Sheep Tag Scoring System?

beer what dont you get? There wont be any perfect system. Its not poxsible to make.

Lakes thats the problem though. The system wont be unbiased. It will make certain players seem a lot better than they are anc vice versa. It will encourage worse playing, less variation and slow down development of st theory.

I ask this again, what is the benefits of irrelevant statistics?

lakes means objective values.

while its true that a score system doesn't accurately reflect skill and promotes bad gameplay, it might be a good idea to implement it at this point. there is 0 hope for reviving ST, doing small stuff like this might delay the complete death of ST and give a temporary increase in activity.

as fun as debating is, its pointless to evaluate map-changes as if ST was a big game and had any future. there is no long-term anymore, do whatever increases activity here and now.

The system will still be biased beacause the process where we decide what "objective values" to take into consideration wont be objective.

If we are going to make mapchanges to sheep tag, why dont we make the ones that will actually contribute to make it a better game rather than arguing that "this mapchange will make st worse, but since its dying anyway why not try it?".

We have mass ideas which are actually founded in good reasoning. If someone can make changes in st, do them instead.

Why is there farmcount in the game then? it's not like I really think someone having more farms than me makes that one better, it's just nice to know and what I read out of it is my own business.

with objective values I mean stuff like death per game or avg distance to sheps, not the aggregated info chakras equation delivered

Amir is right about the short-time effect such a rating would have though

Chakra's suggestion is okay, but I think a more intuitive approach would be to quantify how a sheep does under pressure when around wolves. This is really how I judge how a sheep performs when I'm observing other players: how they cope with danger. When a sheep is off in a corner not near any wolf then that should not add to the score, and when a sheep is surrounded by wolves and manages to escape or find a way to survive that should add to the player's skill.

An algorithm for quantification might be to consider a 'danger potential field' with wolves representing high potential and clusters of farms lower potential with a gradient of increasing potential towards proximity to a wolf and a gradient of decrease around farms. Time spent in positive potential will add to a player's score while zero or negative potential would not affect a player's score. Wolves with more kills and golems/items would also have higher danger potential. Wolves around other would wolves increases the danger potential in the overlapping regions around the wolves.

If you don't know what a field is, just picture an imaginary terrain with mountains, hills, and valleys. The highest point on the mountain or hill would be like where the wolf is, and the valleys would be where the farms are. A person going into higher altitude would be like a sheep going into higher danger potential.

This solves problems with previous statistics such as:

• Farms Per Minute: Doesn't account for player's skill when RTing or using sparse farms.
  
• Save/Death Ratio: Biases players who hide and wait for an opportunity when the pen is danger free to save, and run into masses when threatened to avoid death. It might still be a useful factor, though.
  
• APM: Useless rapid move commands to the same point would bias the score. Not really a useful stat either as FPM is a more useful subset of APM. APM is really only useful for when you're measuring multitasking with many units and not just one sheep.

It isn't the sole descriptive metric though, it would represent how well a sheep survives around wolves (maybe dubbed the 'Survivability' statistic?), but overall game strategy, like map control, would probably need the factors of peak sustained farms per minute, average lifespan, covered map surface area, etc. as a separate number.