Board Thread:Wiki discussions and announcements/@comment-1990160-20140521040036/@comment-5266525-20140523055146

This discussion has really escalated since I last checked on it like several weeks ago.

I'm in favour of keeping gold efficiency's place on the item pages. Let's get that out there first. I feel like its main use as a comparison point between items hold enough worth to keep it here as it stands.

Here's some counterpoints I'd like to make towards arguments for removal made above:

1. Gold efficiency clearly does not represent an item's effectiveness and thus is incorrect.

This is bordering on critical research failure. Gold efficiency is independent of an item's effectiveness and this should effectively be considered an axiom of any debate surrounding it. It doesn't matter if you have good AP ratios or even any AP ratios at all when you buy a, you got more than what you paid for, which is 40 AP for 860g.

I feel like this is even the case for % penetration in the same way that armor does not have diminishing returns - you always have 35% penetration just like buying 100 armor/MR multiplies halves your received damage. The only value to which I think is not a good idea to assign a gold value is flat/percentage movement speed because of movement speed caps.

2. Gold efficiency does not influence item builds in any way.

If not for cost efficiency, why do the majority of laners buy a Doran's item when they start the game? Why is going with an aura tank build a good idea when a champion is behind? Why have the jungler and support roles been historically so hard to balance?

Cost efficiency is likely not at the forefront of a person's mind when they make purchases in the above examples but it is demonstrably a tangible reason for the purchase of certain items.

3. Gold efficiency cannot take into account most passive effects, and thus is misleading at best.

This is the biggest argument in my opinion and one I don't think is entirely inaccurate. It isn't misleading if you are aware of what gold efficiency is and how to manage it since it deals exclusively with quantitative statistics that we can measure, but our way of representing it might be misconstrued as negative feedback; items that do not meet 100% efficiency on stats alone are noted as 'for X item to be 100% gold efficient, the passive must have a value of Y'. This isn't technically incorrect, of course, but it may be taken as a subtle implication that an item can potentially NOT be 100% efficient and that this is a serious detriment towards buying the item. It could also be easy to mistake gold efficiency as being more important than, say, an item's slot efficiency due to its position near the top of every item page, as if to provide heavy emphasis on the stat.

Putting aside the issue that potentially not every item could be efficient (it's why I like having the baseline be the least efficient item of every stat), it's not enough to dismiss this as not the wiki's fault for a reader's mistake. We should be minimizing the chance for error and the use of controversial and vague statements.

As a personal solution, I feel like efficiency is core enough that it deserves a place in the item infobox rather than on the main space, but that's another topic.

4. We do not know how gold efficiency is calculated at Riot.

Aside from the fact that this is also partially in favour of gold efficiency due to implicitly acknowledging that it is used as a balancing tool, I do not feel like this is a reason to refuse to simply use our own model. Regardless of what you use as a baseline the relative efficiencies of items do not change - is always going to be more efficient than, for example, no matter how much gold a point of health is worth.

5. Gold value that changes with situational passive/active effects is nonsensical.

This one doesn't make a whole lot of sense to me. Logically, a person with, say, will have more gold's worth in stats whenever their active is active. This is simply a case of imposing a boundary condition on the original efficiency calculation.

I'm not aware of all the possible different... situations that have been mentioned in this thread and it could be on a case-by-case basis, but it's not inherently incorrect.

6. Against votes are currently leading against For votes.

Addressed mostly to Deshiba. Aside from me throwing in my vote for keeping efficiency, thus bringing keep/delete casts to a tie, voting for Change is inherently a vote to keep it as well, so Will isn't wrong to say the vote's currently 7 to 13. :P