Topics

1

Feedback / [gproxy]

« on: December 15, 2011, 13:56 »

Can when the game is loaded, the gproxy to play the sound "Game started." ?

2

Feedback / Losing connection before the end

« on: October 11, 2011, 17:44 »

It is very unpleasant to lose a connection right before a won game.
Is it possible to implement:
If a player lose a connection no later than 5 minutes before the end, he will gain points if his team wins.

3

Suggestion Board / Friends of the maphackers

« on: September 30, 2011, 17:31 »

If you are a friend and play  with a maphacker and you know that he is cheating, will you be punished?
(didn't find such clause in the rules)

4

Strategy / A guide how to get higher PSR - Play more!

« on: September 26, 2011, 15:10 »

Almost all of the guys with 1900+ psr are teamed up.
You are not so good or neither have 2-3 good people to play with.
What to do to go at least over 1800?
Patience and many games. Your skill is not so important.

The formula calculation the PSR has two things which comes in favor in this situation:

1. Winning psr is with a bigger weight over losing psr. The very first example is if 10 players with 1500 are playing. All of them will have +17/-16 (the base is 33). Winning gives more.

2. Reducing the base
At 1500psr, the base is 33. After 1700, it becomes 10.  It is further reduced if your psr is exceeding the other 4 teamplayer's average psr with 70. If it is over the difference of 187, you will have +0/-0.

Let's say you have something like 1700 psr. You join a "1550+ psr" game (150 below yours).  The game is full and the host do !balance. Since you have high for this game psr, you get weak player in your team - their average psr is way below yours. Thus your base goes very small.  Example:  5 playes with 1550 vs 4 players with 1500 and you with 1680. The game is fairly balanced and you have +1/-0. This is because of reducing the base the rounding it in favour of winning. +1/-0 or +2/-1, it's not hard to find such games. You will lose many such games, but eventually will win many too. When you reach 1750, you aim for "1600 psr" games, and one by one, you will go up to 1800.

So, regular players,
Pick your games carefully and play more, and you will go over 1800 with no doubt.

PS: I bet that's the reason why so many players with ~1800 psr are not good. A fairly decent player should exceed 1800 in less than 100 games.
 

5

Feedback / Stats page - draws=wins

« on: September 23, 2011, 09:48 »

When looking at the table, 3 parameters are flawed: played games, wins % and losses %. That is because a draw game is counted as played and won game.
May you fix it?

6

General Discussions / Whos's playing in PD.eu?

« on: September 19, 2011, 17:32 »

Couldn't find topic about players' distribution based on their location.
I am pretty new here (I like the gproxy), but couldn't not notice that the main european dota players are missing. No DK/SWE/DE/FI/NO.  The players are only from slavic countries and Turkey. Where are the good norderns?

7

General Discussions / A lifetime of a maphacker

« on: September 19, 2011, 16:49 »

Being bored at work did some research: Browsed through the maphack section and checked the last 100 permbanned players there by checking the max psr and played(without the draws) games.

Results
- played games:
A maphacker plays at average 50 games before he gets caught.
- max PSR:
Approximately 40% of the caught maphackers had reached max 1550 psr, 30% up to 1600, 20% up to 1650 and 10% up to 1700(only one had over 1700).

Conclusions

- the good news:
The maphackers are low skilled players who has no other chance to win except to cheat. That's like >99% of them. The other 1% are good or pro players looking to boost their skill and they are hard to catch (TeG-Kanga is an examle). Thus said, without a very good approximation can be said that above 1700 psr, the percentage of maphackers is 1%. So if you are a decent not-cheating player above that limit, you will have fairplay games.

- the bad news:
These 100 maphackers were checked in topics not older than Sept 17. This is like catching 50 maphackers per day. And this number is proportional to the total number of not catched maphackers, which, as it seems, is quite a number.

PS:
Admins,
Can a permbanned player create a new account? When a permban is made do you blacklist his ip?

8

Feedback / [forum] Redirecting after Posting

« on: September 16, 2011, 17:00 »

Wouldn't be better when a post is made the user to be redirected back to the topic instead of to the parent section?

9

Feedback / A new factor evaluating the performance (to be added in the player's statistics)

« on: September 14, 2011, 14:37 »

Two factors are quite accurate no matter the hero's role in the game.

(kills+assists)/deaths
This is already implemented. No matter carry or support, this is very accurate. If you have >4, then you are good.  But this only gives info about your individual skill. You might have in a game 5-0-5, but this would be a bad score if your team is winning like 50-5.

(your kills + your assists)/(team kills)
This evaluates your teamplay. It ignores how much you have died, but shows for how many of team kills you have helped.  Under 50% is considered as not a good teamplay.

So maybe to evaluate both teamplay and personal performance at once a possible factor would be:

P = (your kills + your assists)²/(your deaths)(team kills)

P<1 is considered bad.

What do you think?