Problems I have with the current Cooldown Reduction system + rework idea.

So the current system is simple enough, your cooldowns are reduced by a percentage amount equal to your cooldown reduction and the cap is 40%.

For example, you have an ability with a base cooldown of 2 seconds. You also have 40% cooldown reduction as a stat and because of that the cooldown of that ability is 1.2 seconds because 2 - 40% of 2 ( 0.8 ) equals 1.2.

So, where is the problem with this?

Well, first of all how cooldown reduction becomes more valuable the more of it you have. For example if you have a spell with a 2 second base cooldown and you have 20% cooldown reduction it will be a cooldown of 1.6 seconds. This means that you will be able to cast 37,5 spells per minute rather than just 30, which will be a 25% increase in the amount of spells you cast. Basically, 20% cooldown reduction means 25% increase in spells.

Now you would think that this would also mean that if you get 40% cooldown reduction instead of just 20%, so double as much, you will also increase your spells by double the amount, so 50%. But this is actually not the case.

So let's say you have 40% cooldown reduction on your 2 second spell, which means that the cooldown is now 1.2 seconds. This means you will have 50 spells per second instead of the 30 you would have without any cooldown reduction, but this isn't 50% more spells but actually 66% more spells.

In short, the efficiency of cooldown reduction doesn't scale linearly but exponentially, which means that it's rather unefficient if you don't have a lot of it ( 10%-20% ) but is super strong if you have a lot of it ( 30%-40% ). This is also why the cap exists, if you could reach more than 40% CDR it would become plain crazy, you just need to play a game of URF to confirm that, the 80% CDR you have there increases the spells you cast by a whopping 500%.

Now, why is this bad? Well because you can only really feel good about your cooldown reduction if you have a lot of it. This means that a build path must have either cooldown reduction on everything or cooldown reduction on very few things, because it will only be worth it if you are planning on reaching the cap or at least 30%.

This is for example why in professional play you don't see people sitting on just 10% CDR very often, they usually rush 40% all at once or ignore it altogether.

And I really don't like that. I feel that if you are a Mundo for example and you are sitting on a Spirit Visage and a Warmog's and a total of 20% CDR, you shouldn't feel like you are required to somehow squeeze 10-20% more of it in with weird items or components or feel bad about having an unefficient amount of it to begin with. On the flipside, I don't think it's all that great that mages for example have so much CDR spread unto every single one of their items that CDR becomes less of a choice where you think "Oh, I cast certain spells a lot maybe I should get more CDR" and more of a neat side effect, where you think "so, these are my core items and oh look, I guess I reached 40% CDR too, neat".

What would be the alternative to the system? Well the concept of it already exists in another place, namely with how armor and magic resist works. I actually don't need to explain it here myself, because there already exists a great wiki article on it: http://leagueoflegends.wikia.com/wiki/Armor.

What this means is that because of the diminishing returns armor has, you don't increase your effective health by 1% for your first point of armor, then by 1,01% for the next, then by 1.02% for the next and so on and so fourth, like cooldown reductions increases your effective spells, but consistently 1% for each point of armor you have. This way armor and MR scale linearly, whereas cooldown reduction scales exponentially.

And if you paid attention you will now know the easy fix to the Cooldown Reduction problem - you give it diminishing returns, so that it scales linearly.

We could rename cooldown reduction into something fancy, like casting speed, which uses the same formula as armor and magic resist. This would mean that for every point of casting speed ( or whatever the name is going to be ) you increase the spells you cast by 1%.

This fixes all the problems with the current CDR system - you don't need to worry about overshooting the cap, you don't need to rush 40% just for it to be efficient and you also don't have to feel bad if your build only includes 20% or less CDR. Instead of a secondary stat that is simply randomly slapped unto items, it will become a stat that has its own purpose and will only be on items that encourage you to actually spam spells.

Thoughts?