, 214 average MS
Heavies = 1.40 W/L, 1.09 K/D, 239 average MS
Assaults = 1.15 W/L, 1.72 K/D, 310 average MS


Both you and MOBAjobq have abnormally high statistics in assaults, which proves that even in less skilled players than El Bandito and Mcgral18, the KDK-3 performs well above normal for a player of a given skill level.

It also demonstrates just why you are attempting (and failing hard) to disprove the overwhelming evidence that the KDK-3 is overpowered; you don't want your crutch taken away.

I saw this behaviour all the time in World of Tanks and War Thunder; people that abuse overpowered things will defend it to the death, because their ego depends on feeling like they are doing well at a game.
Take away the one thing that allows them to appear to be doing well, and their ego just can't handle the reality that they just might not be great at the game.

Edited by Zergling, 27 October 2016 - 09:12 PM.

[Mcgral18]

UnofficialOperator, on 27 October 2016 - 08:23 PM, said:

Lol see, when I point out your inconsistency/hypocrisy in your logic, you get angry. I am of course following Zergling's iffy standard of using the leaderboard. The numbers he used does not state origin of whether it is PUG or Group drops either. Go filter out the leaderboards, in one of the seasons, there is a player with 50 WLR. Like I said, if you are comfortable with taking OP's 55 match stats, then it is also fine to take that 50 WLR pilot stats. Again you do not acknowledge this

Again you prove my point. You still do not understand we can't use anecdotal evidence of outlier players. We need data for everyone. Data you don't have. Do you acknowledge this?

It's another Blood Wolf

Burden of proof, bro


I've provided mine
Reported

[Dee Eight]

Leaderboard by W/L ratio, light mech....

Season 4
1 Nova in Water 69 2 34.50 136 10 13.60 71 287

Season 3

1 Araara 24 1 24.00 34 7 4.86 25 297


Season 2

1 QueenBlade 50 1 50.00 77 6 12.83 51 282


Season 1

1 Celyth 39 1 39.00 48 12 4.00 40 257

[Y E O N N E]

-Vompo-, on 27 October 2016 - 08:10 PM, said:

A way to truly test what is op in the game is remove all restrictions from the mwowc finals and see what the teams use.

The fact that they only choose KDKs for the two Assault slots they do have, on both maps, should already tell you what you need to know.

[Mcgral18]

Dee Eight, on 27 October 2016 - 08:37 PM, said:

Leaderboard by W/L ratio, light mech....

Season 4
1 Nova in Water 69 2 34.50 136 10 13.60 71 287

Season 3

1 Araara 24 1 24.00 34 7 4.86 25 297


Season 2

1 QueenBlade 50 1 50.00 77 6 12.83 51 282


Season 1

1 Celyth 39 1 39.00 48 12 4.00 40 257

Unsurprisingly, both competent players, prominently in groups, and in a small sample size (I've never said the OP's isn't small, but it's typical)

[Deathlike]

Dee Eight, on 27 October 2016 - 08:37 PM, said:

Leaderboard by W/L ratio, light mech....

Season 4
1 Nova in Water 69 2 34.50 136 10 13.60 71 287

Season 3

1 Araara 24 1 24.00 34 7 4.86 25 297


Season 2

1 QueenBlade 50 1 50.00 77 6 12.83 51 282


Season 1

1 Celyth 39 1 39.00 48 12 4.00 40 257

These are great players.

You are not.

[SplashDown]

El Bandito, on 24 October 2016 - 07:23 PM, said:

Been playing the KDK-3 in solo-q for a week now. There are no other explanations for stats like this.




Surely it cannot be the mech. There are plenty of forumers telling me Kodiak-3 is not OP. They speak of how easy it is to focus down the mech and how cumbersome the chassis is. I mean, there are two guys below me who did less than 400 damage in KDK-3, thus proves the mech is not OP, right?




Surely it cannot be my skills. There are plenty of forumers telling me one guy cannot make a difference in 12 v 12. Plus my eye sight is bad, and I tend to spread my shells all around like a zinc addicted porn actor. I mean ****, how else I can have an average of 812 damage per game?

Only explanation left is that of luck. Since I play on Solo-q only with KDK-3, I must be extremely lucky to get good teammates 80% of the time! Guess my new habit of praying every morning for competent teammates worked, and MM God is looking out for me, albeit only when I am playing the Kodiak-3.



But seriously, nerf KDK-3. It is an easy-mode one mouse button using OP engine of destruction. And the dakka version is not even its final form.

just cuz you got a 1200+ damage match with 2 kills doesnt mean the KDK3 is broken..the other 2 guys didnt do so well in the same match with KDK3...KDK3 is fine as is......Getting a 1200+ damage match doesnt mean the mech is OP..i get 800 damage mnatch score with my ADDER all the time..is it OP too?

[El Bandito]

UnofficialOperator, on 27 October 2016 - 07:46 PM, said:

@Mcgral18, so El Bandito's sample size of 55 matches is good to use but not that 50 WLR of that particular light mech pilot? Hmm OK, and I'm the one that doesn't understand cherry picking... LOL do you guys even read the whole thread and the intent of OP's post?

100% that pilot didn't play solo-q, therefore his data is not compatible with my analysis. If you want a top MWO player who plays solo-q a lot that would be someone like Jman5, and his average match stats is far below my KDK-3 stats.

SplashDown, on 27 October 2016 - 08:40 PM, said:

just cuz you got a 1200+ damage match with 2 kills doesnt mean the KDK3 is broken..the other 2 guys didnt do so well in the same match with KDK3...KDK3 is fine as is......Getting a 1200+ damage match doesnt mean the mech is OP..i get 800 damage mnatch score with my ADDER all the time..is it OP too?

To how many people must I explain that the screenshot is a sarcastic jab at people who think KDK-3 is not OP just cause potatoes do not do well with it?

Show me your Adder stats, I highly doubt your Adder average damage per match is anywhere near 800.

Edited by El Bandito, 27 October 2016 - 08:59 PM.

[UnofficialOperator]

Deathlike, on 27 October 2016 - 08:38 PM, said:

These are great players.

You are not.

You like many others are missing the point that we are trying to make. We are showing that it is exactly illogical/irrational to base balancing on the stats of a few. You want to balance, then base it on all the data of all players in all tiers.

[Mcgral18]

UnofficialOperator, on 27 October 2016 - 08:49 PM, said:

You like many others are missing the point that we are trying to make. We are showing that it is exactly illogical/irrational to base balancing on the stats of a few. You want to balance, then base it on all the data of all players in all tiers.

No, you're breaking the rule again

Don't Balance by Potato
That only ends poorly

[Dee Eight]

El Bandito...in the four seasons to date, has 132 games total in assault mechs. That might seem a big sample size to you, but to those of us who have handled much bigger numbers..professionally...that's merely a blip on the back page of a report. In the past four months I've more than five times that number of assault matches. I've played a THOUSAND games more in total just in season 4 as Bandito has (1134 vs 129).