PvP Research Project
Go Orioles!!!
Finding good data for a meaningful ranking system is virtually impossible using kills alone.
Sure, the Blaster A (8 kills) LOOKS like he must be a better player than Blaster B (5 kills) who was on the same team in the last match
but Blaster A died 4 times and Blaster B didn't die at all. Without knowing the deaths, Blaster A will get the MVP every time.
have the team's vote for an MVP from the opposing team each match. At the end of the season, the player with the most MVP votes is the winner. In case of a tie, Co-MVP's!
Anyway, I wish you luck.
2 words: emp kill
The best way to get real data is to play. This will likely produce data as skewed as the dev's data mining. Good luck though.
Infinity and Victory mostly
dUmb, etc.
lolz PvP anymore, Market PvP for fun and profit
[ QUOTE ]
2 words: emp kill
[/ QUOTE ]
[ QUOTE ]
The best way to get real data is to play. This will likely produce data as skewed as the dev's data mining. Good luck though.
[/ QUOTE ]
This I know. Which is why I am playing. However, the sheer amount of data necessary is more than any one player can produce based on time and powerset combinations alone. That's why I'm here asking for help from other players.
Adversity
@Chilling Temper - L50 B
Candela - L50 PB
Caustic - L50 C
Crystallizing - L50 T
Crippling - L43 D
Animosity
Caex - L50 B
Psilencer - L37 F
Cooldown - L35 C
Q.Q - L19 S
[ QUOTE ]
[ QUOTE ]
2 words: emp kill
[/ QUOTE ]
[/ QUOTE ]
Got one of those so far, woot.
Yeah, the problem with the League (and well, with any League) is that some teams are coordinated and rolling over other people while others aren't...so you get very lopsided kills.
Is the Blaster/Scrapper/Brute that finishes the League with 80 kills the MvP or is the other player that keeps others bogged down/buffed while evading an MvP?
In the end, I think the finalists in those Leagues will be the most experienced teams...and the ensuing damage spam will indicate a winner.
Questions about the game, either side? /t @Neuronia or @Neuronium, with your queries!
168760: A Death in the Gish. 3 missions, 1-14. Easy to solo.
Infinity Villains
Champion, Pinnacle, Virtue Heroes
As far as coh PvP goes (especially pre-i13), there really is no worth in naming a "best player".
As Terp said, individual kill counts really don't mean anything.
A. They don't take into account the player's deaths durring the match.
B. Getting a kill in a team match only means that you were the last person to attack the team's target - what matters is that the offensive players are ON target. Who gets/doesn't get the kill is unimportant as long as the TEAM wins.
C. Obviously, an offensive player, such as a blaster, will get more kills (in most cases) than a support player, such as an emp/thermal/sonic/kin/ect. That doesn't mean that the blasters are better than the emps.
Rather than select a single MVP, you may want to consider the team as a whole, and chose a "Best Team".
Still, it would be inaccurate to call a player or team "the best" based on statistics and data that don't take into account circumstance.
For example: Suppose Team A has a match with Team B. Let's say that both team A and B are very strong teams made up of competant and experienced players. Team A wins the match with a score of 7-4.
Team C and Team D also have a match. Team C is a fairly competant team, while Team D is inexperienced and uncoordinated. Team C wins the match 23-1.
The problem of course is that according to the scores, Team C is the best team of the four, when in reality Team A and B may have had a shot at beating team C.
Generally, League/Ladder formats work best for deciding which Teams are collectively better because they take into account actual gameplay.
Dear PvPers,
I used to be a steady team arena PvPer about two years ago with my SG on Pinnacle. I took a break and they headed off to WoW. Recently, Pinncale has started a PvP League and it's re-ignited my interested in PvP. Had my first match under the new rules (minus travel suppression) and while my kill totals certainly weren't what they used to be, it was still a relatively enjoyable experience.
Anyways on to the point of this post, the Pinnacle PvP League had discussed a League MVP Award as a possibility. The only problem is that in the current PvP system it's difficult to compare one player to another (especially if they use different characters for different matches). We only get two stats really from the Arena who wins and kills by player (though I recently sent a PM asking for Deaths to be included). I'm an avid baseball fan (Go Orioles!) and feel that some of the metrics they use can be adjusted and applied to our PvP system.
I'm hoping to be able to derive a single decimal value that can be used to easily compare a player's value. The problem with this is there are multiple values to adjust for including but not limited to archetype, powersets, number of players per team, etc. To derive some basic adjustments I'd need tons of data and that's why I'm posting this here.
I'm asking for your help to start compiling some data.
The are a few limitations on data. Please send information on matches that were played Post I13 changes, are held at the Super Heavy Weight level and are a 10 minute timed duel or team duel.
Now on what information I need you to send:
1. A screenshot of the final arena score window with all players and kills listed.
2. List all characters with archetype, primary, secondary and ancillary powersets.
3. List any options that were checked off (eg Travel Suppression, Temp Powers, etc)
4. Date the match was held.
5. Map that the match was on.
Now, on to what you guys get out of it.
First, I'll open up the raw data to anyone and everyone including screenshots. That way you can easily get information about your past performance or how certain ATs powersets have fared in the past.
Secondly, it will hopefully lead to an objective rating system that's better than A > B > C. These ratings can be recalculated on customized datasets such as only including certain matches (for a PvP League Season), only including certain ATs (blaster vs. blaster matches only), etc. I'm hoping this project will help spur interest in PvP.
So to summarize, I'm trying to come up with a better rating system that adjusts for ATs, powersets and other variables. All you have to do is send me an e-mail with a screenshot of the final results of the match and some other info (ATs and powersets, Arena Options, Date and Map) to pvp@psikik.net. Once I have enough data to start with I'll post a link to the website with all the data.
Thank you in advance,
Chilling Temper
Adversity
@Chilling Temper - L50 B
Candela - L50 PB
Caustic - L50 C
Crystallizing - L50 T
Crippling - L43 D
Animosity
Caex - L50 B
Psilencer - L37 F
Cooldown - L35 C
Q.Q - L19 S