Barbs, noobs, screens and fantasies (all QQ in one topic)

**Chaman-bourru** · 13th October 2013, 16:42

Originally Posted by MagnusLL

Ehrr... that one actually shows barb as second last in K/D ratio...

Like Klaston said K/D isn't the only factor and to be honest they are all rather close to each other. Those charts are six/nine months old now so might need a bit of updating but i tried to make them as accurate as possible. They just try to show how each class really does in average during mini.

That being, if you want to prove a point, throughthedoor (or anyone), then get some proper stats. It's rather fast and simple with a spreadsheet application you fill with data after each mini while waiting for the next one.
Throwing a bunch of selected screenshots doesn't really lead anywhere and when you've reached a certain point it might be useful to continue discussing and go past the "i'm right, you're wrong" phase.

**Caudilloo** · 14th October 2013, 11:20

Just like you I am not a brilliant mathematician but I think you made a few errors.

Have included yourself in those stats?

If you have they are not very valid.

This is because half of the hoxes in your calculation are not the 'average hox' on Fury but only one hox namely you.

Assume you are the best hox on Fury but the worst sin (not saying you are but bear with me for the sake of the argument).
If you have a average 5:1 K/D ratio on your hox that means that the rest of the hoxes (aka the server 'average' hox) must have a K/D ratio of 1:5 in order to make the total K/D ratio about 1:1. The average hox then does not have a K/D ratio of 1:1 like you conclude but still of 1:5 with you just being a exception to the rule.

Same goes for sins. If your K/D is about 1:5 that means that the other sins (aka the servers 'average' sin) must have a much higher K/D ratio because 20% of the data you have is from your own sin.

It also means that if you were 'lucky' and you played 7 times with or against the same pom because you happen to play at the same times a day, that 10% of your data about poms is not a average but only data about one pom which happens to be a exceptionally good/bad player and again your conclusion is off.

You most likely have a certain time a day you play mostly so you will meet the same people in minis quite often. I also wonder if those stats would be different if you played during US prime time for example since it looks like they often play classes quite differently.

Long story short: Exclude yourself and other players you counted more then once from your calculations or count them only once.
And also test all times a day to get a more reliable sample of the AoC population.

**kalston** · 14th October 2013, 13:42

Vehl already acknowledged most of what you said. But there is no way of making better stats without spectator mode/replays, unless FC decides to do it themselves. Because when you play minis you will affect how much people kill and die (even if you sit in hide doing nothing you are impacting the game by making teams unbalanced). What he did is the best players can do.

While he is a good player and surely boosted hox stats a little, there are enough hoxes other than him doing great on fury, especially in pugs, so it's really still very interesting stats. Far from perfect, but only funcom can do better

**kennyfisterbotm** · 14th October 2013, 13:47