Jump to content

MVP System Anybody?


40 replies to this topic

#1 Felix Dante

    Member

  • PipPipPipPipPipPip
  • Knight Errant
  • Knight Errant
  • 400 posts
  • LocationDallas, TX

Posted 31 January 2012 - 08:50 AM

I would like to see an MVP system implemented if possible.
:D


I'll leave the details up to the designers however...

But giving the best player from a group an MVP award based on overall performance I think would encourage many to play more often so they can get better at the game...

What does everybody else think?
:)

#2 Nik Van Rhijn

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,905 posts
  • LocationLost

Posted 31 January 2012 - 08:51 AM

What do you mean by MVP? Most Valued Player?

Edited by Nik Van Rhijn, 31 January 2012 - 08:51 AM.


#3 Aegis Kleais

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 6,003 posts

Posted 31 January 2012 - 09:10 AM

View PostFelix Dante, on 31 January 2012 - 08:50 AM, said:

I would like to see an MVP system implemented if possible.
:D


I'll leave the details up to the designers however...

But giving the best player from a group an MVP award based on overall performance I think would encourage many to play more often so they can get better at the game...

What does everybody else think?
:)

The ideal of whom is the "best" player by the masses always equates down to the person with the highest KDR. This is a misnomer to call that person a MVP of a tactical and strategy-based game because simply looking at a KDR does not indicate a player's worth. For example, a person with a 4:1 KDR is worthless if they repeatedly miss opportunities to prevent a team flag from going out or fail to assist our runners in acquiring the enemy flag. In other words, if they don't aid in the completion of objectives, but instead, are just mired in their own personal statistics.

A scoring system could not feasibly be created that would accurately be able to determine actual player value. There's just too many variables to account for.

#4 Paul Inouye

    Lead Designer

  • Developer
  • Developer
  • 2,815 posts
  • LocationVancouver, BC

Posted 31 January 2012 - 10:30 AM

View PostAegis Kleais™, on 31 January 2012 - 09:10 AM, said:

A scoring system could not feasibly be created that would accurately be able to determine actual player value. There's just too many variables to account for.


Oh reallllly?


CASE STUDY: Weighted Stats to Calculate a MVP in a FPS

All of the following scores are weighted between 0.00 and 1.00.

K/D ratio: 0.15
Saviour Kills: 0.20
Objectives Taken: 0.20
Objectives Defended: 0.20
Orders Followed: 0.10
Kill Assists: 0.15

As indicated in the above set of stats, I weighted teamplay higher than simple kills.

#5 Kaemon

    Member

  • PipPipPipPipPipPipPipPip
  • 1,924 posts
  • LocationMN

Posted 31 January 2012 - 10:40 AM

View PostPaul Inouye, on 31 January 2012 - 10:30 AM, said:


Oh reallllly?


CASE STUDY: Weighted Stats to Calculate a MVP in a FPS

All of the following scores are weighted between 0.00 and 1.00.

K/D ratio: 0.15
Saviour Kills: 0.20
Objectives Taken: 0.20
Objectives Defended: 0.20
Orders Followed: 0.10
Kill Assists: 0.15

As indicated in the above set of stats, I weighted teamplay higher than simple kills.


No enemies scouted?


hrmnnn...I'll take it that was for a generic FPS and part of this fictional 'paper' you wrote that we've never seen.

:D

#6 MaddMaxx

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Survivor
  • Survivor
  • 5,911 posts
  • LocationNova Scotia, Canada

Posted 31 January 2012 - 10:43 AM

Oooohhh! A possible look at MM calc stats.

It appears the "Save your ***" stat is a highly regarded one. :D lol

#7 Paul Inouye

    Lead Designer

  • Developer
  • Developer
  • 2,815 posts
  • LocationVancouver, BC

Posted 31 January 2012 - 10:55 AM

View PostKaemon, on 31 January 2012 - 10:40 AM, said:

No enemies scouted? hrmnnn...I'll take it that was for a generic FPS and part of this fictional 'paper' you wrote that we've never seen. :)


Yes.. generic stats. If you really want to read the paper I wrote, I can supply it to you with a NDA requirement.

View PostMaddMaxx, on 31 January 2012 - 10:43 AM, said:

Oooohhh! A possible look at MM calc stats. It appears the "Save your ***" stat is a highly regarded one. :D lol


Of course it is!! :(

#8 L zard

    Member

  • PipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 354 posts
  • LocationOregon

Posted 31 January 2012 - 11:30 AM

I'd agree to an NDA, Paul! Already have one with Matrix....keeps one laughing on the forums, eh?

#9 Kaemon

    Member

  • PipPipPipPipPipPipPipPip
  • 1,924 posts
  • LocationMN

Posted 31 January 2012 - 11:48 AM

View PostPaul Inouye, on 31 January 2012 - 10:55 AM, said:


Yes.. generic stats. If you really want to read the paper I wrote, I can supply it to you with a NDA requirement.



Of course it is!! :D

I'm always interested in the insane manifestos intellectual work of experts in their fields.

NDA away!

Edited by Kaemon, 31 January 2012 - 11:49 AM.


#10 Aegis Kleais

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 6,003 posts

Posted 31 January 2012 - 11:51 AM

View PostPaul Inouye, on 31 January 2012 - 10:30 AM, said:


Oh reallllly?


CASE STUDY: Weighted Stats to Calculate a MVP in a FPS

All of the following scores are weighted between 0.00 and 1.00.

K/D ratio: 0.15
Saviour Kills: 0.20
Objectives Taken: 0.20
Objectives Defended: 0.20
Orders Followed: 0.10
Kill Assists: 0.15

As indicated in the above set of stats, I weighted teamplay higher than simple kills.


So no MVP kudos for:
  • Relaying by recon via voice (an act that cannot be captured by game metrics)
  • Being an effective commander
  • Sacrificing yourself so that another can complete a crucial task
Thing is, there are plenty of things that the game simply cannot account for (despite Paul's OBVIOUS programming skills) which would affect my personal determination of whether someone is a MVP. It's too subjective a task. The game can't give you a MVP via the points you made above, but it should state: "MVP*" and at the bottom there should be a long winded disclaimer stating how the calculation for MVP ignores a lot of intangible factors in gameplay.

I'm your nemesis, Paul. I can't make this easy for you.

#11 Kaemon

    Member

  • PipPipPipPipPipPipPipPip
  • 1,924 posts
  • LocationMN

Posted 31 January 2012 - 11:58 AM

I've always wondered what the limitations are on things like this in a current game engine? Can we not setup predictive events, then daisy chain xp based on a players involvement in those events?

If we have complex objectives, could not the steps to accomplish those objectives be set to rules regarding xp?

i.e. Destroy turret guarded bunker

1. scout spots location (start xp chain)
2. commander decides to take objective (xp setup for following commands)
3. scout calls arty (xp for scout for following command)
4. arty targets turrent (xp for following command)
5. meds attempt diversary attack to allow scout LOS for arty (xp for follow command)
6. arty destroys target (xp for objective accomplished)

then breakdown xp according to involvement/actions taken.

#12 Mchawkeye

    Member

  • PipPipPipPipPipPipPip
  • 883 posts

Posted 31 January 2012 - 11:59 AM

I don't like the idea of MVP.

For no other reason than it makes me sad to be at the bottom of a list...It reminds me of getting picked last at sports in school.

Sigh.

#13 Kaemon

    Member

  • PipPipPipPipPipPipPipPip
  • 1,924 posts
  • LocationMN

Posted 31 January 2012 - 12:01 PM

View PostMchawkeye, on 31 January 2012 - 11:59 AM, said:

I don't like the idea of MVP.

For no other reason than it makes me sad to be at the bottom of a list...It reminds me of getting picked last at sports in school.

Sigh.


lol! Sorry I had to laugh at that post, it reminded me of Charlie Brown for some reason.

#14 GaussDragon

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,183 posts
  • LocationToronto

Posted 31 January 2012 - 12:09 PM

View PostPaul Inouye, on 31 January 2012 - 10:55 AM, said:

If you really want to read the paper I wrote, I can supply it to you with a NDA requirement.


NDA me. The paper, if you please.

#15 Kaemon

    Member

  • PipPipPipPipPipPipPipPip
  • 1,924 posts
  • LocationMN

Posted 31 January 2012 - 12:12 PM

View PostGaussDragon, on 31 January 2012 - 12:09 PM, said:


NDA me. The paper, if you please.


Why do I suddently feel like we're going to be inside the movie Barton Fink once we start reading this?

#16 Aegis Kleais

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 6,003 posts

Posted 31 January 2012 - 12:14 PM

View PostMchawkeye, on 31 January 2012 - 11:59 AM, said:

I don't like the idea of MVP.

For no other reason than it makes me sad to be at the bottom of a list...It reminds me of getting picked last at sports in school.

Sigh.

That's what's great about gaming. Now, your being overweight and having ripe B.O. will be hidden behind the anonymity of the internet! You're gonna get picked last for a legitimate reason now! A severe lack of skill!

Isn't technology great?

#17 Kenyon Burguess

    Member

  • PipPipPipPipPipPipPipPipPip
  • Veteran Founder
  • Veteran Founder
  • 2,619 posts
  • LocationNE PA USA

Posted 31 January 2012 - 12:18 PM

I dont like the term "MVP" is all. hand out awards and medals in a military fashion and please dont make players vote for who was the best.

#18 Dlardrageth

    Member

  • PipPipPipPipPipPipPipPip
  • 1,198 posts
  • LocationF.R.G.

Posted 31 January 2012 - 12:25 PM

View PostPaul Inouye, on 31 January 2012 - 10:30 AM, said:

Oh reallllly?

CASE STUDY: Weighted Stats to Calculate a MVP in a FPS

All of the following scores are weighted between 0.00 and 1.00.

K/D ratio: 0.15
Saviour Kills: 0.20
Objectives Taken: 0.20
Objectives Defended: 0.20
Orders Followed: 0.10
Kill Assists: 0.15

As indicated in the above set of stats, I weighted teamplay higher than simple kills.


Hm, maybe it is me, but that is only 6 factors there, isn't that a bit too simple? Unless you want it mainly for MM purposes instead of ranking ones. I would have expected at least hit/miss ratio and battle survival rate to be factored in. as well.

Also an interesting question would be if this is handled globally or on a "pilot per Mech" basis. Interesting discrepancies are to be discovered in some games when you swap out the vehicle/role for tae same player and look what happens to his stats. So for a MWO model, an excellent player in a Centurion could be barely mediocre in the Catapult. Thus the whole MVP concept might need to be applied per Mech type and not globally to reflect true capabilities there. IMHO.

#19 Paul Inouye

    Lead Designer

  • Developer
  • Developer
  • 2,815 posts
  • LocationVancouver, BC

Posted 31 January 2012 - 12:40 PM

View PostPaul Inouye, on 31 January 2012 - 10:55 AM, said:

Yes.. generic stats.


#20 Paul Inouye

    Lead Designer

  • Developer
  • Developer
  • 2,815 posts
  • LocationVancouver, BC

Posted 31 January 2012 - 12:43 PM

And don't get me wrong people... I'm only showing you a possible calculation for determining MVP without the need to vote... and to squash Aegis' claim that you can't come up with such an equation. HA! Take THAT!





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users