Board Thread:Wiki discussions and announcements/@comment-4881935-20140514222931/@comment-1330314-20140520024106

I think what Deshiba's post is trying to get to is actually completely tangential to the issue of gold value and gold efficiency. When evaluating a on, you're not actually measuring its gold value or gold efficiency, but rather its effectiveness. The fact that an has no use whatsoever on a  does not prevent the fact that you just purchased 435 gold's worth of stats on her, no matter how useless those stats are. In the end, effectiveness is dependent on champions, and not on items. On a more subtle level, a will be far more effective on a  than a mid, for example, despite the fact that both are AP champions, because of Xerath's far superior AP ratios. Does that mean the gold value of an item changes from champion to champion? Absolutely not.

The point I'm trying to make here is that, when evaluating the gold value/efficiency of an item, the whole measurement should be based on the item and the item alone, not on some potential interaction with a champion's kit. That should belong to the champion's own strategy article, where you can specify which items are particularly effective on champion X and which are not (i.e. " is not recommended on "). Again, the only reason this is misleading is because you're assigning properties to gold value that do not exist, and that would be logically impossible to have in the first place (no item is equally valuable on literally every champion out there, and shouldn't be considered as such).

This also extends to the rest of gold value: one of the reasons I think there's a disagreement here is because you're evaluating the effectiveness of an item with respect to one or more specific champions, or are thinking up scenarios that don't really have to do with the item itself. At the risk of sounding like an accountant, the gold value of an item should be equivalent to looking at a spreadsheet, the spreadsheet in question being the stats provided by the item at a given time. If you were to take the enchantment, for example, and you evaluated the item as you are sitting in base, then on your spreadsheet you would see 11405g's worth of movement speed. It doesn't matter where the movement speed's coming from, how long it lasts, or how ridiculous the gold value looks, it is a true and informative statement that the enchant provides that much gold's worth of stats at the time of measurement. If you somehow managed to buy 11405 gold's worth of percent movement speed and sat in base, the movement speed you'd get at that specific time would be identical to the movement speed from Homeguard in every single way.

One of your main arguments, ClariS, stems from simple disbelief: sure, the gold provided by certain item actives, conditional passives or other triggers can be unbelievably massive, but what's wrong with that? Again, the item is, at that specific time, providing that much gold's worth of stats, and censoring it outright just forces the readers to go out and make the calculations themselves. If you specify that the gold value of the item is only appropriate to that situation (i.e. is only worth 4670g while its active is up), informing readers of the item's gold value during that situation is an indisputably true statement, and cannot be misleading as it would specify that the gold value is limited to that situation.

You keep on insisting that the gold value of an item is fixed (a statement that is, by the way, completely unfounded), but it's very easy to prove that this is not true. Again, take : at zero stacks, it gives 40 AP and 30 AD. The gold value of every stat, including these stats, is fixed, so that much won't change. Fast forward eight autoattacks later, and the item provides a total of 72 AP, 30 AD and 32% AS. Again, because the value of stats are fixed, you get a different gold value just by looking at the spreadsheet. More to the point, a Guinsoo's with no stacks and a Guinsoo's with stacks do not give the same stats. You cannot hold one version of the item to the gold value of the other. By your own logic, it would be just as valid to measure the gold value of Rageblade at full stacks and with the surge passive up (5091g) and hold it as the default, under the reasoning that Rageblade having no stacks is itself situational, and therefore not worth mentioning. In fact, considering that the champion using this is going to spend a lot of time autoattacking continuously, and will therefore almost always have stacks up on the item, it might even be the more accurate measure.

Point being, you're not backing up your statements, which also happen to be inconsistent. On the other hand, the above examples clearly demonstrate that the value of an item is not fixed, and therefore that we need to shed light on these situations in the item's cost analysis to inform the readers. It's the current method of obfuscating all of this information that is misleading, not the information itself.