|
https://spreadsheets.google.com/ccc?key=0AhhAugu1Z2aPdDBZc1oxYk9NckNqQ053RzhzY2ZHdkE&hl=en&authkey=CJ37vt8P#gid=0
pm or post an email address and I'll let you edit it
well I posted this in a LOL boards thread that only ever got like four views and now there are twenty people viewing it, so I made it more user friendly and added some features in case it becomes semi-popular
This is not really my work, as you can see - I made it readable, updated it and added some stats but that is all. Who is Wu Thuong? Is he a TL poster? The original version I saw had his name on it but I have no screen name to give e-credit to.
This thing is way better than when I first got my hands on it, but there's still much room for improvement... eg calculating regen properly, maybe calculating movespeed, adding gold values for skills, and so on.
Anyway this thread is for the discussion arising from this, and maybe just for general mathcraft surrounding the game.
|
Eh I can't agree with listing CDR as a base stat. Especially basing it off Kindlegem...
- You can't stack Kindlegems for CDR.
- Stinger gives cheaper CDR on an item that only has base stats as other stats.
- CDR is hardcapped.
Instead, MP5 and HP5 should be included as base stats, as those actually are available as stackble T1 items.
|
I agree with the above post. Kindegem isn't a base item, calculating the worth of CDR with it is like calculating the cost for gold per 10 with Kage's Pick/HoG/Philostone/Avarice. Yet they all have different cost-efficiency.
|
Xypherous posted a while back on the lol boards that you should calculate CDR off kindlegem so I just rolled with it.
I suggest you try the math on say, philo stone using regrowth pendant as a base item. It just doesn't work. The reason, I believe, is that you get increasing returns on larger amounts of regen, so higher regen items are actually less cost efficient as far as base stats vs gold, even though they might be just as or more effective. Notice how regrowth pendant gives less regen per gold than rejuvenation bead, whereas say BF sword gives more ad than longsword per gold.
Right now the only method I've found that consistently works is to cost them at the value of the stats they produce after 55 seconds (the 'per 5' value times 11). No scientific reason to pick 55, it just gives values close to the missing ones on items.
|
Notice how regrowth pendant gives less regen per gold than rejuvenation bead,
Uh, you might want to do your math again. 8 hp5 for 250 gold is 31.25 gold per hp5; 15 hp5 for 435 gold is 29 gold per hp5.
Faerie Charm is 60 gold per mp5 and Meki is ~55.8 gold per mp5, so it looks like it works there too.
|
oh i got the numbers backwards
lol ignore me
|
The spreadsheet seems broken.
A part of it seems to have been placed over the rest, breaking the references in the other cells. I'm not exactly sure how google docs work and where to move the rows so I can't really fix it myself.
|
yep, somebody vandalized it.
I have a backup saved in excel but no idea how to import it properly.
EDIT: Oh hey you can just go back to a past revision if you sign in.
So, no problem.
|
I froze the top row and leftmost column so it's easier to keep track of what you're actually looking at.
|
some recent findings
rank 5 gp buff is about 2500 gold worth of stats, same as blood boil and janna shield
leveling up gives about 550 gold worth of stats not counting regen or the skill point
baron gives about 20,000 (!) gold worth of stats and, uh, gold
On April 27 2011 16:34 crate wrote:
I froze the top row and leftmost column so it's easier to keep track of what you're actually looking at.
very nice!
occurred to me just now that armor pen might be calculated the same way armor is, that is that one point of armor pen = same cost as one point of armor
|
Sona and Taric both have some really sick multipliers:
Sona - 20 AD, 20 AP, 20 Armor, 20 MR - 2153 gold value per person hit, total value = 10765
Taric - 30(60) Armor, 45(90) AD - 2117 gold value per person hit, 4233 (!) value self-buffed, total value = 12701
Bonus fun:
Hyper-Kinetic Position Reverser: 4,026 gold
Defensive Ball Curl: 4,646 gold
Insanity Potion: 4,613 gold (+30% CC reduction, 65 HP5/MP5/Movespeed) [+1690g in movespeed, 839g in health, 650 in mana]
Monsoon: 1549 gold/target [600 HP healed / target, 7745 gold Maximum]
Meditate: 9,291 gold (!!) [+1807 gold worth of HP healed]
Yi OP. Janna pretty OP too.
|
Valuing HP and Shields equally, as well as valuing APen and Armor equally is pretty ridiculous.
|
On May 01 2011 05:20 spinesheath wrote:
Valuing HP and Shields equally, as well as valuing APen and Armor equally is pretty ridiculous.
APen and armor I can understand, but %health effects aside, how is 'gain 300 HP for 3 seconds' any different than 'gain a 300 absorb shield for 3 seconds'?
|
On May 01 2011 07:36 Niton wrote:Show nested quote +On May 01 2011 05:20 spinesheath wrote:
Valuing HP and Shields equally, as well as valuing APen and Armor equally is pretty ridiculous. APen and armor I can understand, but %health effects aside, how is 'gain 300 HP for 3 seconds' any different than 'gain a 300 absorb shield for 3 seconds'?
If you buy HP, you increase your max HP permanently. If you get a heal/shield, you don't increase your max HP at all and you don't have any lasting effects.
Obviously the two things you mentioned are the same, but neither of those can be assigned a value in a reasonable way.
|
why wouldn't you add in HP/5 Mana/5? I mean, they have basic items that just provide them, so it would be hella easy to add them in...
Edit: it's also pretty easy to factor in spell vamp off of Hextech Revolver (like you did with CDR and Kindlegem)
Also you should factor in the AP and Spell Vamp with Will of the Ancients like you did with other aura items.
Another interesting stat to look at is % efficiency (Value/Cost), as this is one of the best ways to drive home the point about doran's items being crazy good.
Also thanks, this is good work, I don't mean to be so critical, just wanna get everything in there that we can.
|
Make regens basic stats. Why aren't they?
Also, what's the reasoning for using Kindle instead of Stinger?
Also why would CDR be a basic stat and regens not, especially since they are unique passives?
Also, why does it have the new HoG stats, but the old LW stats?
|
On May 03 2011 06:26 Mogwai wrote:
Edit: it's also pretty easy to factor in spell vamp off of Hextech Revolver (like you did with CDR and Kindlegem)
Also you should factor in the AP and Spell Vamp with Will of the Ancients like you did with other aura items.
Spellvamp makes some sense because Hextech Revolver is the only "base" item to provide it: You can't get Spellvamp without grabbing a Revolver. It's also not unique or anything, so you can stack it like other base stats.
BUT you can't get Spellvamp items without paying for AP. Physical abilities also use spellvamp, so you might actually want spellvamp on champs that don't have a use for AP. For those champs, Spellvamp is much more expensive.
Then again it's the same issue with Mana items on manaless champs. So yeah I guess I support adding Spellvamp. Much more solid as a base stat than CDR for sure.
I guess it would be nice if there was a table with all the stats in the top row and the suggested gold value of each stat in the second row. Below that you'd have some sort of input fields where you can adjust the values to your likings: set mana/MP5 to 0 if you want to see the efficiency for ninjas, reduce the value of HP5/Lifesteal/Crit if you want to check out the efficiency of items in certain stages of the game (HP5 is bad lategame, lifesteal and crit are bad earlygame).
The reasoning to base CDR off Kindlegem is most likely that HP pretty much isn't wasted on any champ, while ASpd is pretty useless on half of the lineup. Still pretty arbitrary, and the fact that it's unique doesn't help either.
|
Spellvamp doesn't seem worth adding since you get it off so few items... just gunblade and wota, I don't think it's even worth the extra column.
On May 03 2011 07:25 spinesheath wrote:Show nested quote +On May 03 2011 06:26 Mogwai wrote:
Edit: it's also pretty easy to factor in spell vamp off of Hextech Revolver (like you did with CDR and Kindlegem)
Also you should factor in the AP and Spell Vamp with Will of the Ancients like you did with other aura items. I guess it would be nice if there was a table with all the stats in the top row and the suggested gold value of each stat in the second row. Below that you'd have some sort of input fields where you can adjust the values to your likings: set mana/MP5 to 0 if you want to see the efficiency for ninjas, reduce the value of HP5/Lifesteal/Crit if you want to check out the efficiency of items in certain stages of the game (HP5 is bad lategame, lifesteal and crit are bad earlygame).
That's pretty much what the test slots at the bottom are for, you can just plug in whatever numbers you like and it'll compare gold for value. Copy paste the numbers off an existing item to save time.
On May 03 2011 06:30 Shikyo wrote:
Make regens basic stats. Why aren't they?
If you calculate philo stone off regrowth pendant and meki it's worth over 1200 gold (hint: it's not actually worth over 1200 gold). I like the 55 second method but until I have some understanding of how it works and whether it's actually reasonably accurate you probably shouldn't add it.
Also, why does it have the new HoG stats, but the old LW stats?
Because you didn't edit it when you found the error...? It's really simple shikyo, double click the wrong number and change it to the right one.
Also, what's the reasoning for using Kindle instead of Stinger?
already answered
On May 01 2011 18:12 spinesheath wrote:Show nested quote +On May 01 2011 07:36 Niton wrote:On May 01 2011 05:20 spinesheath wrote:
Valuing HP and Shields equally, as well as valuing APen and Armor equally is pretty ridiculous. APen and armor I can understand, but %health effects aside, how is 'gain 300 HP for 3 seconds' any different than 'gain a 300 absorb shield for 3 seconds'? If you buy HP, you increase your max HP permanently. If you get a heal/shield, you don't increase your max HP at all and you don't have any lasting effects. Obviously the two things you mentioned are the same, but neither of those can be assigned a value in a reasonable way.
I agree that these things aren't 'really' the same, but I think the comparison is close enough that it's useful to check these values out.
On May 03 2011 06:26 Mogwai wrote:
Also thanks, this is good work, I don't mean to be so critical, just wanna get everything in there that we can.
The base isn't mine and that's the difficult/nerdy part, I did add a few things but mostly I just updated it from being like six months out of date and made it more readable. I don't think of it as my project lol
|
oh also apparently riot prices movespeed at 7 gold per movespeed unit
|
On May 03 2011 11:09 UniversalSnip wrote:Show nested quote +On May 01 2011 18:12 spinesheath wrote:On May 01 2011 07:36 Niton wrote:On May 01 2011 05:20 spinesheath wrote:
Valuing HP and Shields equally, as well as valuing APen and Armor equally is pretty ridiculous. APen and armor I can understand, but %health effects aside, how is 'gain 300 HP for 3 seconds' any different than 'gain a 300 absorb shield for 3 seconds'? If you buy HP, you increase your max HP permanently. If you get a heal/shield, you don't increase your max HP at all and you don't have any lasting effects. Obviously the two things you mentioned are the same, but neither of those can be assigned a value in a reasonable way. I agree that these things aren't 'really' the same, but I think the comparison is close enough that it's useful to check these values out.
If you want to compare shields and heals to anything, compare them to Potions or Red Elixiers.
On May 03 2011 11:38 UniversalSnip wrote:
oh also apparently riot prices movespeed at 7 gold per movespeed unit
I guess you say that because Boots1 cost 350 gold and provide 50 Movespeed? Well, Boots3 cost 1000 gold and provide 90 Movespeed. That's 11 gold per Movespeed.
Quite obviously Riot's thinking here is that flat Movespeed should be more expensive the more you get.
|
was stalking the red tracker and phreak literally posted 'movespeed is costed at 7 gold per unit'... pretty handy haha
It's interesting they build diminishing returns into both the movement mechanic and the items
Somebody requested sharing on the sheet (which is like admin access I think? spreadsheet admin lol) which I'll happily give, but they gotta post here so I know who it isn't just somebody looking to vandalize it
|
Movementspeed is like range, each additional unit is more valuable than the last. MS can cost 7 gold because you can´t stack boots.
Phreaks quote in particuar was in context of Gold Quints to prove that they are the most "valuable".
|
This spreadsheet is awesome! I am a new player and have been working through the math on this sort of thing for some of the items I consider buying for champions. Now I won't have to do it manually! Thanks.
Also, dumb question, but what exactly does it mean when a field is green or red? I noticed that some fields are positive and some are negative w/o them being green/red, which confused me.
|
On May 12 2011 22:18 JAJAJAGuy wrote:
This spreadsheet is awesome! I am a new player and have been working through the math on this sort of thing for some of the items I consider buying for champions. Now I won't have to do it manually! Thanks.
Also, dumb question, but what exactly does it mean when a field is green or red? I noticed that some fields are positive and some are negative w/o them being green/red, which confused me.
it means the item is unusually efficient or inefficient
|
Is the spreadsheet fucked up for anyone but me?
|
Looks like some tards thought it would be fun to delete everything etc.
Apparently they didn't know that restoring old revisions is just a few clicks, otherwise I doubt they would even have bothered.
|
Wait Tiamat is so inefficient. Korean meta is so confusing -_-
Im also surprised by FoN inefficiency.
|
Movespeed, HP5 and MP5 aren't properly modeled. At 2k HP, FoN's worth 75 HP5, which translates to 825 HP over the suggested 55 second window, which translates to 2128 gold in 'health'.
|
yeah i've never really been satisfied with either method of calculating regen lol... the 55 second window seems to generate unrealistic numbers less often. I'm certain it's not the correct method though
I haven't thought tiamat is really bad for a while btw, just never ran through and changed the colors from the original creator's
|
Wow really cool to see. This will definitely effect my play.
|
On July 31 2011 02:45 Kenpachi wrote:
Wait Tiamat is so inefficient. Korean meta is so confusing -_-
Im also surprised by FoN inefficiency.
Tiamat offers pure intangibles that can't really be quantified by anything. Basically they give heroes who don't normally have the ability to threaten a split push the ability to do so, while farming like mad.
|
what is split pushing o_O
|
On July 31 2011 08:44 Kenpachi wrote:
what is split pushing o_O
pushing 2 lanes at once. ie master yi going off by his own and killing a lane while his team sits and holds 4v5
|
Necroing this thread because the cost efficiency spreadsheet is ruined atm!
|
Fixing it.
Should be fine again. I wonder how long it will last.
|
Dyrus Deleted the sheet....
Not cool
|
On September 04 2011 08:32 jaybrundage wrote:
Dyrus Deleted the sheet....
Not cool
Really?
Restored again. zzzzzzzzzzzzzzzzzzzzzzzz
|
FoN being so cost inefficient is quite shocking..
Though i dunno, its always seemed like one of the best items in the game to me
|
On September 04 2011 18:13 arb wrote:
FoN being so cost inefficient is quite shocking..
Though i dunno, its always seemed like one of the best items in the game to me
The spreadsheet doesn't cost regen and movespeed properly which is a big part of FoN
|
On September 04 2011 18:19 BlackPaladin wrote:Show nested quote +On September 04 2011 18:13 arb wrote:
FoN being so cost inefficient is quite shocking..
Though i dunno, its always seemed like one of the best items in the game to me The spreadsheet doesn't cost regen and movespeed properly which is a big part of FoN
inst the regen even "free" in the spreadsheet?
|
yeah regen simply isn't calculated because the numbers come out as nonsense. There's a way but I don't know it. just have to eyeball the true value for now
|
theres a whole bunch of 'circular dependance errors'
|
suppose I should just disable public editing and give out keys to whoever asks so dyrus can't come in here and fuck it up
maybe i'll figure that out tonight
|
You now have to be logged on to edit it, otherwise no editing restrictions... I can just ban someone who messes it up
|
I can't edit this even when logged in to my google docs.
I just wanted to change Abyssal Scepter to say "Paying for 20 MR reduction aura" instead of penetration... they are different!
The only other item that has MR reduction is Malady and it is correctly stated in the shee
also here is a link if you want to invest some time figuring out how magic damage is calculated
|
If anyone really cares about regen or spellvamp being included I went ahead and uploaded my version of an item efficiency spreadsheet here. I had actually made this before this topic was created but figured it didn't really add much so I didn't bother to post it until now.
Main differences are that I valued items compared to the basic, low-tier items (longsword, amp tome, etc.) and obviously that I included regen and spellvamp (the latter based on Hextech Revolver, assuming that all the combine cost is the spellvamp cost). Also includes an efficiency column which is the actual cost divided by the effective cost (i.e. cost of all the priced stats) to make it a bit easier to compare the efficiency of items with different costs (e.g. it's really hard to compare Rod of Ages to stacking Doran's Rings on USnip's sheet because their base price difference is just so vast).
All items are sorted purely by name. Has no pretty colors.
Comparing to basic items instead of higher-tier single-stat items does adjust the price of AD compared to AP, as my sheet says a BF Sword is significantly more efficient than a Needlessly Large Rod (this is a result of Riot cutting the BF Sword cost a while back without changing its stats). Otherwise it's not a big deal, it just means that items are going to look more efficient in my sheet compared to USnip's.
I might have a few stats wrong (almost certainly some of the misc stuff is wrong, because I don't actually look at that very often myself) because I didn't update this for a while until this weekend; you can tell me if I have anything wrong and I'll probably get around to fixing it.
Won't be publicly editable because I don't want to have to reset it.
|
crate, I'll add your sheet to the OP. I really do think wu thuong was smart not to include regen in the item calculations though because the results are completely meaningless. Look at how mindbogglingly efficient reverie, miracle, etc are when you calculate straight up off pendant/etc, if they were actually worth that much gold they'd be must build on everyone.
On September 07 2011 07:51 redex777 wrote:
I can't edit this even when logged in to my google docs.
ok well the the options are totally misleading than, lol
new setup: post here asking for permission to edit and I'll give it. Needs email address if you don't want crawlers to find it I guess you could PM it to me. fixed the scepter thing btw
|
Well the thing is, look at Archangel's staff with maxed out mana. Supposedly that's disgustingly cost effective as well ... but most of that is because you have an item that gives you 1400 fucking mana which 90% of the champs will never be able to use (so it's wasted). Or look at Rageblade and how it's like the most realistically imba item in the game at full stacks, and even if you can only get to 3 or so it rivals almost any other item this side of stacked up RoA/Bloodthirster ... but most champs can't effectively use all three stats so it's worth much less.
Probably the most useful way to look at these is to break down the items into how much each of their individual stats are worth (e.g. Bloodthirster is worth this much gold for the AD alone and this much for the lifesteal alone) but I'm too lazy to go make that myself at the moment. If you have that then you can decide how much a Gunblade is worth on Pantheon or a Triforce on Lee Sin or whatever.
It's not a problem unique to regen, which is why I think adding it in is just fine. I realize it's more extreme for regen than most stats (regen doesn't do much at all in quick teamfights) but if it bothers you well we have the spreadsheet you're maintaining to look at no-regen costs, which I can see the point of examining.
|
On September 07 2011 09:55 crate wrote:
this is a result of Riot cutting the BF Sword cost a while back without changing its stats
They reduced the AD on BF Sword to 45.
|
On September 07 2011 12:33 crate wrote:
Well the thing is, look at Archangel's staff with maxed out mana. Supposedly that's disgustingly cost effective as well ... but most of that is because you have an item that gives you 1400 fucking mana which 90% of the champs will never be able to use (so it's wasted). Or look at Rageblade and how it's like the most realistically imba item in the game at full stacks, and even if you can only get to 3 or so it rivals almost any other item this side of stacked up RoA/Bloodthirster ... but most champs can't effectively use all three stats so it's worth much less.
Probably the most useful way to look at these is to break down the items into how much each of their individual stats are worth (e.g. Bloodthirster is worth this much gold for the AD alone and this much for the lifesteal alone) but I'm too lazy to go make that myself at the moment. If you have that then you can decide how much a Gunblade is worth on Pantheon or a Triforce on Lee Sin or whatever.
It's not a problem unique to regen, which is why I think adding it in is just fine. I realize it's more extreme for regen than most stats (regen doesn't do much at all in quick teamfights) but if it bothers you well we have the spreadsheet you're maintaining to look at no-regen costs, which I can see the point of examining.
The reason they are cost effective is because of the fact that most people can't use all the stats to full potential. Riot isn't making OP items you know. There's a trade off for everything.
But yeah, this spreadsheet isn't that help ful at all most things you can work out yourself with run-->calc in 20 seconds for the most complicated items.
|
On September 07 2011 17:46 spinesheath wrote:Show nested quote +On September 07 2011 09:55 crate wrote:
this is a result of Riot cutting the BF Sword cost a while back without changing its stats They reduced the AD on BF Sword to 45.
Oh, lol, nevermind then, I guess BF Sword was just always more efficient than NLR.
|
I agree with some people in this thread Hp5 hast to be integrated as well
|
On September 07 2011 12:33 crate wrote:
Well the thing is, look at Archangel's staff with maxed out mana. Supposedly that's disgustingly cost effective as well ... but most of that is because you have an item that gives you 1400 fucking mana which 90% of the champs will never be able to use (so it's wasted). Or look at Rageblade and how it's like the most realistically imba item in the game at full stacks, and even if you can only get to 3 or so it rivals almost any other item this side of stacked up RoA/Bloodthirster ... but most champs can't effectively use all three stats so it's worth much less.
Probably the most useful way to look at these is to break down the items into how much each of their individual stats are worth (e.g. Bloodthirster is worth this much gold for the AD alone and this much for the lifesteal alone) but I'm too lazy to go make that myself at the moment. If you have that then you can decide how much a Gunblade is worth on Pantheon or a Triforce on Lee Sin or whatever.
It's not a problem unique to regen, which is why I think adding it in is just fine. I realize it's more extreme for regen than most stats (regen doesn't do much at all in quick teamfights) but if it bothers you well we have the spreadsheet you're maintaining to look at no-regen costs, which I can see the point of examining.
there's a limit to the information you can draw from these charts, when you reach these limits I think it's better to just let people draw their own conclusions about the items than include information that's actually misleading
|
I don't think including regen is misleading.
|
On September 08 2011 10:44 crate wrote:
I don't think including regen is misleading.
Thing is, regen costs less the more you get of it, because the stat is only REALLY useful in the early game. You could say, for example, that FON or a stacked up Warmogs is like a billion times more efficient on HP regen than a Regrowth pendant, but that doesn't mean much at all.
|
It seems unlikely to me that the whole philo tree is as efficient as calculating regen linearly indicates... not even getting into the other regen items
|
On September 08 2011 11:17 UniversalSnip wrote:
It seems unlikely to me that the whole philo tree is as efficient as calculating regen linearly indicates... not even getting into the other regen items
Well, I guess it could be useful comparing Meki to FaerieCharm to Philo and same with health pendant/bead/philo crap - the early game stuff. And just ignore warmogs and FON and the other weird late game items that happen to have regen but are hard to place a value on that regen.
|
Updated my sheet. Now has some color in the efficiency column. Also new sheet 2 that breaks down what you're actually paying for on an item like I suggested would be helpful.
Bit of behind-the-scenes calculation adjustment as well but you won't notice that (other than it fixed the cost of CDR because apparently Kindlegem got a price increase or something when I wasn't looking, or I entered the data wrong in the first place).
Added the Dominion items as well though obviously those aren't particularly useful at the moment. Didn't add the Dominion items to the "on-hits as AD" section yet.
|
just a quick edit, youre not paying for just 15 apen on brutalizer, but also 10% CD.
|
|
|
|
|
|
|
|
EPT
