215 replies to this topic

[Arn0ldSchwarzenegger]

ShinVector, on 18 August 2014 - 10:09 PM, said:

Because it hard where there are many equal skilled players. The damage and kills get balanced... However mechs like heavies and assaults that can deal a lot of damage in a short period of time can over come this somewhat... Much harder on mediums and lights though..

Tournaments like this, in a light you have to work SOOOOOO hard, paying attention to your radar and what your team is doing constantly. So easy to get caught out, especially with other decent light pilots on the other side have to constantly stay one step ahead, after a days playing in a light in this tourny i was shattered... I honestly didnt want to play on monday because i was getting sooo tired of it. I wanted a relaxing weekend but I worked harder playing this than i did at work during the week!

[Kaeb Odellas]

This was real fun. It's a lot less stressful when you know you're not going to make it to the top and can settle for top 15. I think I clocked in at somewhere around 20-ish hours this weekend. If I'd played more on Sunday and played at all on Friday I'd probably have made top 5 in the IS Light category. If I'd won my last round (5 kills, 4 assists, 763 damage) I would've been 7th. It was a real heartbreaker.

I think the dumbest round I played was a Terra Therma match where I found Shadow Hawk, Wang, and Kit Fox standing in a single 200 meter file on one of the rock ramps leading up to the volcano, all focused on the volcano action and oblivious to what the others were doing. I managed to kill the Shadow Hawk with 2 rear CT alphas without him or the others noticing, disarmed the Wang and killed him by legging without the Kit Fox noticing, and then the Kit Fox by legging, though another Kit Fox did notice and tried to protect him in vain. I think I ended up killing that dude too. Hilarious round. I think I had something like 7 kills and 4 assists or something.

[Eglar]

The only thing I found disturbing is that having a tie gives you 0 points while losing still gives you 5 points. I know it's not a huge issue but at times confusing and I've had 2 matches during the last 4 tournament where the 5 points would have made a difference.

[Baron Blitz Fokheimer]

As someone who did pretty well, I have to say that the scoring system is an absolute joke. Including kills in the calculation is ridiculously stupid, much less emphasizing them. Assists are hardly relevant too.

If you want to make it fair, use a tonnage-adjusted damage, or a damage-per-target metric, because the player who softens up the entire enemy team throughout the round and then scores the win by taking out the last two enemies two-on-one is worth more than someone who stops shooting at an enemy during the period of time when their CT goes from yellow to dark orange. And then include points for various support actions, while also making wins count for a more significant portion of the final score.

Or at least start awarding kills to those who do the most damage, and not those who land the final blow, although this would create its own, separate set of issues.

Let's make a quick comparison:

1 kill, 11 assists, 1250 damage, victory = 233.3 points
8 kills, 2 assists, 900 damage, loss = 245.0 points

Anyone who doesn't see an issue with this is blind.

Start scoring people based on how useful and efficient they are, and not on how often the stars align in order for them to have those "perfect" matches.

Edited by Baron Blitz Fokheimer, 19 August 2014 - 12:21 AM.

[Navid A1]

well... it was fun. Congrats to the winners.
I ended up 25th on the clan assault leader board with 2241 points.
hmm... not bad, considering that i have never touched arty/air and did it in a non-meta classic prime dire wolf.

Also, i agree that the scoring system could have been chosen better... so many times my hiding team mates showed up at the last moment to collect my kills.!

Edited by Navid A1, 19 August 2014 - 12:31 AM.

[Mr David]

I found the tourney to be an interesting play experience. The most notable difference in gameplay was the aggression displayed by all players. Everyone was jumping on crit components in a heartbeat to get that kill. Great focus fire all around, but tragically that meant many short matches that didn't allow enough time to run up my damage scores. Other players were just as intent on stealing the same kills I tried to steal too. Conquest was always played as a deathmatch, which broke my strategy of engaging small groups aggressively and then moving on to the next to maximize kills and assists. Nearly every match had multiple leaderboard guys on either side which made for tight games. They were fun matches for sure, but not wanted when the goal is inflated scores to get $3 worth of MC.

I found these play qualities developed over time. I jumped into the top 15 after about 20 matches and hit #7 after 25, all of which seemed like typical matches. I only gained about 30 points in the 6-8 hours following hitting #7 and couldn't get a solid scoring match to save my life. The competition really heated up! Cheers to the winners and placers. This was a fun,frustrating, challenging, and unique experience for me. I always had other plans during previous tourneys, but I enjoyed the couple of hours I put in to this one.

[iRON KANSAS]

Is tournament ended?

[Surien]

I have to say I still didn't completely understand scoring system (I've no problems with maths but I'm not a native english speaker, maybe it does something).

I guess the system takes your 10 best matches and do a sum of these matches

This is the formula:
(Kills × 20) + (Kill Assist × 10) + ((Damage Done - Team Damage) ÷ 15) + (Wins × 20) + (Loss × 5)

If I do for example a match with 5 kills, 9 assists, 650 dmg and 40 team damage (I know it as a really nice match) it will result into:

100 for the kills, 90 for the assists, 40 for damage (650-40/15) + 20 for the win = 170

To be in the top 15 the only chance then is to make more than 5 kills and 9 assists and 1000 damage in these 10 matches, am I right?

Forget the previous lines, my example match (wich was not so bad) was 5 kills 6 assists 640 damage difference (more or less) and win so: 100+60+40'67+20 = 220,67 (Thanks Baron Blitz Fokheimer)

Another question, what do you think is the best escenario mode to this contest, skirmish maybe?

Thanks to all!

**: I made a 160th mark, with 1768 points. I think it's nice but wow guys, the top 15 sound tooooo far!

Edited by Surien, 19 August 2014 - 02:15 AM.

[xhrit]

Baron Blitz Fokheimer, on 19 August 2014 - 12:18 AM, said:

If you want to make it fair, use a tonnage-adjusted damage, or a damage-per-target metric, because the player who softens up the entire enemy team throughout the round and then scores the win by taking out the last two enemies two-on-one is worth more than someone who stops shooting at an enemy during the period of time when their CT goes from yellow to dark orange. And then include points for various support actions, while also making wins count for a more significant portion of the final score.

80 points of damage in the rear center torso is enough damage to kill pretty much any mech. I rarely score more then 800 points of damage, even if I kill the entire enemy team myself. Any sort of metric involving damage is skewed against people who actually aim to kill.

Surien, on 19 August 2014 - 01:54 AM, said:

If I do for example a match with 5 kills, 9 assists, 650 dmg and 40 team damage (I know it as a really nice match) it will result into:

100 for the kills, 90 for the assists, 40 for damage (650-40/15) + 20 for the win = 170

To be in the top 15 the only chance then is to make more than 5 kills and 9 assists and 1000 damage in these 10 matches, am I right?

I made #32 on the clan light list with 1,812 points. I could get 6 kills easy enough, but I just couldn't dish out the damage...

Edited by xhrit, 19 August 2014 - 02:08 AM.

[Baron Blitz Fokheimer]

Surien, your math is really bad. First of all, you can't have 5 kills and 9 assists in a match. Second, even if you could, the answer for your example should have been 250.67 points.

To put things in perspective, the top Clan Assault guy has 3025 points. That's the equivalent of making 6 kills, 6 assists, doing 1,500 damage with no team damage, and winning the round. Ten times.

xhrit, on 19 August 2014 - 02:03 AM, said:

80 points of damage in the rear center torso is enough damage to kill pretty much any mech. I rarely score more then 800 points of damage, even if I kill the entire enemy team myself. Any sort of metric involving damage is skewed against people who actually aim to kill.

Right, I understand. But do you do this in a heavy, or an assault mech, on a consistent basis? Because that's where the weight adjustment comes into play. If you take out 10 guys with backstabs, at 80 damage a pop, in a smaller mech, then your results will be appropriately weighted. Something tells me you don't go around coring spines in an Atlas.

Edited by Baron Blitz Fokheimer, 19 August 2014 - 02:10 AM.

[Surien]

Baron Blitz Fokheimer, on 19 August 2014 - 02:04 AM, said:

Surien, your math is really bad. First of all, you can't have 5 kills and 9 assists in a match. Second, even if you could, the answer for your example should have been 250.67 points.

To put things in perspective, the top Clan Assault guy has 3025 points. That's the equivalent of making 6 kills, 6 assists, doing 1,500 damage with no team damage, and winning the round. Ten times.


Right, I understand. But do you do this in a heavy, or an assault mech, on a consistent basis? Because that's where the weight adjustment comes into play. If you take out 10 guys with backstabs, at 80 damage a pop, in a smaller mech, then your results will be appropriately weighted. Something tells me you don't go around coring spines in an Atlas.

Man, you're right, I don't know where from I've taken that numbers LoL. And the assists thing, yes It's a typo I was 5 kills and 6 assists

Thanks anyway!

[Thrudvangar]

xhrit, on 19 August 2014 - 02:03 AM, said:

80 points of damage in the rear center torso is enough damage to kill pretty much any mech. I rarely score more then 800 points of damage, even if I kill the entire enemy team myself. Any sort of metric involving damage is skewed against people who actually aim to kill.



I made #32 on the clan light list with 1,812 points. I could get 6 kills easy enough, but I just couldn't dish out the damage...

wasn't it you on canyon running around the edges of the map in circles with your ecm spider as the last one of your team like 6 minutes avoiding to get killed because of your K/D ratio? i think yes!

[Arn0ldSchwarzenegger]

The thing is, when i drop in a tournament, I dont change my style, i dont go out steal kills or play differently. I play as I always do, its just how I do it. Same games work in your favour. Sometimes you make the perfect decisions and do amazing, sometimes you get caught out and fk it up. I Make maybe slightly more risks in a tournament because if it pays off then you do epic but as for my style, it doesnt change, I dont pick a certain game type. I just drop and pew pew. The team winning is still priority.

[Tarogato]

shad0w4life, on 18 August 2014 - 02:53 PM, said:

It all comes down to who has the most free time really, you need to get some drops on good maps etc.

I agree. If you have enough time to play, you'll eventually get 10 good matches on Frozen City, which is better odds to score high than if half your good matches were on a hot map. To make the event briefer however (best 10 out of first 50 matches) would make it slightly luck based on maps dealt.

zaepp, on 18 August 2014 - 02:15 PM, said:

in my opinion, the current Scoring-System for Tournaments is heavily flawed because it largely encourages

- kill-stealing
- touching every opponent (just a little)
- trying to get that next "magic" (aka "derpy") game

...
Dealing at least 51% of all damage on a specific component, that leads eventually to a kill, gets rewarded like a kill.

I was actually thinking about this yesterday when I was in game: what if we were rewarded significantly per damage dealt on the component that eventually led to the takedown? Then, if you "stole" a kill off of somebody you hadn't even touched, you wouldn't get as much credit as the person(s) who whittled the core down. And also, the calculator would figure in component destruction. This puts emphasis on pinpoint accuracy as well as consistency.

Ex. just in case:
Mate R cores out a DWF's RT, Mate L cores out the LT, and Mate C cores out the CT. Mate Z comes along and finishes both side torsos and he gets credit for the kill and damage done. Mate C doesn't get the bonus for damaging the killing component, because the CT never popped, he gets damage and a standard assist credit. Mates L and R get a hefty bonus for critting out the side torsos because they led to the kill.

[Catra Lanis]

mwotourney by Videopac1, on Flickr

Edited by Catra Lanis, 19 August 2014 - 03:44 AM.

[Baron Blitz Fokheimer]

If they insisting on keeping kills as part of the formula, then they should at least be weighted by the amount of damage done as a proportion of the total damage done to the enemy team.

So if all twelve enemies die, here are two examples:

1. Player A scores 6 kills, more or less stealing them. Player A did 500 damage of a total of 2,500 done to the enemy team. This means that on average, one enemy needed 208.33 damage to die, but the player did only 83.33 damage per target. Player A's kill score gets multiplied by (83.33/208.33). Normally. this would equate to 153.33 points using the current formula. Adjusted, it would become 81.33 points.

2. Player B scores 4 kills, playing normally. Player B did 850 damage of a total of 2,500 done to the enemy team. This means that on average, one enemy needed 208.33 damage to die, and the player did 212.50 damage per target. Player B's kill score gets multiplied by (212.50/208.33). Normally, this would equate to 133.33 points using the current formula. Adjusted, it would become 134.93 points.

Of course, the formula would need to be more complex, in order to not punish players who get kills on almost dead targets, or reward players who inefficiently chip away at their targets to buff up the multiplier. Maybe use a logarithmic scale for the multiplier? And definitely add weight to the multiplier by taking into consideration the tonnage of the mech used.

Still, something along these lines would be fundamentally more fair than what we have now, because no matter how you go about reasoning about the system, a person who whittled down an enemy's armor and exposed the juicy, red core, was more valuable to the team than someone who slid lasers across the almost-dead target at the right moment.

I could spend a bunch of time on ironing out the formula, but they're not paying me for my time. In fact, I'm paying them. So they should get to it.

Edited by Baron Blitz Fokheimer, 19 August 2014 - 03:54 AM.

[Arn0ldSchwarzenegger]

Its unfair in all directions in some manner, adapt and overcome. Only you know if your a kill stealer or someone who deserves their place, only cheating yourself!

[Baron Blitz Fokheimer]

Do you know how people adapt to and overcome fundamentally flawed systems? By fleecing the crap out of them. Which is exactly what I, and pretty much everyone in the top ranks has done. At the end of the day, the system doesn't differentiate between players in terms of how much honor they possess. And no one can see how anyone else made those points either. Morality becomes very loose when there's anonymity involved.

[ShinVector]

Is there a point to all this whining ?
How many competitions let the competitors decides the rules ?

Can't compete just try harder next time..

On another note.. Couldn't do as well as I hoped.. But congrats to the winners.
All I know is, most of the time I came across people from the light leader board... I gave them a hard time. LOL.

Edited by ShinVector, 19 August 2014 - 04:39 AM.

[Cpt Zaepp]

Eglar, on 18 August 2014 - 09:53 PM, said:

Make different leagues than but i am not willing to give "worse" players that edge.

And I am not willing to give "good" players an edge, just because they ended up in a match full of "lesser talented players".

Eglar, on 18 August 2014 - 09:53 PM, said:

"the best because you prepared your elo by losing 200 matches in a row".

I don't even come up with such ideas. But losing 200 matches in a row, is that even possible? And since nobody knows actual Elo and matchmakers prediction you wouldn't even know if you had really decreased your Elo.

So,I don't think so.

But yes, there are always people who exploit the system, whichever system you have.

Edit: BTW, artificially lowering your Elo in order to get into more "derpy" matches is already possible. Or simply take a smurf-account.

Edited by zaepp, 19 August 2014 - 05:27 AM.