97 replies to this topic

[Moment Killer]

I had fun with the tourney I suppose but a few things I wasn't completely on board with.

- Tag/Spot in the equation. As this is something that is dependent on your teammate having missiles, it doesn't completely apply to the individuals skill level. Which brings me to the next thing.

- Being in a tournament against people who arent. It's a different game when you're fighting against other people fighting for the same thing. It makes it difficult for the tournament folk having to fight against premades and people maybe just running a game to check out a new tweek on their mech.

I might even think either a tournament bracket would be nice. Or perhaps a 4 v 4 with the people in the tourney. The reason is I feel the less people on the drop the more it shows an individuals skill.

I would have a question to ask also. Is there any info out on when team tournaments will be something to look forward to?

[Kaeb Odellas]

WOW

Okay, I did not expect that! I suppose it was kind of weird that my great last-minute 8-K/A + 12 TAG/Spot game only seemed to bump my score up by 80 points.

While it's true that the scoring system didn't actually reward particularly effective play, it was by no means easy to pull off those high scores. Games without LRMs on your team were basically a wasted match, and there were quite a few of those on the heels of the LRM nerfs. And unless you had multiple LRM mechs on your team, getting all those spotting/tag assists required very particular maneuvering and situational awareness on your part. You also still had to land a hit on all of your enemies, and then ensure that they actually died in order to get the points for the assist.

So not only did you have to play this crazy TAG minigame, you still had to go on and basically win the match, too, or at least play a very close game. The scoring rules definitely added an extra layer of challenge, which while not entirely unwelcome, could have been better handled I think.

It wasn't quite the ridiculous timesink that the first tournament was, but you still had to spend quite a bit of your weekend on matches to get the right team compositions too, unless you were simply dropping out whenever you get an LRM-free match, but that would make you a butthead.

Congratulations to the other winners! Good games, all!

Edited by Kaeb Odellas, 26 March 2013 - 05:15 PM.

[w00tzor]

Thanks guys! I Played really a lot to improve my score near the end, i was scared to lose just at the end for those 10 points!

Gratz to all guys that i met and i fought, gg!

[TOGSolid]

OOSIKS
OOSIKS
OOSIKS
OOSIKS

Grats Kaeb!

[jakucha]

Zolaz, on 26 March 2013 - 03:11 PM, said:

Not impressed with with this tournament or the way that PGI supported it.

That being said, I do want to offer my congrats to all of the winners for sticking it out. I do wish that the game would have recorded my matches better and that there was more transparency in seeing my score. Being able to see a breakdown of my top 20 matches would have been awesome.

Don't worry about it, it's mostly to collect stats and data.

[Kashaar]

Oosiks! Wheee! We're all very proud of our very own Kaboodles, and will from now on tolerate his incessant bitching about Raven-3Ls with immeasurable patience and grace

[Naja]

Definitely a good learning experience for the Devs on scoring systems, looking forward to future tournaments.

And of course, congrats to the victors

[Targetloc]

Rewarding both sets of winners... classy move. Thumbs up, PGI.

[Homeless Bill]

You guys get better every tournament, and the new format is how it should be (best 10 out of X rounds).

One suggestion is that you should base score on role warfare. Having NARC/TAG/Spotting Assist bonuses for heavies is kind of ridiculous. I know that within every weight class, different 'mechs fill different roles, but I think it's clear that heavies shouldn't be acting as LRM spotters.

Off the top of my head, lights should be rewarded more for capturing / spotting, mediums should be rewarded fairly equally for everything, and heavies / assaults should focus on damage output, component destruction, and kills.

One glaring problem I see that current undermines any scoring system you implement is that damage and kill assists aren't very good statistics.
-Kill Assists: Whether you took an Atlas down 1v1 and a Raven stole the kill shot or you touched someone with a small laser, you get an assist. Assist score shouldn't be the number of dead people you touched. It should be some metric based on how much you helped kill other 'mechs.

Suggestion: You could base it on the percentage of damage a player did to the target 'mech (or whatever). The point being that there needs to be a distinction between a poke and a beatdown.

-Damage: As a sniper, you will generally do less damage than other people, but that doesn't mean you're less useful. Quite the contrary, I'd rather have 300 damage of sniper damage concentrated on torsos than 600 damage of SRM spam.

Suggestion: You guys should make a system that differentiates useful damage from damage that doesn't matter. I'd base it on damage done to the component that killed the enemy (center torso most of the time, left or right for XL, leg or arm if there was an ammo explosion), but for simplicity sake you could also just go with torsos (maybe center only for STD, all three for XL) and head.

The guy that cores out the Atlas' CT for me deserves a cookie; the clown that then manages to take off both the left and right torsos of that same Atlas without killing it deserves only my shame.

I know it'll never be perfect, but I appreciate you guys putting these things on and actively listening and improving. Keep up the good work!

[BlueSanta]

I'm glad you discovered a problem and corrected it, and are rewarding both sets of winners. That's the kind of ethics and behavior I'm looking for from PGI. Thank you for being up front about the error.

The ten best matches format is the way to go, but still, you had people exploiting to get the best possible results (saw a Jager with dual Gauss and a TAG just to get the TAG bonus). I'm glad you rewarded kills and kill assists equally, though.

[Zabadoo]

Good match guys.
Now for the constructive criticism.
How about instead of a 3 day run of grinding and figuring out the loopholes on how to get the best score. This is what you do.
The same thing almost. But.... Big BUT, you only count the 1st say 30-40'matches while using the top 10. It makes it fair to people who have a life who can't play 3 days straight.
Just looking at your scoring it would take a miracle to get 500 let alone the average 700 from those 7000+ scores. Without help from people who had sync dropped.

[Mattiator]

And this only provides further proof that the Dragon is the best heavy.

[Loc Nar]

My tag line for the weekend...

Big maps, big mechs, big fun.


MOAR!

[Nikoliy]

Perhaps a good way to keep people from trying to exploit the scoring system is to keep the scoring system a secret until the tournament is over. If PGI sets up 3 scoring systems witch prioritize support, firepower & kills with an additional overall scoring board. That way people will play to the best of their ability.

[Felix Reynolds]

Congratulations to all the winners, I would also like to see an expanded leaderboard as I was one place away from having a spot in your shiny post prior to the recalc and my vanity needs to know...

Have put a longer response in the feedback thread, but overall really well done PGI, can definitely be improved but is a definite improvement over previous ones.

Biggest factors for me to address are: requirement of a 'randomized' component in your team (LRMs) in order to get the best possible score, and the fact that grinding still helped immensely- I would prefer to see a change whereby the best 10 out of a capped number of games (perhaps 25) are taken, and what the 'opt-in' button does is starts the recording.

Ie, Jim Bob practices in his Dragon on Friday and Saturday, then on Sunday sets aside 3-4 hours and is feeling good about himself, so he clicks the 'Opt-In' button and now the next 25 matches he plays in any of his Heavies are used to count his score. After 25, games go back to being normal.

I feel that would remove the grind and the 'wait to see if the match is going to be worth it' mentality that many had when waiting to see if they had gotten dropped with a group of 7 AC20 Jagers and knew the whole match was a scrub because they wouldn't be able to get TAG/Spotting assists.

But, again, bravo and thanks for a fun weekend!

[Mister Blastman]

Congrats to everyone! I enjoyed it! I did make a mega-post here:

http://mwomercs.com/...-world-winners/

But, to summarize for PGI, my thoughts on the tournament are as follows:


I felt that the format this time around was far better than the last couple of tournaments we have had. Unlike the first one that rewarded quantity far above quantity and the second one which had goofy scoring--this third one definitely came closer to showcasing true skill and ability unlike the others. It also wasn't perfect as who would of thought heavy mechs need to carry tag and spot targets? I didn't. Regardless... Why do I feel it took more skill:

a. Your team had to win the match. If they didn't--then you couldn't possibly get all 8 kills/assists. It was paramount that you tried your hardest to make the enemy die. No longer could you run in, touch every enemy and then either suicide in a blaze of glory or quit the game. You had to stick it out and preferably, if your really wanted the game to count--survive til the end. This alone had a huge impact on how it was played. I liked this the best.

b. Tagging with a big mech is HARD! You had to get close to the enemy and expose yourself long enough for the missiles to hit. Yes, you could tag them at the last moment before they hit but still, it required you to reach far outside your comfort zone and put your mech at risk to the enemy.

c. Spotting was also hard and required exposure when you normally wouldn't.

d. 10 best games played. You couldn't play all weekend and expect to come in first just due to racking up points. You had to perfect your game. You had to do it again if you felt you could do better. Sure, people get lucky but to get lucky ten times in a row? Not nearly as likely.

Now, with those points said, there are some things I did not like and feel PGI could have done better/differently...

a. Tag/Narc... doesn't belong on Heavy Mechs. Maybe lights, but not heavies.

b. LRM spotting after an LRM nerf? This required luck more than anything. Not to get an assist--but to get players on your team with LRMs! I remember many a game sighing to myself as I looked at my teammates, asking them if they had missiles only to get a no from them. This alone was a big problem with the tournament and did not reward skill so much as it was more dependent on the luck of the draw.

c. ELO bracketing or lack of. The tournament grouped everyone together in one group. However, some of us have very high ELO scores (or at least we /think/ we do). Who knows. In the future, it might be good to have PGI bracket us within our ELO category and compete against like-ranked warriors. This may or may not be possible but it is worth considering. At the high end of the ELO ladder the battles can be very intense... or not so much. When they aren't, it is when you get a lot of rookies on a team versus a lot of good or above average players. This typically results in being rolled and isn't fun for anyone--both the winners and the losers.

d. Perhaps in the future base points on a combination of match score / kills / assists / damage done (to a lesser degree) and if you team won/lost. I think having a mechanic in the system that requires you to influence your team to victory will prevent a lot of farming that we have seen in the past. Probably also contain a heavier weighting towards match score as well.

e. I'd love to see a 1on1 tournament of some sorts. Long gone are the days of honorable dueling. It was a time-honored tradition back in the day in both Mechwarrior 2 and 3 that Warriors would test themselves versus others with only their mech and weaponry and noone else. If not a tournament, a ladder perchance. I know Mechwarrior has /always/ been about team play in leagues, but a great number of us from the old days value the thrill of the duel above all other. It is an amazing test of a pilots stamina and focus. This of course would require pre-arranged games, a lobby or ability to drop solo versus a solo combatant...

and, with that said...

f. A tournament of teams, say 4 man or 8 man.

[Targetloc]

Mister Blastman, on 26 March 2013 - 07:27 PM, said:

e. I'd love to see a 1on1 tournament of some sorts. Long gone are the days of honorable dueling. It was a time-honored tradition back in the day in both Mechwarrior 2 and 3 that Warriors would test themselves versus others with only their mech and weaponry and noone else. If not a tournament, a ladder perchance.

Maybe they'll do a C1 duelling tournament after they release the Stormcrow.

[Mister Blastman]

Targetloc, on 26 March 2013 - 07:50 PM, said:

Maybe they'll do a C1 duelling tournament after they release the Stormcrow.

That would be awesome!

[Kaeb Odellas]

Nikoliy, on 26 March 2013 - 06:38 PM, said:

Perhaps a good way to keep people from trying to exploit the scoring system is to keep the scoring system a secret until the tournament is over. If PGI sets up 3 scoring systems witch prioritize support, firepower & kills with an additional overall scoring board. That way people will play to the best of their ability.

No way, man! How can you play a game when the rules are a secret? The scoring system was posted with the announcement, so anyone who actually paid attention to the rules could have figured out the "winning" strategy (TAG AND SHOOT EVERYONE). I actually didn't understand the system for the first couple of hours and only looked up the rules again after seeing everyone's ridiculous high scores. I actually lucked out, thinking the Dragon ladder would be easier to compete in since most people would probably pick the heavier chassis, when it turns out the Dragon was actually the best mech to win the whole championship with (TAG assists are much easier with a 100kph+ mech).

Tournaments need a better scoring scheme, not a secret one.

[Kageru Ikazuchi]

@PGI:

Great job - much better than the last two rounds ... keep improving!

Scoring criteria aside, I recommend a best X of Y format ... given equal skill and luck between two players, the guy who plays more (i.e.: best 10 games out of 200 ... top 5% of his games) will have a better score than someone who plays less (i.e.: top 10 of 50).

Finally, I agree that spotting, TAG, and NARC assists probably shouldn't have been so prominent in the scoring.

@ Others:

In general, I'm not sure that Elo brackets won't be exploited (either by trying to tank one's Elo or creating a new account right before an event, for example) without some careful and/or sneaky planning.

(Edit: added this bit)
One idea ... reward both the best overall and use a handicap system (i.e.: a multiplier for your score based on your Elo) that "should" reward the best performance vs. handicap during that particular event.

Still open to exploiting, second guessing or violently loud QQing ... but this is the internet ... what do you really expect? A fair fight between equally matched foes and a "good game" from the loser?

Edited by Kageru Ikazuchi, 26 March 2013 - 10:39 PM.