Ballistic Pro Ladder
-
- Disappeared Administrator
- Posts: 4196
- Joined: Fri 19 Mar , 2010 1:21 am
- Location: Earth
- Contact:
Re: Ballistic Pro Ladder
It's not implemented yet, although it is planned based on weapon groups (hangs on Aza right now).
- CaptainXavious
- V.I.P. Member
- Posts: 269
- Joined: Tue 07 Jun , 2011 2:49 am
- Contact:
Re: Ballistic Pro Ladder
Awesome, can't wait!
- iRobot
- Junk Administrator
- Posts: 3909
- Joined: Fri 06 Jan , 2012 10:37 am
- Contact:
Re: Ballistic Pro Ladder
Can't wait to see all that Firestorm and M50 usage.
- Azarael
- UT2004 Administrator
- Posts: 5365
- Joined: Thu 11 Feb , 2010 10:52 pm
Re: Ballistic Pro Ladder
Weapon groups, guys. Assault, SMG, etc.
- iRobot
- Junk Administrator
- Posts: 3909
- Joined: Fri 06 Jan , 2012 10:37 am
- Contact:
Re: Ballistic Pro Ladder
Without sounding ungrateful, what would be the point of that?
We already know that 95% of players just roll with an assault rifle.
If you knew per weapon, rather than per group, then you would be able to use it for balance purposes.
We already know that 95% of players just roll with an assault rifle.
If you knew per weapon, rather than per group, then you would be able to use it for balance purposes.
-
- Disappeared Administrator
- Posts: 4196
- Joined: Fri 19 Mar , 2010 1:21 am
- Location: Earth
- Contact:
Re: Ballistic Pro Ladder
Both ways have their purpose. Speaking more practically, the groups is nearly finished, and it's not that much work to make that per weapon either.
- focus
- Member
- Posts: 232
- Joined: Wed 15 Aug , 2012 10:28 pm
- Location: Moscow, Russia
- Contact:
Re: Ballistic Pro Ladder
My suggestion is to calculate player's skill not just averaging all his deeds over all time playing here, but weight more recent actions more then old ones.
For example, you can use an exponential curve for weighting over time.
50% of skill is formed by actions not more than 4 months ago. Rest 50% is formed by all remaining actions.
Additionally, i believe that there can be better approach to calculating skill then based on kill, death, thaw points.
IMO, approach with recording who killed whom is more fair. For example, if in two players fight with same skill,
one killed another, then he is a better player. Balancer can record this statistics and adjust skill for those
players. More than that. If gizmo kills me, thats an routine matter, and his skill shouldnt move much.
But if i kill him, i should get an considerable boost in skill.
For example, you can use an exponential curve for weighting over time.
50% of skill is formed by actions not more than 4 months ago. Rest 50% is formed by all remaining actions.
Additionally, i believe that there can be better approach to calculating skill then based on kill, death, thaw points.
IMO, approach with recording who killed whom is more fair. For example, if in two players fight with same skill,
one killed another, then he is a better player. Balancer can record this statistics and adjust skill for those
players. More than that. If gizmo kills me, thats an routine matter, and his skill shouldnt move much.
But if i kill him, i should get an considerable boost in skill.
- iRobot
- Junk Administrator
- Posts: 3909
- Joined: Fri 06 Jan , 2012 10:37 am
- Contact:
Re: Ballistic Pro Ladder
I'd like a decay system. As it is now, I can just get to the top of the ranking and never play again, and retain #1 forever until someone passes it.
It should go down gradually.
It should go down gradually.
-
- Disappeared Administrator
- Posts: 4196
- Joined: Fri 19 Mar , 2010 1:21 am
- Location: Earth
- Contact:
Re: Ballistic Pro Ladder
Such approach is far more complex to implement and while it's true that it would eliminate the possibility of farming (which is now being punished manually if goes beoynd lameness, such as winning team joining etc.). I'm not convinced we actually need that.focus wrote:My suggestion is to calculate player's skill not just averaging all his deeds over all time playing here, but weight more recent actions more then old ones.
For example, you can use an exponential curve for weighting over time.
50% of skill is formed by actions not more than 4 months ago. Rest 50% is formed by all remaining actions.
Additionally, i believe that there can be better approach to calculating skill then based on kill, death, thaw points.
IMO, approach with recording who killed whom is more fair. For example, if in two players fight with same skill,
one killed another, then he is a better player. Balancer can record this statistics and adjust skill for those
players. More than that. If gizmo kills me, thats an routine matter, and his skill shouldnt move much.
But if i kill him, i should get an considerable boost in skill.
I'm intending to make a decay system based on additional reducing the weighting rate based on inactivity (fully reversible). However given the fact that the tracking is on only for 14 days, I'll give priority to finishing the team balancer.iRobot wrote:I'd like a decay system. As it is now, I can just get to the top of the ranking and never play again, and retain #1 forever until someone passes it.
It should go down gradually.
- Butcher
- V.I.P. Member
- Posts: 893
- Joined: Sat 15 Sep , 2012 1:31 pm
- Location: Germany
- Contact:
Re: Ballistic Pro Ladder
Hi All,
I suggest a number of games buffer to get an average of the skill of the player in a reasonable movable scale. Say last 10-20 games movable average only. Players that game often will not see big changes in their scores because of the weighted average carried on during all the time. The kill,death and thaw method is good to find a reference value for the skill level.
Maybe like this we will not see the same players in the same positions statically, and there is also a chance of being in the list (I can dream too )..
I suggest a number of games buffer to get an average of the skill of the player in a reasonable movable scale. Say last 10-20 games movable average only. Players that game often will not see big changes in their scores because of the weighted average carried on during all the time. The kill,death and thaw method is good to find a reference value for the skill level.
Maybe like this we will not see the same players in the same positions statically, and there is also a chance of being in the list (I can dream too )..
"An BW match is a test of your skill against your opponents' luck."
Who is online
Users browsing this forum: No registered users and 1 guest