Leaderboards are flawed?

[Kanachi]

I’ve kinda gone into this mid way in another thread, but I figured it may be worth talking about on its own.

I think the current implementation behind the Leaderboards is flawed? It rewards longevity not ability. A farer system would be to calculate a player’s average points once they have played 10 or more games. Before you have played 10 games your score fluctuates too much to draw an accurate average and thus you do not appear on the leader board.

As i posted in the previous forum. Who is more skilled the guy who has played a million games and won 100 or the guy who has played 11 games and won 9?​

​

The first guy has a win % of less than 1%... so in other words he sucks​

The second guy has a win % of 81%... should he really have to win 90+ games just to overtake the other guy.​

​

The longer the games run the more out of touch the leaderboards (in their current form) will be.​

​

Also if you have not played for X amount of time (say a month) your score is hidden until you start playing again. This will remove people who have long since left the game from the boards making way for others over time and keeping the leaderboards fresh.​

​

Also if people want to become legends of the game and leave thier mark somewhere then have world records or something: "most kiss in a game" or "quickest win" for example. Or you could run top 10 torniments each month to win "player of the month" status and keep you name in the record books as it were.​

 

[Borissssss]

Basing the leaderboards on win/loss ratio alone would not work, it would mean that a few lucky people that played a few times and won a few would hold the top place, not by skill but by luck as many of the game go by. It would make people not want to play once they had won a few games, as even the best players have a few unlucky bad games which would water down their leaderboard ranking.

Ok so a few people have quite a lot of time to spare and spend it on the survival games, why should they be punished, anyone who spends time on a skill gets better at it. Even if the top player has played a lot, it means they have all the more experience for it and are therefore more likely to win, than someone who has played a few games with 1 lucky win.

Although the awards would be a nice addition.

 

[Kanachi]

Basing the leaderboards on win/loss ratio alone would not work, it would mean that a few lucky people that played a few times and won a few would hold the top place, not by skill but by luck as many of the game go by. It would make people not want to play once they had won a few games, as even the best players have a few unlucky bad games which would water down their leaderboard ranking.

Ok so a few people have quite a lot of time to spare and spend it on the survival games, why should they be punished, anyone who spends time on a skill gets better at it. Even if the top player has played a lot, it means they have all the more experience for it and are therefore more likely to win, than someone who has played a few games with 1 lucky win.

Although the awards would be a nice addition.

Sorry i think you misunderstood.

You would obviously exclude people from the leadboards if they had played less than X games (i plucked 10 out of thin air, but you could up that). Obviously this is logical because, as you say, someone with 1 awesome lucky game would be top. So your issue would be addressed.

and i deffonately disagree on your second point... The current system only rewards longevity of play not your overall performance. If I play a million games i'm likely to win a few just by sheer luck... this does not make me a good player, simply someone with alot of time on their hands.

Only people with artificially inflated scores would loose out as a result. This would certainly be a better measure of skill.

 

[Borissssss]

To get a good average you would have to do a lot of games still 10 wouldn't be enough, everyone when they first start out are usually bad, maybe after 30 or so. Although score already stands as a measure of skill, you just need to look at that, a high score is very hard to maintain. I still thing that the leaderboards should be based on victories though. maybe if each victory gave a bonus to your overall score, it might work to measure the leaderboards by score.

 

[bart]

I think they should use the same system as halo does. When peole quit a few times before dying they should lose points.

 

[Kanachi]

To get a good average you would have to do a lot of games still 10 wouldn't be enough, everyone when they first start out are usually bad, maybe after 30 or so. Although score already stands as a measure of skill, you just need to look at that, a high score is very hard to maintain. I still thing that the leaderboards should be based on victories though. maybe if each victory gave a bonus to your overall score, it might work to measure the leaderboards by score.

Your looking at the problem as it stands now when an accumulated score can still be caught up by a new player... think of it in a few months (or a year) when those score become utterly unasailable to a new player.

Also high scores are hard to maintain because some people have more time on there hands than others. I may be the best player ever but if i can only play once every few days my score would struggle to keep up with someone who is not really that good but plays day and night all day winning the odd few every 50 games or so.

Check any sport, players are assessed on their consistancy... how many acurate passes do you make per game? whats your % of cross court backhand winners? whats your pitching or batting average?... etc...

You may well be right about 30 games being a good starting point though.

 

[Borissssss]

At a high level it is really difficult to keep skill at a high level, at the moment i have a score of about 700 (it might have changed by now because i really do love playing a lot ) Every player that kills me that is bellow the score of 300ish, which is most players, detracts 50 score from me even a cheap kill at the start and i'm down 50 points. Winning a game only gets you about 70 points, so if you want to remain on top you almost have to win every other game.

But yes i do agree with you they could measure the leaderboad by different multiple factors, but I'm not sure that's likely going to happen. In most video game leaderboards its is pretty much always the person who plays the most that is the highest. I still feel that score is a good measure of ability because I am sure having to try hard to keep it up, with every small mistake being heavily punished for. I mean I get 2 points a kill usually and -50 a death if score is no measure of skill i don't know what is.

The leaderboard could have a win/loss column but I do not think it should be the top priority. Yes noobs can get a high score if the beast a few games, but 'beasting a game' does take skill too. This is an interesting topic you have brought up.

 

[Ber Ric]

Maybe could be a combination of wins/losses, kills/deaths, time survived/total game time, total non-empty chests looted / total chests looted, and so on... (things that I'd like to be able to see somewhere outside the game as I speak of it... general stats are good for self-satisfaction/dissatisfaction ^^)

 

[Kanachi]

At a high level it is really difficult to keep skill at a high level, at the moment i have a score of about 700 (it might have changed by now because i really do love playing a lot ) Every player that kills me that is bellow the score of 300ish, which is most players, detracts 50 score from me even a cheap kill at the start and i'm down 50 points. Winning a game only gets you about 70 points, so if you want to remain on top you almost have to win every other game.

But yes i do agree with you they could measure the leaderboad by different multiple factors, but I'm not sure that's likely going to happen. In most video game leaderboards its is pretty much always the person who plays the most that is the highest. I still feel that score is a good measure of ability because I am sure having to try hard to keep it up, with every small mistake being heavily punished for. I mean I get 2 points a kill usually and -50 a death if score is no measure of skill i don't know what is.

The leaderboard could have a win/loss column but I do not think it should be the top priority. Yes noobs can get a high score if the beast a few games, but 'beasting a game' does take skill too. This is an interesting topic you have brought up.

To be honest if your a good player and your maintaining your status i.e defeating the odds then my system would work out well for you. It would simply be a more accurate representation of your skill over time.

Also just because something is done often is not an argument that it is done correctly. You have to ask "what are the motivations behind why things are the way they are?"

For example in a game where you pay to play over time it makes sense "as a business" to reward your players who have played the longest, i.e keep them near the top so they keep playing... and thus keep paying. An MMO is a great example of how the rich get richer and no noob can ever catch up to their level if they didn't start playing on release. Things are the way they are because its all about business. However, anyone who understands maths can surely see the flaws inherent such systems.

The only people who will loose out are those with artificially inflated scores due to the "length of time they play vs the number of points/win/whatever they collect" %.