Board Thread:Wiki discussions and announcements/@comment-1990160-20140521040036/@comment-4091261-20140523015020

I don't think removing gold efficiency is the right response. I can see why you want to remove it because of the ever so growing variability of the statistics, but removing gold efficiency isn't the right response. Adding information is the correct response to this situation. Of course it will be hard, of course people will have high expectations of the wikia. Though to simply scrap the idea and think of a new idea in the near future isn't the right response. Doing so will create anarchy and further decrease the our reputation, and yes I said our because I personally do adore this wiki.

To clear up some confusion allow me to be so bold as reply to couple worthy mentions in this thread. Just to show I've been lurking and reading everyone's threads and so I don't sound like a pretentious butt- cause I will sound like a pretentious butt.

TehAnonymous wrote:;Discussion || || || || || ||  \/  \/     \/    \/    \/    \/ Where are you in dis conversation!! D:

Willbachbakal wrote: Wouldn't retaining the information and keeping it tucked into a dark corner of the wiki just be a less optimal way of keeping gold efficiency? Yes, but it is the easier way to do things. In democracy, the easiest thing that everyone can agree with even in some degree is the best way. Speaking of democracy.

Yatsura wrote: Wikias are not meant for interpretation or censorship of any kind.

Wikias are meant to state raw facts, proven by generally accepted sources. So how do Wikias deal with plagarism? Oh yeah, they censor it. Wikias are a democracy, freedom with rules so it doesn't screw itself over. If it feels that it's rules are being questioned, then things start to happen. It's only natural that the questioning of the balance of a well known statistic we have is causing this much racket.

Deshiba wrote: Unless this gets fixed, we will be spreading inaccurate information. Which is something that affects the reputation and reliability of this wiki. Many gold efficency stats are not inaccurate information. That is something you don't understand. What needs to be fixed are subjective stats such as: with it's active, multiplicative percentage statistics, and base stat steroids with a duration. is an important mention because it is only 3 autoattacks or spells considered autoattacks that apply it's bonus. Then there is any percentage statistic because it is subjective to any flat statistics and is multiplicative with the exception of lifesteal, spell vamp, and other stats I can't remember off the top of my head.

Typhron wrote: Reminds me of people using Damage parsers the wrong way. Numbers can easily be abused, especially if you consider the calculations vs application argument. You're awesome, you know that? But yeah, exactly what I said in my previous argument in a much simpler fashion. Some people just use the numbers given in a wrong way. Though honestly, there are a couple statistics that are a bit off in this wiki.

ClariS wrote: And the thing about comparisons, to truly and accurately make comparisons, all possible ‘unassigned’ variables must remain the same at all times. If not, we have an unknown value that we have no idea how to adjust for. This makes any answer with these unassigned variables nothing more than a guess (which we aren’t suppose to be doing on this wiki). Are guesses bad? Nope. You really take this seriously and I admire that, I always have, though a guess isn't bad. As long as it's explained clearly. That is what Probability & Statistics for, making guesses that can be explained based on very stable inferences. This tacks on with why duration is a problem. The one-dimensional side of gold efficiency also tacks in with duration which assumes infinite duration. Actives, unique passives, and auras have one thing in common: they have a variable duration. Though one thing must be understood when it comes to gold efficiency, not utilizing your statistics also counts as a variable duration. Every champion applies statistics in variable ways and it shouldn't be attempted to measure that. Rather what should be measured is what these stats have in common with every champion which I find this system to accurately display regardless of the flaws.

So, what does it all boil down to? Well it boils down to one thing really, statistics aren't perfect. Nope, they never were, and they never will be. People find the flaws in our statistics and bash on it. Ok, go ahead and bash. In a democracy setting, it is amazing enough that we created such a system. It's like part of the constitution of this wikia, an implied power, to give to the populous the informations based on golddddd! Cause remember, sometimes the simplest things are the hardest to find. So let's begin FULL BLOWN OVERDRIVE ANALYSIS!

Q.What is gold efficiency?

A.How much you get out of your gold in it's most basic terms.

Q.Why is it flawed?

A.Because of the many assumptions to compare the gold values across other statistics.

For example, trying to compare the monetary value of ability power in comparison to attack damage. How do you compare that exactly? They are used in a totally different way so how do you determine which is better? Because setting the standard at the base stat makes assumptions that one entire statistic is more useful than the other.

Another problem is that the duration of when you can use certain statistics such as: auras, which require you to be near the champion with the aura; unique passives, which require certain steps to activate or apply and may or may not have a cooldown; and unique actives, which require the physical usage of the item and has a cooldown on when you can use the item again.

A new problem is changing statistics. In other words, homeguard and furor. They don't have a set movement speed boost because they both rapidly decrease so setting the value at the maximum point isn't accurate because it is highly variable and most of the application of boost may not be at the maximum point on top of having a duration.

Q.How do we fix this?

A.Oh man. Oh man, oh man, oh man. This is a big one. One that I don't even think I can fully answer; but it's one I will try. Ok, one statistic I think that is possible is percentages. Now everyone knows it is variable according to other stats making it a hard statistic to determine, but it is definitely possible. Why is that? Because of Algebra.

Q.Dacluck you mean Algebra?

A.Ever heard of the substitution rule? Well we'll be using it to apply gold efficiency because it is a lot simpler than you think. Let's go cray cray with maths now. Percentages are parabolic functions when used in relation to base statistic and bonus statistic. In other words a 1% increase on 100 ms which is 1 bonus ms is nowhere near the same 1% increase on 1,000,000 ms which is 10,000 because it's not a linear relationship, it's exponential. So what is our y-intercept? Well, it's 0. Because all of your base ms is used in calculating bonus ms. So let's put this in a formula. Btw I'm sorry it's in text format because I don't know how to put it in formula format.

x * .01% = y

Eh? That's all? Yeah, it really is. let's put it in practice. 100 is x and 10% is your %. 100 * .01(1) = 1. HOLY CRAP that worked! Err... Isn't that what you always do when finding a percentage of something? Well yeah, of course it is. But let's take it a step further instead of stopping here.

x = 13 gold/ms

WOAH GUISE HOLD THE PHONE! WHAT DAFUQ AM I DOING, DAT X ONLY BELONGS TO MOVEMENT SPEEEEEEDDDSSSSS! Ahem... Well, what do we have trouble doing? Finding gold efficiency of %'s, so this is the obvious way of putting gold efficiency in our problem. Let's see how this pans out.

13 gold/ms * .01(1) = .13 gold/ms

HOLY SHIZNITES! Wait... What does this mean? Oh, it's simple really. When you buy anything dealing with %'s or ratios... You are not becoming more "gold efficient" because if you had 0 of that statistic, there would be no gold efficiency. No... Rather you are doing something crazier. You are essentially BUYING GOLD EFFICENCY! Oh yes, you heard me correct, buying gold efficiency. Cause what makes a % different from a ratio? Heh, just the form, in other words they are literally the same thing in a different form. So what the hell did I just figure out back there? Well, for every 1% of movement speed you gain, you get .13 gold/ms. Doesn't sound like much really. But I mean, rack up 400 flat ms and then 400 ms * .13 gold/ms = 52 gold. HOLY FUDGENUTS THAT'S HUGE. That literally makes it higher than the gold efficiency of critical strike. But anyways, let's put this in a scenario to exemplify how this can be used. So, you are and you have the highest base movement speed at 355 ms. You decide to buy  because you want to tread on Runeterra like Mercury. That gives you a bonus 45 flat movement speed. In this case your movement speed was worth 585 gold since 13 * 45 = 585. Now you decide, hmm, dis ain't enough speeds, I should get a zeal. And so you do, giving your 5% movement speed. In this case 5 * .13 gp/ms = .65 gp/ms. Now you multiply that .65 gp/ms with all your flat ms which means 45 + 355 = 400 ms. .65 gp/ms * 400 ms = 260 gold. So your 5% ms managed to reap 260 gold. Then 585 + 260 = 845 gold. That is how much all that movement speed you bought was worth. There are a couple restrictions but that is general jist of this. Stating that .13gp/ms is gained for every % ms you have is very accurate because essentially 101% gold efficiency from what would be 100%. In other words, 13.13/ms instead of 13gp/ms alone. That is what this wiki strives for right? Accuracy.

In regards to penetration, you can base that on armor because essentially, armor penetration is the loss of enemy armor in calculation for damage. This means that and it's armor penetration is saying that it removes 200 gold from the enemy since armor is 20gp/ar. Which makes SO MUCH SENSE seeing as though the rest of it's stats is 91% gold efficient, adding 200 gold to the item makes it 106% gold efficient which does display more accurately how it compares to other items. Same with magic penetration on which would add 300 gold into the mix giving it a 93% gold efficiency showing that it is still not as great as the other items but it's not as terrible as the 72.5% efficiency as it showed before. For the % penetration of each, you would go through the same process as I did before and to make it more simple, just move the decimal place 2 spots. In other words .2gp/ar and .2gp/mr.

Q.What do I expect to gain from this seemingly irrelevant fix to the problem?

A.Quite simple. I fixed a major problem that was unable to be fixed before. I'm not even a moderator, blogger, or big editor in this wiki. Heck I only made 221 edits so far! Yet I solved a problem that was a big issue in terms of gold efficiency. What I'm saying is that these problems that people argue about with our gold efficiency methods can be solved. It's a lot of work, but it is possible. I know that I am not giving you a direct answer and that's because I simply can't give you one, I evened warned that I didn't have a complete answer to the problem. I don't think we should leave gold efficiency in the cupboards as Willbachbakal said. I know it's painful to see the trouble that was caused because of our inaccuracies but we simply got to confront these problems that are pushed on us. And from it, I think we'll make strong progress in teaching the community how balanced these items really are. If anyone is afraid of our honor as a wiki being lost, then know this. The only thing you need to fear, is fear itself.