127 replies to this topic

[KaptinOrk]

martian, on 29 December 2021 - 01:02 AM, said:

Going with c) means that you will have to do some mental and gameplay adjustments:

• in T3 you can meet more competent opposition, some T1 players included, so you will have to play more cautiously. Some of your habits from T5 and T4 may no longer be viable.
  
• modify your 'Mech builds. Just that Shadow Hawk of yours: The first thing that caught my eye was that pair of Small lasers. If you are engaging enemy 'Mechs on 300-400 metres with your Ultracannon and MRMs, then those small lasers are unused and pointless. And if you have some enemy 'Mech in the point-blank range, those small lasers are not going to help much. Not even a light 'Mech with a decent pilot is going to be deterred by them, no matter if they worked in T5 or T4. Remove them and use that saved ton for something else - something what would be more useful in T3 games.

I actually posted the wrong Shadowhawk, I had a few test builds saved and grabbed the wrong one. Here is the Shadowhawk I meant to post, it has 2xMLas and a slightly smaller engine: https://mwo.nav-alph...6f81f9fc_SHD-5M

Regarding my tactics, I mostly play the Blackjacks lately, I like the medium fire support role. I try to hang with the heavies and assaults and shoot their targets while keeping lights off their backs. I make a point to be engaged as often as I can, shoot first, shoot often, keep the heat gauge up until I need to reposition.

Why is cautious play the better option? I consistently lose when I'm on cautious teams, either the team runs around in circles, avoiding the enemy while getting picked off or the aggressive players try to get the team to form up and make a stand at a defensible position only to get cut down while the cautious player abandon us. The only games I consistently win are games with aggressive teams that push that aggression from the beginning.

[martian]

KaptinOrk, on 29 December 2021 - 10:02 AM, said:

I actually posted the wrong Shadowhawk, I had a few test builds saved and grabbed the wrong one. Here is the Shadowhawk I meant to post, it has 2xMLas and a slightly smaller engine: https://mwo.nav-alph...6f81f9fc_SHD-5M

Well, at least you upgraded those Small lasers.

KaptinOrk, on 29 December 2021 - 10:02 AM, said:

Regarding my tactics, I mostly play the Blackjacks lately, I like the medium fire support role. I try to hang with the heavies and assaults and shoot their targets while keeping lights off their backs. I make a point to be engaged as often as I can, shoot first, shoot often, keep the heat gauge up until I need to reposition.

That's good.

KaptinOrk, on 29 December 2021 - 10:02 AM, said:

Why is cautious play the better option? I consistently lose when I'm on cautious teams, either the team runs around in circles, avoiding the enemy while getting picked off or the aggressive players try to get the team to form up and make a stand at a defensible position only to get cut down while the cautious player abandon us. The only games I consistently win are games with aggressive teams that push that aggression from the beginning.

I did not mean "cautious" as avoiding the contact with the enemy team as it often happens.

I meant "cautious" as waiting for a little while to find out where the enemy team actually is and what 'Mechs and loadouts the enemy players have.

In T3 games you can meet some competent players from higher Tiers who know how to punish your mistakes, for example in positioning.

[ScrapIron Prime]

martian, on 29 December 2021 - 10:22 AM, said:

In T3 games you can meet some competent players from higher Tiers who know how to punish your mistakes, for example in positioning.

This.

Pretty much every game I get owned, I find that I pushed out someplace I shouldn't have been, hesitated too long before moving, or exposed in too predictable a pattern. And a good player punished me for it.

A good 5M build there, by the way. A bit fragile with the XL engine if you're sticking around for multiple shots from the UAC, but good damage. My take on it JUST uses UAC/5's and a light engine, less front loaded damage but more durable. https://mwo.nav-alph...9ae84d86_SHD-5M

[Horseman]

KaptinOrk, on 29 December 2021 - 10:02 AM, said:

Why is cautious play the better option? I consistently lose when I'm on cautious teams, either the team runs around in circles, avoiding the enemy while getting picked off or the aggressive players try to get the team to form up and make a stand at a defensible position only to get cut down while the cautious player abandon us. The only games I consistently win are games with aggressive teams that push that aggression from the beginning.

Aggression without control is just suicide rushing.
You want to engage when you are at an advantage and in weapons range, and disengage if you find yourself at a disadvantage or out of weapons range.
If you engage too early or too aggressively without support, your mech will get shot out from under you in a few seconds. If you're engaging as one of several mechs, the enemy is less likely to focus fire on any single given mech (yours included)

[Michael Abt]

KaptinOrk, on 28 December 2021 - 05:27 PM, said:

I feel like I'm kinda stuck in limbo between T4 and T3. I can do well in a T4 game in an AC/2 Blackjack BJ-1 or a weird Shadowhawk and the matchmaker thinks I'm worthy of T3, but once I'm in T3, I'm getting shredded by PPFLD meta builds left and right, so I get booted back down to T4 to repeat the cycle. I have no interest in playing the meta tryhard game, I just want to screw around with dumb 'mechs and casual players, is there a tier for me?

The way I see tier is an adeption of von Clausewitz. Tiers are a multiplication, and skill is just one factor. There are couple factors, but the only other one you can control is the means, in case of MWO, the mech you choose. If you pick a weak mech then you have to rely more on your skills, and random things like team compositions, picked mech, etc may have a greater impact.

My personal solution to your question is rather simple. I am sticking to a very, very few personal mechs, and I don't care that they are sub par. With that factor being a constant i am improving my skill with those chasis, and that pushes me forward. Yes, it takes longer to move in tier now, and I am accepting the challenge.

Spoiler

To give an impression of my dedication - as of this posting, i made14.900 matches:
HBK-4G(F) 8.146 matches
JR7-Oxide 3.151 matches

Under the old PSR these mechs got me to T1. In the current system i am in the yo-yo state between T3 and T2 atm.

This brings me to my main criticism of the new systems. I do believe that the current system, while overall being fairer and better, shifted the focus away from player skill to the mech(s) piloted. Many players argue it repesents player skill better now, but I do think it is the opposite, because the choice of mech matters so much more now in that multiplication i mentioned above.

Maybe I will switch to better performing mechs for a while to get a stronger foot into T2 again, and then I do have a buffer for learning and adepting my playstyle with my prefered mechs in that new enviroment. It is difficult to do that when you bounce between tiers.

Otherwise, Martian and others already gave some good advice.

[KaptinOrk]

Horseman, on 29 December 2021 - 12:01 PM, said:

Aggression without control is just suicide rushing.
You want to engage when you are at an advantage and in weapons range, and disengage if you find yourself at a disadvantage or out of weapons range.
If you engage too early or too aggressively without support, your mech will get shot out from under you in a few seconds. If you're engaging as one of several mechs, the enemy is less likely to focus fire on any single given mech (yours included)

I prefer to play medium fire support roles, I hang with the team, shoot their targets and try to amplify damage where I can. The issue that I have is that some teams just won't engage, even when we're in a position to overwhelm the enemy. I will fully admit that I'm overly aggressive sometimes and I am definitely punished for it, but I can't play fire support if there's no fire to support.

I've lost too many games with 3 or 4 'mech leads because my team decided to scatter and run away instead of committing to a push and mopping up.

Michael Abt, on 29 December 2021 - 12:08 PM, said:

The way I see tier is an adeption of von Clausewitz. Tiers are a multiplication, and skill is just one factor. There are couple factors, but the only other one you can control is the means, in case of MWO, the mech you choose. If you pick a weak mech then you have to rely more on your skills, and random things like team compositions, picked mech, etc may have a greater impact.

My personal solution to your question is rather simple. I am sticking to a very, very few personal mechs, and I don't care that they are sub par. With that factor being a constant i am improving my skill with those chasis, and that pushes me forward. Yes, it takes longer to move in tier now, and I am accepting the challenge.

Spoiler

To give an impression of my dedication - as of this posting, i made14.900 matches:
HBK-4G(F) 8.146 matches
JR7-Oxide 3.151 matches

Under the old PSR these mechs got me to T1. In the current system i am in the yo-yo state between T3 and T2 atm.

This brings me to my main criticism of the new systems. I do believe that the current system, while overall being fairer and better, shifted the focus away from player skill to the mech(s) piloted. Many players argue it repesents player skill better now, but I do think it is the opposite, because the choice of mech matters so much more now in that multiplication i mentioned above.

Maybe I will switch to better performing mechs for a while to get a stronger foot into T2 again, and then I do have a buffer for learning and adepting my playstyle with my prefered mechs in that new enviroment. It is difficult to do that when you bounce between tiers.

Otherwise, Martian and others already gave some good advice.

I fully believe in playing fun, personal 'mechs, if I'm not having fun with it, then why am I playing the game? I've been trying the same approach as you, forcing myself to git gud with my bad 'mechs instead of taking the easy route with an AMS boat Corsair or generic meta Clam alpha build #23.

The tier yo-yo is the most frustrating part, I only get a few games in T3 before I'm stomped back down to T4. I want to practice in the higher tier to actually improve, not club cadet seals.

[crazytimes]

pbiggz, on 29 December 2021 - 09:07 AM, said:

Lack of accurate reporting means that the game isn't presenting as complete a summary of performance as it could. What I would personally like to see is a bar indicating a team's total armour. There's a non-zero number of games that looks like 12-0 stomps, but where the team that got 0 kills, ended up chewing through 80-90% of the 12 kill team's total armour. Just because they didnt get kills, doesn't mean they didn't perform well across the board. You need the kills to win but damage is still an excellent indicator of performance and the way end of match stats are presented now, it sometimes minimizes that performance, making close matches seem like complete clownshows. A 1-12 match where the winning team gets 12 kills but lost 99% of its armour would be just as statistically unlikely as an 11-12 match, essentially, a near draw, but the game doesn't present that in a meaningful way.

I don't get why there aren't more detailed metrics available. There are match IDs, the data is recorded, having a web interface to look up match data based on match ID would make perfect sense. Even if they anonymised other players, the data is all there... just hidden as part of PGIs maximum opacity policy.

That said- it's perfectly possible for the losing team to do way more damage than the winning team. That can come down to aim, weigh disparity or PPFLD vs. spread weapons. I posted a screenshot last year where my losing team did around 1500 more damage total, but was down around 200 tonne and too many spread weapons.

From a PSR perspective- if you're consistently on the losing team but doing big damage you're A going to go up still and B probably needing to re-evaluate your play style.

Edited by crazytimes, 29 December 2021 - 03:21 PM.

[Sjorpha]

Michael Abt, on 29 December 2021 - 12:08 PM, said:

This brings me to my main criticism of the new systems. I do believe that the current system, while overall being fairer and better, shifted the focus away from player skill to the mech(s) piloted. Many players argue it repesents player skill better now, but I do think it is the opposite, because the choice of mech matters so much more now in that multiplication i mentioned above.

The system measures neither the strength of your mechs or your skill, there is actually no way for it to separate those two things. It measures your results.

What a matchmaker needs to know is your expected impact on matches, so that it can put against players with the same expected impact. It doesn't actually matter for matchmaking if that impact comes from skill or mech choice or whatever.

Edited by Sjorpha, 29 December 2021 - 05:39 PM.

[pbiggz]

crazytimes, on 29 December 2021 - 03:21 PM, said:

I don't get why there aren't more detailed metrics available. There are match IDs, the data is recorded, having a web interface to look up match data based on match ID would make perfect sense. Even if they anonymised other players, the data is all there... just hidden as part of PGIs maximum opacity policy.

That said- it's perfectly possible for the losing team to do way more damage than the winning team. That can come down to aim, weigh disparity or PPFLD vs. spread weapons. I posted a screenshot last year where my losing team did around 1500 more damage total, but was down around 200 tonne and too many spread weapons.

From a PSR perspective- if you're consistently on the losing team but doing big damage you're A going to go up still and B probably needing to re-evaluate your play style.

Pretty much exactly right. Presenting those stats doesn't require the game to be changed, it just requires some UI work. The game already rewards you for that performance anyway. You can lose and get an up arrow. You can win and get a down. A significant number of people feeling bad about stomps would feel less bad if they understood the match results better. Some matches are closer than the scorecard makes them seem.

[GoodTry]

KaptinOrk, on 29 December 2021 - 03:08 PM, said:

I prefer to play medium fire support roles, I hang with the team, shoot their targets and try to amplify damage where I can. The issue that I have is that some teams just won't engage, even when we're in a position to overwhelm the enemy. I will fully admit that I'm overly aggressive sometimes and I am definitely punished for it, but I can't play fire support if there's no fire to support.

That's a great way to play IMO, and I feel your pain on teams that aren't aggressive. It makes everything so much harder. You don't have to go full NASCAR to be aggressive.

KaptinOrk, on 29 December 2021 - 03:08 PM, said:

I fully believe in playing fun, personal 'mechs, if I'm not having fun with it, then why am I playing the game? I've been trying the same approach as you, forcing myself to git gud with my bad 'mechs instead of taking the easy route with an AMS boat Corsair or generic meta Clam alpha build #23.

I agree you should play whatever is fun and try not to worry about tier. All tier is in reality is an automatic difficulty adjustment.

If you are going to "force yourself to git gud" at something though, make it a good meta build off of Grimmechs, not a build that is actually bad. You're not going to get good by intentionally playing a bad build. Plus that's no fun.

KaptinOrk, on 29 December 2021 - 03:08 PM, said:

The tier yo-yo is the most frustrating part, I only get a few games in T3 before I'm stomped back down to T4. I want to practice in the higher tier to actually improve, not club cadet seals.

Just group with a couple of tier one players until you are sick of having your face smashed in :-). If you don't know any, join a unit - there are still units out there recruiting new players who are friendly but aren't yet skilled.

Edited by GoodTry, 29 December 2021 - 10:37 PM.

[Horseman]

KaptinOrk, on 29 December 2021 - 03:08 PM, said:

I prefer to play medium fire support roles, I hang with the team, shoot their targets and try to amplify damage where I can. The issue that I have is that some teams just won't engage, even when we're in a position to overwhelm the enemy. I will fully admit that I'm overly aggressive sometimes and I am definitely punished for it, but I can't play fire support if there's no fire to support.

I've lost too many games with 3 or 4 'mech leads because my team decided to scatter and run away instead of committing to a push and mopping up.

Sometimes it happens, yes. The most you can do is wait for opportunities to get in additional damage (=score) and survive long enough to get as many of them as possible so that you're not losing PSR after the match ends. Proximity towards friendlies triggers multiple score tickets that help with that.

[martian]

Horseman, on 30 December 2021 - 12:10 AM, said:

Sometimes it happens, yes. The most you can do is wait for opportunities to get in additional damage (=score) and survive long enough to get as many of them as possible so that you're not losing PSR after the match ends.

If it is possible, I concentrate on getting at least one honorable kill, so the game does not end with the awkward 0:12 result.

Horseman, on 30 December 2021 - 12:10 AM, said:

Proximity towards friendlies triggers multiple score tickets that help with that.

It works only if the PUGs do not scatter in all directions ... which is the more common case.



But you are right, those rewards help a bit.

And it is always better to see the green arrow than the red arrow.

[Michael Abt]

Sjorpha, on 29 December 2021 - 05:37 PM, said:

The system measures neither the strength of your mechs or your skill, there is actually no way for it to separate those two things. It measures your results.

What a matchmaker needs to know is your expected impact on matches, so that it can put against players with the same expected impact. It doesn't actually matter for matchmaking if that impact comes from skill or mech choice or whatever.

Are you familiar with von Clausewitz? He says the outcome, the result, is a multiplication of willpower and means used (simplified).

In MWO the part of the calculated impact you can influence yourself is your skill and your mech. You are correct that the MM doesn't take those into account directly. It takes your PSR/tier into account, and as I explained, that is the result of skill and mech used.

Just read through this thread. What is one of the most common suggestions? "Use a better mech" Or read through the "lights OP" thread. People posting diagrams, proving that different weight classes yield different average matchscore, which results into PSR/tier. Obviously the mech does have an impact, albeit over time, and not directly.

The main critic I do have is that under the new system the mech has a much greater impact now on the result, and skill is less important. Even the Jarl's list acknowledges that and uses an adjusted matchscore. It would be nice if the MM would use an adjusted matchscore as well.

[Horseman]

Michael Abt, on 30 December 2021 - 02:32 AM, said:

The main critic I do have is that under the new system the mech has a much greater impact now on the result, and skill is less important.

Under the old system neither mech nor skill were very relevant - it was designed to float everyone to T1 over time, and the static tresholds for that were below average player performance (and that's before we got the ~10% match score inflation after the Skill Tree).
We've seen pilots in the 38th percentile - and I believe even lower - making it to T1. One would have to be a total stinker of a potato to not make it to T1 eventually.

Your premise is theoretically correct in that it's easier to put up big numbers in bigger mechs, but practically I see many heavy and assault mechs underperforming even in T1 - it's not difficult for a competent light or medium pilot with a decent build to put in enough numbers to completely outshine them despite the class disparity.

Edited by Horseman, 30 December 2021 - 03:49 AM.

[Michael Abt]

Horseman, on 30 December 2021 - 03:47 AM, said:

Under the old system neither mech nor skill were very relevant - it was designed to float everyone to T1 over time, and the static tresholds for that were below average player performance (and that's before we got the ~10% match score inflation after the Skill Tree).
We've seen pilots in the 38th percentile - and I believe even lower - making it to T1. One would have to be a total stinker of a potato to not make it to T1 eventually.

Your premise is theoretically correct in that it's easier to put up big numbers in bigger mechs, but practically I see many heavy and assault mechs underperforming even in T1 - it's not difficult for a competent light or medium pilot with a decent build to put in enough numbers to completely outshine them despite the class disparity.

In a previous post i said that overall the new system is fairer and better, so there is no disagreement. The one thing better under the old system was that the mech wasn't that important.

I used Jarl's List just as an example for adjusted matchscore, but we could also base it around Solaris devisions. There mechs are sorted according to performance, not weight class. Therefore, your practical observation is not a contradiction to the theory, it is its confirmation.

The better the mech, the more likely it is to get a good matchscore, consistently.

My point is that an adjustment to the matchscore based on mech performance increases the incentive to play more/other mechs as well without being punished.

PS thought: A variation of that discussion is that other game modes don't offer enough rewards to do anything other than skirmish, e.g. you can play a light, cap conquest, win the match for your team, being the pilot of the match, and your PSR still gets a huge hit because you did zero damage.

Edited by Michael Abt, 30 December 2021 - 05:42 AM.

[martian]

Michael Abt, on 30 December 2021 - 05:41 AM, said:

PS thought: A variation of that discussion is that other game modes don't offer enough rewards to do anything other than skirmish, e.g. you can play a light, cap conquest, win the match for your team, being the pilot of the match, and your PSR still gets a huge hit because you did zero damage.

When piloting a light 'Mech in Conquest, your task is not to cap every single base. If something, you just have to make sure that your team will not lose on caps.

As long as you keep one eye on the gauge that shows the control over bases, you are free to fight enemy 'Mechs as you wish

[Michael Abt]

martian, on 30 December 2021 - 06:00 AM, said:

When piloting a light 'Mech in Conquest, your task is not to cap every single base. If something, you just have to make sure that your team will not lose on caps.

As long as you keep one eye on the gauge that shows the control over bases, you are free to fight enemy 'Mechs as you wish

Neither did I suggest that capping is the only job for lights in conquest, or to cap everything, nor is your statement a contradiction of my statement.

See, if the enemy team has lights capping, your lights have to do exactly what you suggested, make sure your team doesn't lose on caps. I am pretty certain you also experienced games when the last mech to kill is a light who did just that. He may have been off better shooting stuff right from the start. Even when the match result, loss, is still the same, in terms of his own PSR his result would have been better.

There have been enough threads discussing game modes debating these issues though.

Edited by Michael Abt, 30 December 2021 - 06:57 AM.

[KaptinOrk]

GoodTry, on 29 December 2021 - 10:36 PM, said:

That's a great way to play IMO, and I feel your pain on teams that aren't aggressive. It makes everything so much harder. You don't have to go full NASCAR to be aggressive.



I agree you should play whatever is fun and try not to worry about tier. All tier is in reality is an automatic difficulty adjustment.

If you are going to "force yourself to git gud" at something though, make it a good meta build off of Grimmechs, not a build that is actually bad. You're not going to get good by intentionally playing a bad build. Plus that's no fun.



Just group with a couple of tier one players until you are sick of having your face smashed in :-). If you don't know any, join a unit - there are still units out there recruiting new players who are friendly but aren't yet skilled.

I feel like NASCAR doesn't really happen in T5/4, I've really only seen it in the T3 games that I play. I don't get why it happens, the enemies are in front of us, why are we turning and running instead of shooting them?

I don't play intentionally trash builds, mostly off-meta mechs without the best quirks, hardpoints or the current best weapons. I know that DPS builds with high facetime are bad in the current PPFLD meta, but it won't prevent me from playing my favorite BJ-1 with AC/2s.

I've considered joining a unit, I just haven't found one yet. Maybe I'll make that a New Years thing.

Horseman, on 30 December 2021 - 12:10 AM, said:

Sometimes it happens, yes. The most you can do is wait for opportunities to get in additional damage (=score) and survive long enough to get as many of them as possible so that you're not losing PSR after the match ends. Proximity towards friendlies triggers multiple score tickets that help with that.

martian, on 30 December 2021 - 12:50 AM, said:

It works only if the PUGs do not scatter in all directions ... which is the more common case.

The "scatter immediately and get picked off" games are the bane of my existence, we're always left 1-2 'mechs down within the first 2-3 minutes of the game and by that point, it's effectively over for us. Such is life for a support player. No amount of voice or text chat will convince players not to YOLO into the enemy and die before doing 50 damage.

[dubstep albatross]

Horseman, on 30 December 2021 - 03:47 AM, said:

Under the old system neither mech nor skill were very relevant - it was designed to float everyone to T1 over time, and the static tresholds for that were below average player performance (and that's before we got the ~10% match score inflation after the Skill Tree).

Your comment here, particularly the last part in parenthesis, caught my attention. Would it be possible for you to provide a reference (link, post, etc) for the match score inflation statement? Not only am I interested in the data and results, but also in the methodology.

I ask because I have done my own data collection (and experiments, with statistical analysis) regarding the impact of playing a fully skilled mech versus its fully unskilled counterpart. I examined a number of metrics (match score, damage, psr, kills, survival, to name a few). In multiple experiments, I found about a 10% increase in match score for a fully skilled chassis. When segmented in various ways, the results get even more interesting.

So, like I said, your comment caught my attention.

[martian]

Michael Abt, on 30 December 2021 - 06:56 AM, said:

Neither did I suggest that capping is the only job for lights in conquest, or to cap everything, nor is your statement a contradiction of my statement.

Sorry, this part of your post confused me:

Michael Abt, on 30 December 2021 - 05:41 AM, said:

PS thought: A variation of that discussion is that other game modes don't offer enough rewards to do anything other than skirmish, e.g. you can play a light, cap conquest, win the match for your team, being the pilot of the match, and your PSR still gets a huge hit because you did zero damage.

Michael Abt, on 30 December 2021 - 06:56 AM, said:

See, if the enemy team has lights capping, your lights have to do exactly what you suggested, make sure your team doesn't lose on caps. I am pretty certain you also experienced games when the last mech to kill is a light who did just that. He may have been off better shooting stuff right from the start. Even when the match result, loss, is still the same, in terms of his own PSR his result would have been better.

I think that this is okay. There have been also games when capturing bases lead to victory, even if the team failed in the direct combat.

If this is just a random defeat because his "half-time fighting/half-time capping" tactics did not work in this particular game, then some small PSR loss means nothing.

And if he always uses the "caps only" tactics in Conquest, his PSR goes down and eventually he will move to a Tier that is the most suitable for him and his play style.