68 replies to this topic

[Moonlander]

Well, we know IS mechs may be struggling with some chassis and there's no denying it. But there's no need to keep pointing out stuff like this when we're having an IS quirk pass really soon and they're aware of the faults in the IS mechs... or what they feel is holding them back.

Nonetheless, this tournament was not a good representation of anything, tbh. So much "kill stealing" (and I refer to people running up and just alphaing a torso you've opened or just standing in front of you when a mech is 1 ML was dying so you can't fire) and/or people hiding out til the end and just cleaning up heavily battered mechs with their fresh (Ohai Twolf pilot, I'm looking at you... you know who you are... I watched you do it on several matches). Not to mention, the countless numbers of LRMs used lately is just ********... but it's the new thing to do (even moreso than before) since you're rewarded so great for using TAG/Narc and then the LRMs for assists. You couldn't get people to hold locks before the new rewards systme, for an LRM boat, to save your life without being called every name in the book. Now, everyone has LRMs, TAG and NARC.

Anyway, it's nice to see a breakdown since I was gone all day and didn't see the end results.

[DYSEQTA]

Based on your colouring you seem to be confused about the Mad Dog. It is a 60 ton Heavy not an Assault.

I can see how you got confused though as the noumber of times I get dropped into the assault lance in mine is amazing. That said though it makes sense as the Mad Dog makes a great Assault support mech.

[Kjudoon]

DYSEQTA, on 27 October 2014 - 11:10 PM, said:

Based on your colouring you seem to be confused about the Mad Dog. It is a 60 ton Heavy not an Assault.

I can see how you got confused though as the noumber of times I get dropped into the assault lance in mine is amazing. That said though it makes sense as the Mad Dog makes a great Assault support mech.

Won't be great for much longer unless this LRM nerf talk backs off and they undo the LRM5 nerf.

[Alistair Winter]

Very hard to extract anything useful from these numbers.

• Tournament doesn't take Elo into account, so winners are not necessarily elite players.
  
• Since the winners are not necessarily elite players, it means they don't necessarily know how to use their mechs to full effect.
  
• Since the winners are not necessarily elite players, the same goes for their opponents. Ipso facto, their opponents may not know how to counter their mechs. (E.g. LRM boats doing well in low Elo matches)
  
• Players gonna play, which means the distribution of players to each chassis is anything but random. For instance, a lot of players may try to grab the #1 Locust or Awesome rank, since one mech is incredibly bad and the other is sort of a cult classic (and... also pretty bad). In other words, the skill distribution between the chassis is probably not completely even.

While there is some correlation between these scores and the commonly accepted mech Tiers, there's also a bunch of surprises (e.g. Awesome significantly better than the Cataphract) which may be explained by the factors above.

[Deathlike]

Alistair Winter, on 27 October 2014 - 11:58 PM, said:

Very hard to extract anything useful from these numbers.

• Tournament doesn't take Elo into account, so winners are not necessarily elite players.

Fair enough. However, if the goal requires you to win a lot, your Elo is going to go up as a result of the tourney... when all is said and done.

Quote

• Since the winners are not necessarily elite players, it means they don't necessarily know how to use their mechs to full effect.

That is unlikely the case. Most people would consider tweaking builds so that they are most comfortable at making it work for them. It is likely some variant of a "optimized build" would occur in fact.

Quote

• Since the winners are not necessarily elite players, the same goes for their opponents. Ipso facto, their opponents may not know how to counter their mechs. (E.g. LRM boats doing well in low Elo matches)

While not guaranteed to be elite players, their Elos would show otherwise. In fact, it is a likelihood that most tourney players would have met each other at one time or another... and possibly in other mechs (either for or not for the tourney, or for grinding)

Quote

• Players gonna play, which means the distribution of players to each chassis is anything but random. For instance, a lot of players may try to grab the #1 Locust or Awesome rank, since one mech is incredibly bad and the other is sort of a cult classic (and... also pretty bad). In other words, the skill distribution between the chassis is probably not completely even.

That may be true... but for the most part... for someone to score high in the Locust, they would actually have to work at it pretty significantly (has anyone taken the Locust seriously in general?)

There is some randomness factor to allow for epic scores to occur, which has to happen at least some random point, so it's not impossible.

Quote

While there is some correlation between these scores and the commonly accepted mech Tiers, there's also a bunch of surprises (e.g. Awesome significantly better than the Cataphract) which may be explained by the factors above.

For the Cataphract, the dominant build when it was "top tier" was when PPCs and ERPPCs were out of whack. Given that the PPC projectile nerf has hurt that weapon the most, and most top tier builds are dependent on them, it is actually far more difficult to put the kind of numbers that they were normally used to previously. Same could be said for the Victors (the meta-Victors are harder to succeed in tourney competition, but not optimized dropdecks).

The build that is "best" for a chassis is not necessarily "tourney optimized". I've learned this from playing enough of these tournies.

[Ghogiel]

bluepiglet, on 27 October 2014 - 06:23 AM, said:

A few things I can get from the simple chart:

1) Damage/Kill based tournament is heavy weight favouring, and boring.

It's assault favoring not so much heavy favoring, as the best lights are beating out some of the best heavies in the game.


The orion doing as well as it did is surprising the lurm scrubs dragged the average over a CTF and even a firestarter

[El Bandito]

Alistair Winter, on 27 October 2014 - 11:58 PM, said:

While there is some correlation between these scores and the commonly accepted mech Tiers, there's also a bunch of surprises (e.g. Awesome significantly better than the Cataphract) which may be explained by the factors above.

Laser vomit build were dominant in this challenge, and Awesome got heat reduction quirk.

[Alistair Winter]

Deathlike, on 28 October 2014 - 12:12 AM, said:

Fair enough. However, if the goal requires you to win a lot, your Elo is going to go up as a result of the tourney... when all is said and done.

To the point where you're actually competing against the best players online at any given moment? It seems unlikely.

Deathlike, on 28 October 2014 - 12:12 AM, said:

That is unlikely the case. Most people would consider tweaking builds so that they are most comfortable at making it work for them. It is likely some variant of a "optimized build" would occur in fact.

Making it work for them, yes. But a person's optimized build depends on their skill level. An LRM Stalker may be the most effective build for one guy, while a quad ER LL / LL Stalker may be the most effecitve for another, and a laser / SRM build may be the most effective for a third. But against the very best players in the game, some builds that are effective against low Elo opponents are going to be ineffective.

For me, the quad LL CPLT-K2 is by far the most deadly of all my Catapults. Would an elite player choose the same variant and build for this tournament? Maybe not.

Deathlike, on 28 October 2014 - 12:12 AM, said:

While not guaranteed to be elite players, their Elos would show otherwise. In fact, it is a likelihood that most tourney players would have met each other at one time or another... and possibly in other mechs (either for or not for the tourney, or for grinding)

I'm sure they've met each other at one time or another, but that doesn't mean they're all consistently being matched up with the best players available at any given time.

No one outside PGI knows enough about Elo and matchmaking in this game to be sure of this stuff. And when we're not sure, it's hard to analyze these data.

Deathlike, on 28 October 2014 - 12:12 AM, said:

That may be true... but for the most part... for someone to score high in the Locust, they would actually have to work at it pretty significantly (has anyone taken the Locust seriously in general?)

Their past experience isn't really relevant to this point. I'm just saying that it's possible all the most skilled players are not evenly distributed across different mechs. Maybe all the pr0s like to play the Dragon. Or the Kintaro. We just don't know. But it's pretty safe to say that the number of people trying to reach the top 10 spot was considerably higher for the Dire Wolf than for the Vindicator. And maybe higher for the Awesome than the Hunchback too.

Deathlike, on 28 October 2014 - 12:12 AM, said:

For the Cataphract, the dominant build when it was "top tier" was when PPCs and ERPPCs were out of whack. Given that the PPC projectile nerf has hurt that weapon the most, and most top tier builds are dependent on them, it is actually far more difficult to put the kind of numbers that they were normally used to previously. Same could be said for the Victors (the meta-Victors are harder to succeed in tourney competition, but not optimized dropdecks).
The build that is "best" for a chassis is not necessarily "tourney optimized". I've learned this from playing enough of these tournies.

In regards to the former point, I agree that the CTF-3D isn't what it once was. But the CTF is still pretty good.
In regards to the latter, that's probably true. Which makes it even harder to extract useful info from these data.

[Alistair Winter]

El Bandito, on 28 October 2014 - 12:28 AM, said:

Laser vomit build were dominant in this challenge, and Awesome got heat reduction quirk.

Laser vomit builds are always pretty common though. Unless PGI makes a weird challenge with lots of emphasis on TAG / NARC / LRM, there will be lots of laser boats.

[Deathlike]

Alistair Winter, on 28 October 2014 - 12:33 AM, said:

To the point where you're actually competing against the best players online at any given moment? It seems unlikely.

Well, "yes" and yet "no".

Since other people have lives (I know I don't quite have that yet), just playing during a tourney and dropping solo nets me a Lord, or some other competing tourney goer. I know enough that my Elo is "adjusted" because of the frequent number of drops during one.

There are rare occasions where I get put into a match where everyone's Elo is probably nowhere near mine (it happened late this tourney, where I got a crazy game that netted me what I needed).

Quote

Making it work for them, yes. But a person's optimized build depends on their skill level. An LRM Stalker may be the most effective build for one guy, while a quad ER LL / LL Stalker may be the most effecitve for another, and a laser / SRM build may be the most effective for a third. But against the very best players in the game, some builds that are effective against low Elo opponents are going to be ineffective.

For me, the quad LL CPLT-K2 is by far the most deadly of all my Catapults. Would an elite player choose the same variant and build for this tournament? Maybe not.

It's hard to explain loadouts explicitly w/o getting into the minutia.

For instance, I've seen some of the tourney goers in the Quickdraw bracket carry virtually the same loadout (well, at least for weapons). That suggests that it is "close" to the optimal loadout for the tourney (there may be better builds, but it's more or less what I thought it would be). In the Wolverine bracket however... the tourney leader (and winner) used exclusively a short ranged build (3 med, 3 ASRM6s) with the 7K. I dabbled but ended up primarily using the 6K (the energy arm, no JJ variant). These two builds have vastly different characteristics... and while I'd probably get a higher score with the 7K, I don't have the kind of patience to go that route (not that I'm afraid to brawl, but tourney mech building has taught me consistency above all else).

Quote

I'm sure they've met each other at one time or another, but that doesn't mean they're all consistently being matched up with the best players available at any given time.

No one outside PGI knows enough about Elo and matchmaking in this game to be sure of this stuff. And when we're not sure, it's hard to analyze these data.

I'm not saying they all meet each other indefinitely... I'm only suggesting for the time brackets and zones involved... some who is going the silly 24/7 route will see a lot of the tourney goers. Someone who is playing in a specific time zone will obviously see a lot of the players that play during said hours. It's natural.

Quote

Their past experience isn't really relevant to this point. I'm just saying that it's possible all the most skilled players are not evenly distributed across different mechs. Maybe all the pr0s like to play the Dragon. Or the Kintaro. We just don't know. But it's pretty safe to say that the number of people trying to reach the top 10 spot was considerably higher for the Dire Wolf than for the Vindicator. And maybe higher for the Awesome than the Hunchback too.

Well, popularity dictates the # of players in a particular bracket. Someone wise enough to go a different bracket instead of going for popularity (say, Dragon over Timberwolf) will have a far easier time in getting over "the hump" in terms of the players involved vs going "easy mode" but having to work far more hard since there are more people competing and likely doing the same thing.

For instance, I've seen Edmeister ran a Cataphract during this tourney. I believe he gave up on that soon enough. Right now, the current "meta" does't favor the classic PPC/ERPPC builds used in previous tournies. He ended up going Timberwolf (didn't see him on other boards though). So, go figure.

Quote

In regards to the former point, I agree that the CTF-3D isn't what it once was. But the CTF is still pretty good.
In regards to the latter, that's probably true. Which makes it even harder to extract useful info from these data.

You can still extrapolate a few obvious things from the data, but there's some oddities that would probably need more looking into to make a proper analysis.

For instance, given that the current "meta" is pro-laservomit, even the lowly Battlemaster that still looks poorly upon is actually pretty energy friendly... (not PPC friendly, but laservomit friendly). Then again, they are used for missile boats just as well...

I haven't seen enough Battlemasters in my tourney matches to make an informed opinion though.

Edited by Deathlike, 28 October 2014 - 12:58 AM.

[El Bandito]

Alistair Winter, on 28 October 2014 - 12:33 AM, said:

To the point where you're actually competing against the best players online at any given moment? It seems unlikely.

Yes. I have seen Awesomes in the same matches as Proton, Odwalla, Bishop, Jman5 etc...

Edited by El Bandito, 28 October 2014 - 01:12 AM.

[Deathlike]

El Bandito, on 28 October 2014 - 01:10 AM, said:

Yes. I have seen Awesomes in the same matches as Proton, Odwalla, Bishop, Jman5 etc...

Yes, sadly. I've also seen you too.

I don't remember if you were in a Victor (I'd have to look @ my screenshots to verify).

I've seen the tourney leaders in Orions with missiles boats.

Yes, I said it... missile boats (in XL no less).

[El Bandito]

Deathlike, on 28 October 2014 - 01:14 AM, said:

I don't remember if you were in a Victor (I'd have to look @ my screenshots to verify).

Dragon Slayer, yes. I didn't know you liked the Wolverine, gratz.

Edited by El Bandito, 28 October 2014 - 01:26 AM.

[Lily from animove]

Kjudoon, on 27 October 2014 - 06:44 PM, said:

And there should be even fewer producing assaults and at slower rates. It's one of the reasons why CW will absolutely need the return of R&R and monitoring the actual number of mechs out there, and pricing according to a REAL and DYNAMIC economic model that covers the logistics of , faction, distance shipped, rate of production and more. Otherwise there is no purpose for capturing a planet let alone not nuking them from orbit.

the CW dropship mechanics will actually prevent this and force a lot people to use lights and mediums as well, except you wanna come with 4x60t mechs.

Hmm can I call my dropship "Dogpound Asylum" and put 4 MDD's in?

[Viges]

The true name of this topic is

THE FALL OF THE VICTOR

[PhoenixFire55]

With few exceptions this seems about right to detrmine which mech is the best mech. And again, with few exceptions this seems about right in terms of which mech is supposed to be the best mech. Kinda lame tho to see the summoner right amongst its IS counterparts and Victor that far down the ladder.

[Stijnovic]

bluepiglet, on 27 October 2014 - 06:23 AM, said:

...The numbers are the means of the tops 15 players of each classes...

Could you add the min/max values and the standard deviation, and also a plot for the latter? The closer the values are, the more people are probably playing that chassis.

(or do you have the excel file available?)

Edited by Stijnovic, 28 October 2014 - 03:08 AM.

[Widowmaker1981]

Kjudoon, on 27 October 2014 - 07:35 PM, said:

I do all the time. It's just smarter to kill a mech without him having a chance to shoot you back. Works for every military and soldier in the world, should work just as well here.

This 'honor' based "I should be able to see my assailant" argument is full of cowdung.

YES. People need to realise that in a game where you cannot heal and have only 1 'life' sensible tactics involve attempting to do damage without receiving any/much in return. Its not cowardice, its having brains.

[bluepiglet]

Updated.

[Lily from animove]

Widowmaker1981, on 28 October 2014 - 03:21 AM, said:

YES. People need to realise that in a game where you cannot heal and have only 1 'life' sensible tactics involve attempting to do damage without receiving any/much in return. Its not cowardice, its having brains.

the reason why the nova is bad, to deliver damage you need to expose a lot of your mech which is the widest amongst the mediums. And then most of the time this makes you get return fire, and with your fat nose and CT its nearly impossible to miss vital parts. SCR look at it n the list, it easily dodges and spreads 3x as much damage. And on top you are in one of the most slowly mediums you can get, making many position changes quite hard compared to the other mechs capabilities. The JJ's rarely help here honestly. A turkey on a roof is way too much of an easy target than a real advantage.

Edited by Lily from animove, 28 October 2014 - 04:27 AM.