Tribes 1 Legendz

Current points system based on kills means nothing. Implementing the following statistic would do the following:

- give a single statistic that actually gives accurate indication of a players overall value to a team.
- Encourages players to do what is actually best for the team as that is the only way to maximize this statistic.

I call it "net caps per hour". It goes like this: When a player logs on, you track the time he is logged on and add it to his previously stored total log in time. You also track how many total caps have ever been scored by that players team and by the opposing team while he has been logged on. Then you compute the stat as follows:

= [ (total caps ever made by the team player is on) - (total caps ever made by team player is not on) ] / [total hours ever logged on]

This gives you a number that can be either positive or negative.

- A positive score indicates his team tends to score more caps than the other when he is on that team.
- A negative score indicates his team tends to be outcapped by the opposing team when he is playing.
- A score of zero would be the average player.
- the more positive the score, the better the player. The more negative the player, the worse the player.

This stat would clearly show who is causing teams to win and who is screwing them up. This stat could reduce base killing, screwing around, and it would reward those who get stuck having to set all the D that actually causes the team to win.


Thanks,
TEXAS

When a player joins they normally get sent to the team in need. This will affect their accumulative score. How will matches be determine if one were to join the losing team within minutes of ending a match?

That will tend to balance out overall. All players are equally subject to getting unlucky. On average it's all the same. It is like in the NBA when coaches make bad calls, you don't worry because you know the other bad call worked in your teams favor 5 minutes ago. In the short term the stat will not be so meaningful for the reason you mention, but after a player logs like 10 hours, it ought to be a really good indication as that effect will wash out.

This is not real life.

Fox legend is my bitch in the arena

You smurf in the arena so you don't get humiliated when you get beat.

vistalize wrote:Fox legend is my bitch in the arena

Texas_Crash wrote:Current points system based on kills means nothing. Implementing the following statistic would do the following:

- give a single statistic that actually gives accurate indication of a players overall value to a team.
- Encourages players to do what is actually best for the team as that is the only way to maximize this statistic.

I call it "net caps per hour". It goes like this: When a player logs on, you track the time he is logged on and add it to his previously stored total log in time. You also track how many total caps have ever been scored by that players team and by the opposing team while he has been logged on. Then you compute the stat as follows:

= [ (total caps ever made by the team player is on) - (total caps ever made by team player is not on) ] / [total hours ever logged on]

This gives you a number that can be either positive or negative.

- A positive score indicates his team tends to score more caps than the other when he is on that team.
- A negative score indicates his team tends to be outcapped by the opposing team when he is playing.
- A score of zero would be the average player.
- the more positive the score, the better the player. The more negative the player, the worse the player.

This stat would clearly show who is causing teams to win and who is screwing them up. This stat could reduce base killing, screwing around, and it would reward those who get stuck having to set all the D that actually causes the team to win.


Thanks,
TEXAS

Well I think if this could be done, We could add the distance each player snipes there targets in fog or no fog and figure out if he or she could be running happy mod, So Tex be careful what you ask for!

In thinking about this more, I think this stat when implemented needs to:

- Count wins rather than caps. A game can end 8 to 0, and I think it would be unfair for someone to pad their stats with so many caps, when it was only 1 win.

- Only apply the result of a game to a person's stats if they were logged in *to the same team* for a minimum amount of time, say 20 minutes. No fair to join last minute and get a win you didn't deserve, or to change teams last second. Also not fair to drop and rejoin hoping to change teams. Doing this would also make any matches that were a blow-out and lasted less than 20 minutes not be counted for anyone, which I think is also good.

- Only apply a result to a person's stats if the teams were relatively balanced during the match. For example, you can total each teams cumulative "man hours" (the sum of everyone's login times, including people who dropped). Then check if they are say within 20% of each other. If not, reject the whole game and don't update anyone's stats.


So once again, what I am suggesting is a statistic that looks like:

percentage wins = [total "valid" wins player has had] / [total "valid" games a person has played]

By "valid", I mean the win meets the criteria I've listed above.

Also, this stat should report total number of "valid" matches ever played. So for example:

Texas: 0.55 34
guy: 0.20 5
yup: 0.73 129

Reporting the total number of matches lets you know that "guy" who has only 20% wins could have just have bad luck, as he has only ever logged 5 matches, and that "yup" is very outstanding, as he has played 129 matches yet kept such a high win percentage at 73%.

So this stat is actually a little more intuitive than the previous "net caps" stat I described. It is just a "percent wins" stat. So if a guy has 0.50, it means he wins about half and loses about half. If he has 0.4, it means he wins about 40% and loses about 60%. If he has 0.7, he wins about 70% and loses about 30%. Etc.

This stat would:
- Deter base killing, as people would be hurting their own stat
- Deter people from just giving up if they start losing, as it hurts their stat to lose due to laziness.
- deters people from logging out and and logging in, as they will never log a continuous 20 minutes, and their stat show that they never seem to complete a "valid" match. So if a guy has been on the server for a long time, but never shows many completed matches, you know he either:
a) is a base killer and keeps dropping
b) is a chronic team changer
c) drops and rejoins to get out of jail
d) drops anytime his team starts losing
So this could help show who really are the spazzes that ruin the game for others.



Texas

excellent idea's and wishful thinking!

There should also be a script that doesn't allow one team to be stacked with the best players vs a team full of noobs. I'm not sure how or if the system makes any decision about who goes on what team before a match starts but I often end up on a team full of bums.