EPT[D] Points system - Page 10
| Forum Index > SC2 General |
|
pieisamazing
United States1234 Posts
| ||
|
InRaged
1047 Posts
On September 21 2010 02:04 ToxNub wrote: No. Please observe the following. 5 games per week (0% loss) = 50 points + 50 point bonus pool = 100 points 10 games per week (40% loss) = 10 points + 50 point bonus pool = 50 points 100 games per week (40% loss) = 100 points + 50 point bonus pool = 150 points 10:100 is a far bigger difference than 50:150. Thus 1. The system helps players who cannot play often (above a minimum threshold). 2. The system REALLY helps players who are ranked far too low for their skill, but cannot play too often. (In this example, doubling his gains). But sure, everyone is going to ignore this and just continue to post shit from their ass. Except in order to redeem your whole bonus pool you need to earn as many points as it's accumulated, cause bonus pool merely doubles your winnings. So your second example would have only 20 points  | ||
|
Champ24
177 Posts
On September 21 2010 03:47 SCdinner wrote: I think this problem is better than ones that would exist without bonus points. For example: someone with the beta who learned the fundementals playing a bunch of games against new players and getting a high rating after a couple days. Then he never plays again and in a couple years his name is up there at the top with a bunch of people that are WAY better than him. I think a better comprimised system would be for your bonus points to come from your actual points so after a day you gain 12 bonus points but loose 12. That is why you have deflation similar to WC3. It forced you to play or you get screwed for inactivity. It worked. Really well. And you didn't have an issue with inflation since the top players eventually hit a plateau. | ||
|
JustPlay
United States211 Posts
It's intentionally designed to give people a false sense of progress no matter how stagnant their play is. This would be fine if points and the league you are in weren't the only measure of skill, but it literally is the only measure of skill because they don't show your "true" rating anywhere. Bonus points are designed to get people on and playing games. The only real improvement they could make is to show actual ratings in some way. I'm not sure how they would do this in a way that most people could ignore, but it's not my job to figure that out. | ||
|
InRaged
1047 Posts
| ||
|
Champ24
177 Posts
On September 21 2010 04:47 InRaged wrote: Champ24, point decay from WC3 was based on punishment. This system is based on reward. This psychological trick makes it so much better than WC3 system even though it has some flaws like inflation. Just look at this thread alone and see how many people think that this system helps those who don't play often (even though it does not). WC3 system on other had only negative perception I would agree with your assessment if the top players in the world eventually hit a plateau in points so an accurate determination of every players skill in respect to each other can be concluded. Unfortunately since blizzard is giving us the silence treatment on specifically how this system works we can only speculate. | ||
|
AcOrP
Bulgaria148 Posts
so difference is 240 points ahead for the weaker skill player and more points With bonus pool 100 games = 48 points + for example if the bonus pool can give you 5 double point games this is extra 60 points total 1080 points 10 game = 24 points +60 84= 840 in 10 days now compare 1080 to 840 then compare 480 to 240 So do you still think that 52% winrate guy that can play 10 times more games should have double the points of someone with 60% winrate that can't play much. If we add few days without playing things get even worse without bonus pool. | ||
|
JustPlay
United States211 Posts
| ||
|
Kryptonite333
42 Posts
www.sc2ranks.com check your spot there if your interested how youd compare to other players. I think it increases your motivation when your points keep rising always  | ||
|
Logo
United States7542 Posts
On September 21 2010 04:44 JustPlay wrote: The inflation is terrible and offers no static assessment of skill. Outside of that the system is fine. It's intentionally designed to give people a false sense of progress no matter how stagnant their play is. This would be fine if points and the league you are in weren't the only measure of skill, but it literally is the only measure of skill because they don't show your "true" rating anywhere. Bonus points are designed to get people on and playing games. The only real improvement they could make is to show actual ratings in some way. I'm not sure how they would do this in a way that most people could ignore, but it's not my job to figure that out. I agree with this. It's so very annoying. While ladder is pretty meaningless having a small sense of progression does help guide you. With inflation you don't really get any idea of how your rating is changing relative to others so it becomes pretty meaningless. This is doubly true when/if you take a break from laddering and it can take a significant amount of time for your score to properly adjust to inflation. | ||
|
scDeluX
Canada1341 Posts
MAKE A GLOBAL RANKING FOR DIAMONDS PLAYERS. keep everything as-is and make another "section" where you can see the global ranking. You will know if you've improved or not by playing. This would also implies that points stop inflating that much and around 2000 would be the cap, otherwise its ridiculous. | ||
|
TheRabidDeer
United States3806 Posts
On September 21 2010 02:04 ToxNub wrote: No. Please observe the following. 5 games per week (0% loss) = 50 points + 50 point bonus pool = 100 points 10 games per week (40% loss) = 10 points + 50 point bonus pool = 50 points 100 games per week (40% loss) = 100 points + 50 point bonus pool = 150 points 10:100 is a far bigger difference than 50:150. Thus 1. The system helps players who cannot play often (above a minimum threshold). 2. The system REALLY helps players who are ranked far too low for their skill, but cannot play too often. (In this example, doubling his gains). But sure, everyone is going to ignore this and just continue to post shit from their ass. (assume 10 points per win, 10 points per loss) With bonus pool: 10 games per week (40% loss) = 20 points + 50 point bonus pool = 70 points 100 games per week (40% loss) = 200 points + 50 point bonus pool = 250 points Point difference: 180 points Without bonus pool: 10 games per week (40% loss) = 20 points 100 games per week (40% loss) = 200 points Point difference: 180 points The point spread is the same in any and all cases. This should be obvious to you if you know anything at all about math since you are adding a constant number to everything. Realy ToxNub is right I don't understand why you care about the infloation at all everyone get the same bonus pool so it doesn't realy change the ranking. And just check the diamond league I am mid-low diamond with winrate 52-53% If someone 52% winrate play 100 games per day thats 48 points in 10 days this is 480 points. Someone else with 60% winrate play 10 games per day so 24 points in 10 days thats 240 points(this is the case without bonus pool) so difference is 240 points ahead for the weaker skill player and more points With bonus pool 100 games = 48 points + for example if the bonus pool can give you 5 double point games this is extra 60 points total 1080 points 10 game = 24 points +60 84= 840 in 10 days now compare 1080 to 840 then compare 480 to 240 So do you still think that 52% winrate guy that can play 10 times more games should have double the points of someone with 60% winrate that can't play much. If we add few days without playing things get even worse without bonus pool. No, he wont have double the points of somebody with a 60% winrate that cant play much. You also cant look at just the win rate. The way things work is that you rise and rise in points until you are facing people that are near the same skill as you. Then you start to lose more, and eventually you reach a 50/50 win loss for that rating. However, it wont look like a 50% win/loss on your profile because while getting up to that point you might have had an 80% win/loss. So, the more you play just reduces your win/loss %. So these top players that you see with a 52% win rate with huge numbers of games played are only that low of a win % because of the number of games they have played. | ||
|
Shakes
Australia557 Posts
On September 21 2010 04:30 InRaged wrote: Except in order to redeem your whole bonus pool you need to earn as many points as it's accumulated, cause bonus pool merely doubles your winnings. So your second example would have only 20 points No it doesn't work like that, in earning those 10 points he probably gained gained 55 and lost 45 (non bonus) points. So you still get 55 + 50 - 45 = 60 points. | ||
|
xtfftc
United Kingdom2343 Posts
| ||
|
AcOrP
Bulgaria148 Posts
On September 21 2010 08:28 TheRabidDeer wrote: (assume 10 points per win, 10 points per loss) With bonus pool: 10 games per week (40% loss) = 20 points + 50 point bonus pool = 70 points 100 games per week (40% loss) = 200 points + 50 point bonus pool = 250 points Point difference: 180 points Without bonus pool: 10 games per week (40% loss) = 20 points 100 games per week (40% loss) = 200 points Point difference: 180 points The point spread is the same in any and all cases. This should be obvious to you if you know anything at all about math since you are adding a constant number to everything. No, he wont have double the points of somebody with a 60% winrate that cant play much. You also cant look at just the win rate. The way things work is that you rise and rise in points until you are facing people that are near the same skill as you. Then you start to lose more, and eventually you reach a 50/50 win loss for that rating. However, it wont look like a 50% win/loss on your profile because while getting up to that point you might have had an 80% win/loss. So, the more you play just reduces your win/loss %. So these top players that you see with a 52% win rate with huge numbers of games played are only that low of a win % because of the number of games they have played. 70:250=1:3,5 20:200=1:10 point difference is 180 points in both cases Thats the simple idea of it. And there are players that maintain 60% winrate... diamond has alot skilled players that play more custom games than ladder. System will match them with equal but system may also match them with some weaker players this cause higher winratio. the 50/50 thing is joke... | ||
|
ThunderGod
New Zealand897 Posts
| ||
|
Drazzzt
Germany999 Posts
IF there is real point inflation due to the bonus pool AT ALL, is still not 100% clear as it is mainly a question whether for the favor-calculations your points are corrected for the bonus pool (paralleluniverse believes this) or not (I believe this). If they are corrected then there should be a bonus pool inflation, if not then there isn't and the inflation seen is only due to a larger player pool PLUS a larger skill "bandwith" in the player skill distribution, as the system is rather a relative than an absolute system (like ELO), which means point differences are important. Therefore, to judge your skill it is probably better to look at the point difference from the top players rather than the absolute value. You can read the posts of parallel universe and me on the previous two pages to learn more about what I am talking about. Again, as it still seems unclear: As far as I know there is NO PROOF at all that the seen point inflation is due to the bonus pool and is therefore infinite (grewing the points with 0.5*t, with t being the hours since release). | ||
|
Champ24
177 Posts
| ||
|
Snowfield
1289 Posts
Also; EVERYONE GETS 1 BONUS POINT PR HOUR, ITS THE SAME, EVERYONE HAS THE SAME AMOUNT, doesn't matter if you play or not if it turns out that you can gain points indefinetly, this works if ladders gets reset every now and then | ||
|
Grummler
Germany743 Posts
On September 21 2010 15:52 ThunderGod wrote: I thought everyone knew: it's 1 bonus point per hour. On September 21 2010 21:36 Snowfield wrote: Also; EVERYONE GETS 1 BONUS POINT PR HOUR, ITS THE SAME, Wrong, its 1 point per 2 hours. | ||
| ||
