FreeBeer wrote:QUOTE (FreeBeer @ Dec 26 2010, 12:27 AM) That doesn't help, if I've read SgtBaker correctly.
I haven't run any formal tests on it (or seen the SQL), but I'm pretty sure your AS is only changed based on the amount of time you've been in the game and seems to be a linear relationship. Do you have the link to his post handy?
Assuming that's the case, it's unfortunate that it leads to such behavior (as well as stacking in a way that's not too visible on AS). Personally I don't care about rank as such, except that I absolutely detest the (4) that pops up next to the name which I'm trying to avoid.
Last edited by Spunkmeyer on Sun Dec 26, 2010 3:43 pm, edited 1 time in total.
Want bigger games? Log on to play at the official game time: 9pmET/8pmCT/7pmMT/6pmPT every day of the week. Also Saturdays 8pm UTC.
No, I don't have a link. And I may have interpreted the discussion (partly on Teamspeak, partly in the wiki and in forum responses) incorrectly. It's my understanding that rank is continually compared against everyone else. I do recall that they wanted to make sure that quitters were dealt with appropriately - they are definitely tagged with a loss if the team eventually loses.
It seems to me that if the game is going south on you, and folks are quitting, you're actually better off (albeit slightly) if you stick with the game. As your team AS drops, you're now being stacked against for that final part of the game and AS takes that proportion of stack into account.
Unless I've totally got their intention and implementation wrong. I don't claim to be an expert here. I have not seen the code, nor do I understand the underlying math behind it. I do know that much consideration was given to late joiners and early quitters and I'm confident that they handle this gracefully and with the intent of accurately capturing each player's relative contribution. The goal, after all, is to provide a means to produce balanced games by providing the players and comms with a metric in which to gauge that.
For that to work, AS calculations would have to have data on who joined & left when. I don't think that's the case. The only data available is time played. So it doesn't matter if you joined or left at the 30 minute mark of an hour long game - all AS has to work with is that you've been present half the time during the game and adjust your AS accordingly.
Want bigger games? Log on to play at the official game time: 9pmET/8pmCT/7pmMT/6pmPT every day of the week. Also Saturdays 8pm UTC.
You could make games more unpredictable by surprising me by NOT SUCKING.
The reason games are predictable is because everyone sucks too badly to pull off anything unpredictable.
You can look at two teams and say, "this team sucks more than the other team, so we can safely predict that they will lose"
and "these teams suck too badly to play X tech or Y tech, so we can safely predict that they will go Z tech"
IT'S YOUR FAULT GAMES ARE PREDICTABLE
BE UNPREDICTABLE BY $#@!ING UNINSTALLING
Last edited by Mastametz on Mon Dec 27, 2010 12:37 am, edited 1 time in total.
Spunkmeyer wrote:QUOTE (Spunkmeyer @ Dec 26 2010, 06:08 PM) For that to work, AS calculations would have to have data on who joined & left when.
That is correct.
QUOTE I don't think that's the case. The only data available is time played.[/quote]
That is incorrect. TAG logs all major and most minor events. Joiners, quitters, etc. are included in that data. (I've read through the logs, so I know it's there.)
AS is good at predicting the outcome of games when the score that AS assigns the teams is vastly different. It is, as always, up to the community to work on even games.
Get over yourselves, don't try to win arguments on the internet where the option of a punch in the mouth is unavailable
"It is not that I cannot create anything good, but that I will not." And to prove this, he created the peacock.
Spunkmeyer wrote:QUOTE (Spunkmeyer @ Dec 21 2010, 01:44 PM) Everyone who commands should understand the relative strengths of factions, the effects of money settings on factions and the balance of key players on each team, and make a strong effort to have these right for an even game. Everyone else should point out problems with these and not just hop on board for an hour of predictable futility.
The good people are so good they massively unbalance the game. 9/10 I can join a game that is even or loss for my team and turn it into an overwhelming victory.
I'm sorry I don't remember any of it. For you the day spideycw graced your squad with utter destruction was the most important day of your life. But for me, it was Sunday
Idanmel wrote:QUOTE (Idanmel @ Mar 19 2012, 05:54 AM) I am ashamed for all the drama I caused, I have much to learn on how to behave when things don't go my way.
FreeBeer wrote:QUOTE (FreeBeer @ Dec 26 2010, 08:46 PM) That is incorrect. TAG logs all major and most minor events. Joiners, quitters, etc. are included in that data. (I've read through the logs, so I know it's there.)
Yeah, I don't know if/how it's all stored though. Anyway I pm'd TE to clear it up and AS only takes into account the fraction of time you've been in the game, regardless of when you join or drop.
spideycw wrote:QUOTE (spideycw @ Dec 27 2010, 02:13 PM) The good people are so good they massively unbalance the game. 9/10 I can join a game that is even or loss for my team and turn it into an overwhelming victory.
And how do good people figure into this?
Want bigger games? Log on to play at the official game time: 9pmET/8pmCT/7pmMT/6pmPT every day of the week. Also Saturdays 8pm UTC.
girlyboy wrote:QUOTE (girlyboy @ Dec 21 2010, 06:56 PM) Why is it a bad thing that you can predict games by seeing who's playing? I mean... isn't that a good thing, really, telling you that player skill is what determines the outcome of games of Allegiance?
That. QFT.
If Baker's data is correct (which I don't question), my interpretation is that it proves both that Allegiance is highly player-skill based (yay!) and that AS correctly reflects player skill (yay!).
Last edited by Cortex on Fri Dec 31, 2010 12:46 am, edited 1 time in total.