Board Thread:Wiki discussions and announcements/@comment-3308937-20150122092301/@comment-1330314-20150126013608

Delucion wrote: Willbachbakal: Hmmm, isn't cost, though, ultimately tied to effectiveness? Especially with so many stats clumped together on items. On Athene's, for instance, parts of its value can be completely discounted on manaless champions, but what if the value for the rest of it is high? I feel like this is important in cases like buying sheen on a manaless champ like Akali - there's a question of whether or not it's worth it, despite unnecessary mana.

You could technically argue that the mana on, while being unused, is still there, so regardless of who you purchase the item on you're still getting the same stats in the same amount on your item (the same goes for on ,  on , etc.). This also basically boils down to how we measure gold value on the wiki right now: an item's cost analysis is meant to be true regardless of which champion or build you pick it on, which means that effectiveness doesn't really factor into an item's value. This is also why the system you're proposing would, if correct, revolutionize the way we look at efficiency, because if we had a model for accurately measuring the effectiveness of an item on any build and/or champion (i.e. "plug in your champion and items and see exactly how much value this item has") then we'd be able to include that factor as part of our analyses. Instead of just factoring in stats, which is our current model, we'd be able to include synergy and champions in order to make each analysis tailored to every situation. That's also why I'm going in full inquisitor mode regarding the app you linked, because everything about it needs to be justified very well to integrate properly into the wiki without anyone else questioning its validity.

NeonSpotlight wrote: As someone who thinks that "cost efficiency" has been given a lot more importance/time than it actually deserves I still stick by what I said in the last discussion which was to get rid of the misleading "cost efficiency" and replace it with statistical/quantifiable values and to make that more of trivia info than an actual part of the article. With that in mind I think I like your #5 option the most if we do just use quantifiable stats (no things like Guardian Angel passive). Keeping along this train of thought I think that the ideal information we would give would just be things like "40 ability power = 870g" and not even touch on % statistically efficient since a lot of people get caught up on those % numbers.


 * 1) 5 seems to be popular! I'm really happy about this, because I think this would be the best way to take cost analyses off the article space for each item while still keeping them for those interested. I also think that we should get around to handling stats more like champions, and this would be the best fit: there's a ton of redundancy on the wiki due to items being copied across multiple articles, and it's a pain to update these whenever an item gets changed because it means having to perform multiple edits for the same thing across several pages (this is also why there's a lot of info that's out-of-date). If each item were to have its own data section, we could easily reimplement their presentation across all item-related articles (including the item's page itself), and modifying the item's data would update it across the whole wiki. On top of this, I think the cost analysis process can be completely automated, so given a template with an inbuilt series of rules you could plug in any item's data and have it calculate its cost analysis. Merging the cost analysis template into an item's infobox would allow both to run on the same data set.

Regarding percent statistical efficiency, I think it would be good to discuss its meaning and application(s) in greater detail, whether or not we're removing it (it would help make the decision). I'm personally in favor of keeping it, since it gives an idea of an item's budget and how its power is being distributed at any given time, but if it's bringing more confusion than information then it should definitely be removed.