Get the Balance Right

[tokamak]

Ya that’s another thing. Xp includes the context in which you’re killing people. I had forgotten about that.

[spookify]

[QUOTE=tokamak;511900]I’m honestly wondering who of us is confused here.

xp/min doesn’t say anything about the time played [/QUOTE]

XP / “MIN”

MIN = Minute which is a unit of Time…

Here (This is From a Few Days Ago)

Lets find my “MIN”

As you can see I am level 20 with 1,636,539 Total XP
My SPM or XP/MIN is 577

Math! How Long Have I been playing?
Answer = 2,836.29 Total Minutes Played.

XP (1,636,539) / “MIN” (2,836.29) = XP/MIN @ 577

Note: There should be two different stats! Ranked and UN-Ranked STATS!!!

Interesting W/L is a Ratio! Makes sense!

Bronze 1 Rank (spookify) (Screen shot above)
This math is so wrong hahaha!! But SD can fix it!!!

1 * ((1.99 * 10 *5%) + ((((42% + 35%) / 2) * 100) * 10%)) + ((577 / 10) * 30%) + (((2.4% * 100)) * 55%)))
Rank * (K/D) + (Acc and HS Acc) + (SPM) + (W/L)

(20.90 + 42.35 + 75.01 + 372) * 1 = 510.26 Player Rank

Multiplier of Importance:
K/D 5%
Acc 10% (and Averaged)
SPM 30%
Win/Loss 55%

I would also add Total Wins!
or Total Season Wins

510.26 Player Rank + 200 Wins = 710.26 Player Rank

Something Like that…

OR SD could do something again like Clash of Clans that you put your “Wins” or Trophy’s on the Line every Match…
You win you get 5 Trophys… You loss, you loss 5 Trophys…

[Destroy666]

Every currently available stat has its flaws.

K/D:

• selfish players can freely switch to the dominating team and boost their K/D
• someone who always plays as defender will have better K/D
• there are players with good K/D but no value for the team. For example, yesterday I played a 6v6 in Underground and my attacking team had 2 Vassilis camping near spawn… They obviously ended up with the highest K/D since the others had no chance of doing anything alone near the capture objective.
• respawn selfkills, which are sometimes crucial for winning maps and used by obj-players, are counted as deaths

W/L:

• selfish players can freely switch to the dominating team and easily win maps
• maps are not well-balanced considering attacking/defending - for instance, winning Underground in a 8v8 as an attacker is nearly impossible
• W/L gets increased when you win against no opponent
• correct me if I’m wrong, but a player can join 30 seconds before a map ends and still gets W/L ratio boosted, which is stupid

XP/MIN:

• yet again, selfish players can freely switch to the dominating team and boost their XP/MIN
• you may throw ammo packs near spawn the whole map and gain quite high XP
• XP/MIN gets increased when you win against no opponent
• correct me if I’m wrong, but a player can join 30 seconds before a map ends and still gets 1000 XP/MIN, which is stupid

Thus I think introducing a 4th stat is inevitable. Let’s call it Player Rating, gained/lost after each map. What should it consider?

1. Its points should be based on all three stats - W/L, XP/MIN and K/D.
2. Points percentage should be based on the time played. So if a team finished a map in 6 minutes and PlayerXYZ played only 3 minutes, he should get/lose only 50%. For the W/L aspect at least.
3. Arithmetic means of both teams’ Player Rating.

• If team A has similar PRTAM (Player Rating Team Arithmetic Mean) to team B’s, they should get regular points.
• If team A has much bigger PRTAM than team B, they should get less points depending on the PRTAM difference.
• If team A has much smaller PRTAM than team B, they should get more points depending on the PRTAM difference.

4. Map-specific W/L. The stat should be globally saved after each map result.

• If the attacking team wins more often than the defending team on map X, they should get less points depending on the map-specific W/L ratio.
• If the attacking team wins less often than the defending team on map X, they should get more points depending on the map-specific W/L ratio.
  This point doesn’t apply to stopwatch, of course.

Then balance teams based on that Player Rating.

Some people here may know a similar system from ET’s mod - etpub. It’s called PRW and works very well for a team-based FPS, in my opinion.

[INF3RN0]

Ratings are primarily going to be used for matchmaking correct? So then ignore the past. If ELO will be used to balance/shuffle clan pub servers then maybe some of the concern is relevant, but overall I doubt it. Once again, there should be two main multipliers… Damage Per Minute and Score Per Minute. DPM is much more weighted than KDR or Accuracy. SPM will increase if the kills are at the objective, as well as if you are supporting or playing the obj. Overall the highest rated players will have the highest W/L+DPM+SPM. These are the only truly relevant variables and are also hard to fake. They should also inversely effect your rating too, so if you do win but have a low DPM and SPM- then it becomes an obvious sign that the win was not due to your contribution and thus should not be rewarded equally.

[tokamak]

[QUOTE=Destroy666;511918]XP/MIN:

• yet again, selfish players can freely switch to the dominating team and boost their XP/MIN
• you may throw ammo packs near spawn the whole map and gain quite high XP
• XP/MIN gets increased when you win against no opponent
• correct me if I’m wrong, but a player can join 30 seconds before a map ends and still gets 1000 XP/MIN, which is stupid[/QUOTE]

These obstacles are the easiest to control for, compared to the other alternatives.

[Smooth]

For ‘Public Skill Ratings’ we won’t be using Kills or Accuracy as an indicator of player skill because those are already accounted for in the XP system and many of our Mercs aren’t focussed in those areas.

The XP system does need improving and tweaking to get it balanced between all Mercs (and exploit/farm prevention where necessary) there will be more work along these lines in the future.

For ‘Matchmade Skill Rating’ the system will be based of Wins and Losses taking the skill of your team-mates and opponents, similar to the Elo rating system but accounting for the differences our game has, such as 5v5 teams rather than 1v1.

Quoting Anti since he said most of this already.

[QUOTE=Anti;511873]We already shuffle teams between matches based on what we call a ‘Public Skill Rating’, an aggregate rating of XPM across multiple matches. We use this system very much for the reasons tokamak mentioned.

It’s by no means ideal though, we have a number of things we still need to do. One issue is that we have to shuffle in the lobby at the lobby’s start so that players can pick their Mercs based on their side. This causes an issue where people decide to leave/join the session after the shuffle, messing the balance up - ideally we’d shuffle on match start but this would impact Merc choice a lot.

As others pointed out we also don’t normalize XPM by team, which might be a good idea. Right now we kind of let the aggregate rating across last 10 matches address the variance.

I think the biggest issue is that we still don’t have a shuffle/vote shuffle mid match and with the current small beta population this means there isnt a ready supply of players to come in and fill the gaps.

All of this is known however and on our list of things to improve, but any other suggestions you folks have would be good.[/QUOTE]

[spookify]

[QUOTE=Smooth;511943]For ‘Public Skill Ratings’ we won’t be using Kills or Accuracy as an indicator of player skill because those are already accounted for in the XP system and many of our Mercs aren’t focussed in those areas.
[/QUOTE]

None of the mercs are…

ZING!!!

[tokamak]

[QUOTE=Smooth;511943]For ‘Public Skill Ratings’ we won’t be using Kills or Accuracy as an indicator of player skill because those are already accounted for in the XP system and many of our Mercs aren’t focussed in those areas.

The XP system does need improving and tweaking to get it balanced between all Mercs (and exploit/farm prevention where necessary) there will be more work along these lines in the future.

For ‘Matchmade Skill Rating’ the system will be based of Wins and Losses taking the skill of your team-mates and opponents, similar to the Elo rating system but accounting for the differences out game has, such as 5v5 teams rather than 1v1.

Quoting Anti since he said most of this already.[/QUOTE]

Yeah this makes the most amount of sense.

Getting ELO right for team-wide games is incredibly hard though. I imagine that’s going to take quite a bit of tweaking and headaches before that one works right.