Anyways, I realized something. Say you are 1650 PSR. But you always, always play alone. Say your opponents are 1650 too, but they always, always play together, the five of them. Since probably they have the advantage of the teamwork, which the PSR doesn't represent, it would mean you are better than them. It would mean that if you were to play with a team you would outplay them..
This. The PSR would work if in every game the teamwork would be the same, only then the true skill of players could be measured. Ofcourse picks would be of influence as well.
I believe the picks would be a fundamental diffence, since they would be the most important variable as PSR and teamwork would be roughly the same. However, the "skill of picking heroes" would, in long-term, also be reflected in PSR.
The PSR system has its obvious flaws, but I like it. It reminds me an aspect of poker. Short-term, luck matters more than skill. Long-term, skill prevails over luck. Luck being environmental aspects like if you are tired, unfocused, bad computer, lag, etc. If the teamwork flaw would be fixed, then it would be a nearly ideal system.
I don't care which Christian branch you belong to, unless you belong to one that acknowledges that the entire friggin Bible is a fairy tale that was written by a bunch of different idiots and then voted by a congress on which gospels and crap were actually worthy of being in the Bible, you're an idiot. You're a person who doesn't think for himself and clings to the stupid beliefs his parents forced on him as a child, or you're the even worse kind of idiot who actually started believing this crap as an adult. Either way you're an idiot.
even ur teammate left u there and u are dead...is nth v ur PSR dude...if they make u die and u lose the match...they lose as well...if ur temp make a good ult and win the match without any kill...u can still raise ur PSR... what will different was KDR not PSR... -__-"
No man, never killsteal! And when I say killsteal, I mean obviously wasting ultimates or something.
Some people may hate me for this. But the stupidity of some players really just shocks me. I'm really tired of "stupid" people who just flame and talk arrogantly. KILL DEATH RATIO DOES NOT SHOW HOW GOOD YOU ARE.
Sure, they can tell you SOMETHING, but people interpret it the wrong way. First of all, PSR is the most accurate way to tell how well someone does in public games. So, even if they do feed and die 50 times per game, but still have a PSR 1800, would you kick that player out of the match? Well I wouldn't because there's obviously something about that player that makes him win a lot. (Ignore the possibility that he may be exploiting the system lol).
If you looked at the PSR and want to look at more stuff, THEN you go look at stats. When you look at stats, however, you do NOT look at kill death ratio ONLY. You look at EVERYTHING. S2 Games is awesome enough to show us like 10 different aspects of statistics. They did this for a reason. For example, if someone has a PSR of 1500, but has a K/D/A ratio of 10:1:0 with a exp rate of 400 per minute, then you can immediately have a good guess of how the player plays like. In this case, it's likely a combination of conceding a lot, missing out in team fights, running away or standing off to the side of a team battle and killing everyone after others have died, or they're a kser. You determine this from the low deaths, no assists whatsoever, the high amount of kills, and the low exp rate.
Another example. If someone has a PSR of 1800, a KDA of 1:10:10, and a exp rate of 600, you can easily determine that the player is likely a good support player. He's always in for team fights (hence the high exp rate and high assists), he dies a lot, showing that he's either doing his job as a tank or initiator well, and also that he lets the carries on his team or such get the kills. His PSR shows that he plays well, and hence he wins a lot.
Don't be a statwhore. And don't encourage stat whoring. Unless you can read the stats properly, you should stick to looking at PSR. Statwhoring is the kind of act is what will create a bad community, which is exactly what destroyed DotA's online public community. It encourages people to kill steal each other, and ruins the fun of a team game. It also encourages people to concede when it's like only 15 minutes in the game, and their team will clearly win lategame because the enemy team has all early game int heroes.
Note: Btw, I understand what kill stealing is. It's understandable in a huge team fight that a Legionnaire uses his ultimate to quickly turn the 5v5 into a 5v4. However, a Thunderbringer who greedily watches a Devourer ganking a lone hero and uses his ultimate at the last second is "kill stealing". It's like betraying your team, as you wasted your ult that you could have used for something better to help the team, and you've also wasted exp since you weren't even at the location of where Devourer was fighting. Sure Devourer still gets some exp for being near, but the killer still gets more exp than others.
Last edited by Ragnarok17; 10-18-2009 at 01:40 PM.
Agreed, Ragnarok17. Just a minor correction, when Thunderbringer kills someone by ulting and is not near a 1000 radius range, he won't receive any xp. Just the gold and the stats.
And a point of opinion, I don't think supporters and tankers dying show that they are doing their job well. If you have 10 deaths and the game drags to an hour, you will be mostly underleveled and without items. That way you won't benefit your team as much as you could late game. It's not that they should not die at all. It's just that even if it's a tanker/support, it's not worth to suicide in order to kill 2 heroes, usually. It's a different history when you die, your whole team survives, and the other whole team dies.
Host your own game and check the stats of every player (there's a UI mod to do it quicker), looking at K, creep denies, win/loss and xp/min. Kick those who are way out of your skill level.
Start off doing this with the green game (noobs only) and easy mode, if you are winning vs these players with similar skill then move up to noobs allowed EM, then non-easy.
Not sure how matchmaking will work since PSR is based solely on win/loss? A noob stomper with 70%+ wins is obviously playing against people a lot worse than him but that doesn't make him a better player than someone who plays vs pr0 players only with 50% wins. The noob stomper will then have higher psr unjustly and be matched against better players while the 50% wins decent player may get matched with someone on a lower skill level who also has 50% wins.
Last edited by deweylewie; 10-19-2009 at 08:56 AM.
well my psr is 1614 atm and my skill level is around 1850sh i think
i just lose to many points to teamplayers that go in when they are all miss / never talk/ don't have a microphone/picking more agility heroes when we have 4..../ and never ever type something i'm beating 1800 which have the same skill or WAY below (somehow they went to 1800 by good teammates i guess)
psr sucks they should make games with the number of games a person played
like 100+ games only
200+ games only
would be better imo atleast u know they aren't that nobbish then
and ofcourse not on another account and trying to bash u
the point is, that aiming for better PSR is wrong.
when u stuck at 1600 it's just your level, dont complain about it. this just shows youre a mediocre player, not bad or pro.
i got stuck at the 1700ish as well, sometimes slightly over, sometimes slightly below.
all i can say is, that i'm okay with this rating, i get good games, where most of people share my skill level, therefore games are balanced and ussualy exciting.