Statistics

[owned]

IMO, just providing the raw data would be best.

Why not both?

I just don't think that we could really make a formula that would be accurate, and especially not one that would also stay accurate as tactics changed. Creating one and putting it on the wiki as "fact" in my opinion would be disingenuous.

[Hoax]

stats ftw

remember when alex got that golden question mark

[Tank Program]

If there are recordings available for every match, you could use those to generate statistics.

[Concord]

or the server admins, who surely must keep logs, could just pass those along.

[Rudycantfail]

would be fun also to chose an mvp, (most valuable player) for each ladle , or atleast for the final matches.

[Desolate]

lol Rudy

Tournament MVP voting?

[apparition]

I've always wanted to see screenshots and data of all the Ladle matches, so that'd be awesome to post up on the wiki.

Don't ya think all kill points should be included, even those after the match is done? I mean if someone is going to be proud of kill points they're not going to discount those after the match ends.

I think the easiest and quickest way for the stats to be compiled and generated would be using the server-integrated script related to INW/Concord/Free Kill. Then post the statistics to the wiki in addition to screenshots maybe? That'd be rad.

We had MVPs in AFL using votes on the wiki. I'd like to do that, but the only way I think that'd be possible is to use recordings/stats or just the final matches since no one is going to be able to see everyone play.

[Concord]

Individual MVP awards are even more meaningless in Fortress than in other team sports. And in the absence of everyone watching every player in every game or decent statistics, the awards are a popularity contest at best and something really bad, void of any meaning or significance at worst.

[apparition]

Don't really care if we nix the MVP thing.


Could we get the statistics initiative initiated by next ladle?

[Concord]

I'll demonstrate the statistics I've been working on once I get them worked out over the upcoming weekend. Someone with scripting abilities can run with that, assuming server admins supply logs.

[Mecca]

I've always wanted to see screenshots and data of all the Ladle matches, so that'd be awesome to post up on the wiki.

I have screenshots of the majority of the matches we played (I think I missed the final match vs tx and the final match vs pru) that I can upload later.

[dlh]

I can provide some stats from http://gridstats.com. Server admins would only need to provide their ladderlog.txt file.

The ladderlog.txt file would need some editing to ensure it contains only Ladle matches. It could be marked with an authority blurb message to show the approximate start time of the Ladle.

The biggest hurdle is getting server admins to actually provide this data.

[owned]

I was going through old topics and I came across this one. It seems like the only reason this idea was thrown out was because no one was willing to make a script. Is there anyone right now willing to make such a script?

Also, we don't need to stop at statistics for specific ladles. We could find stats encompassing all ladles such as:

Total Ladles Active in (for Clans, and another one for players)
Total Ladles Won (For clans, and another one for players)
Most Points in a Ladle Match
Most Points per match in a Ladle round
etc.

What do you all think of this idea?

[dariv]

I like the idea, but have no idea how to code in this way.

For a MVP, is the idea to create a score for each player to indicate usefulness per unit of time? eg. "kills - deaths + zones ganked" per minute of gameplay? Then discount people who played less than, eg, 30 minutes, to remove anomalies... I'd love to know stuff like this! Though it is hard to judge true value, especially as a good holer can change the game but gets no points from it.

[Olive]

the results would be unrealistic.

for instance, its easier to core dump (no pun intended) vs. una than ct. Also, defenders usually get more kills than sweepers, though its the sweepers who lure them into the defender's wall.

or we take the results purely statistical. Furthermore, for voting, to be least biased one shouldve spectated all players on all matches, impossible.