I am currently in the process of developing an item valuation tool. The tool is not intended to make suggestion whether an item is an improvement for you... it is there to analyze whether an item has received favorable rolls that could benefit a certain class/build.
This tool will be able to analyze an item based on a combination of the following points:
- Item Type
- Item Level
- Rolled Affixes
- Rolled Affix values (compared to their min/max values)
Based on this input it can give an estimate on the quality of the item. Currently I have the tool working and my intention is to share this with the community as I think it will make life a whole lot easier when you're doubting to vendor/ah an item.
Let's for instance look at some random item that I can plug into the tool:
Once plugged in the tool it will tell me that it has a 58,64% quality indicator for these affixes, based on the “completion percentage” per affix (str: 37,47%, vit: 82,61%, GF: 50% and AR 63,77%). There is some more logic as to the number of affixes on the item but I’ll leave that out to keep the discussion focused J. This in itself is a pretty good indicator, however I would like to take this a step further.
So what?
This is where you (the community) come in. I would like to ask every class/build to provide a rating to the below affix list in order to determine weights for each affix per class combination. Since each class/build will weight affixes and their respective values differently I would like the tool to give a quality indicator on a class/build basis. Since it will take a lot of work determining these weights I thought it would be a better and more reliable way to “crowdsource” this to the community.
If you would be so kind and awesome to fill in the list of all the different affixes and give a rating between zero (0) and one (1) where 0 is not important and 1 is the most important then we’d be well underway in making all our lives a little bit easier . I would like to ask you to fill in from which build perspective you have filled in this list. Thanks for the feedback!!!
The list can be found in the following google spreadsheet:
Mind the following things:
- please fill in your ratings in the appropriate class sheet (found at the bottom of the docs) since we don't want barbarians preferring int for instance
- please fill in the name of the build for which you filled in the ratings
Thanks for that Zubacz! I know that the data that is being deliverd is subjective, but that's ok, as long as we have a big enough set I think it will be reliable enough to use. Be sure to check in once i post the tool!
I can understand that there is some skepticism surrounding the usefulness and reliablity regarding this "weighting" exercise, so let me give a little bit more background.
First of all there are other tools out there that can indicate whether an item is considered an upgrade to you; this tool looks at items decoupled from any other contextual values that may be present for your character/build . It is to identify which stats in general on a class to class basis are more important.
Secondly please do not start discussions about the statistical relevance of this excercise; I know it's not a perfect represenation/sample, however it's better then nothing and would greatly help the quality of the "quality indicator" (sorry for that one :))
TLDR: Consider this a tool where your character is not important (I know that's hard for some of you, lol) and your gear is not being considered. It's purely looking at an item on it's own and the respective quality of the roll... interdependency between the affixes will be incorporated with the stat weights; this will be done on a class to class basis
Yea and what you'll have is a useless tool, telling you what? Nothing relevant, or outright wrong information. So it's not better than nothing, it's worse than nothing.
I know your intentions are good, but what you're doing is just gonna mislead people.
Wow... the internet never seizes to amaze me. You haven't seen the tool nor do you understand what it is intended for and yet your already are and condemning me for creating it. Let me try to explain one more time.
It looks at what is possible on a certain ilvl/item combination (both the affixes and their respective possible values) and checks how well the item rolled compared to what it could have rolled. The weights come in to identify stat importance (which can be used to develop interdependencies with other stats) and this will be based on information gathered from the community. I understand that this information will be far from perfect, however it's a start and we can work on it from there.
Considering the fact that a lot of information regarding affix/value rolls on items is not available to the majority of the people in a quick and easy way I believe this will not be a worthless tool. If you don't get it or you don't believe in it that is fine, however please refrain from flaming. Instead try a constructive approach, I am more then happy to listen to suggestions.
It is to identify which stats in general on a class to class basis are more important.
This is where it goes wrong. For me, attack speed is not important at all, i actually try not to get any at all because it makes me run out of arcane power since i dont have any signature spell in my build. For other people with different builds, attack speed is very important. You see, there is no such thing as a general quality indicator based on class.
We cant give a stat a rating without taking our spec into consideration, it just doesn't work like that.
If people want to see the quality of their item, they just need to type in the stats in the AH and see what similar items sell for; thats their value right there (value = quality for obvious reasons).
Alright, thanks for the feedback, I understand what you guys are getting at. So maybe we can make it go a level deeper and specifiy the weights on a build to build basis. I do understand the difference between the builds and their respective favoritism towards certain stats over others. I will update my posts accordingly. Please fill in the list from the perspective of your build (for instance melee wiz, or blizz/hydra) and mention for which build it is. These builds can be incorporated in the tool to identify if an item is good for a certain build.
Again thanks for the feedback, I know it's still work in progress, but I do believe this could be a good tool to use.
Perhaps instead of having the tool dictate to the user the stats they should use per their build/spec, the user could manually set their own stat weights?
It seems to me this would make the tool much more useful and flexible in many more situations. IE "Currently I am very low on X, so even though Y is usually better for me, I should increase the value of X to help me decide on this piece of gear."
If this were one day paired with a simulation tool (like simcraft for WoW) the tool would be able to automatically set stat weights for people with a reasonable degree of success- until then I'm not so certain you can.
Perhaps instead of having the tool dictate to the user the stats they should use per their build/spec, the user could manually set their own stat weights?
It seems to me this would make the tool much more useful and flexible in many more situations. IE "Currently I am very low on X, so even though Y is usually better for me, I should increase the value of X to help me decide on this piece of gear."
If this were one day paired with a simulation tool (like simcraft for WoW) the tool would be able to automatically set stat weights for people with a reasonable degree of success- until then I'm not so certain you can.
The tool is not intended to make suggestion whether an item is an improvement for you... it is there to analyze whether an item has received favorable rolls that could benefit a certain class/build.
@ Zubacz: thanks for clarifying this. mad props to someone who can shine a light in dark cornes (the internet)
The point is, the values of the stats are wrong since people value them differently. An item with attack speed might be worth alot on AH because its good for some people, but the value in this tool is average because people valued it low (because its bad for them). And people cant just say, "alright, i'll rate this based on no spec" because that doesn't make sense; how do i rate 2 stats against each other based on no spec, it cant be done.
Another thing about the AH value of items, is that its more the combination of stats on an item that makes it worth alot. An item could easily have alot of "high quality" stats based on this tool, but if the combination of those are bad, then it will not sell, even if all the stats on their own are great.
Thirdly, why would you need a tool to determine AH value ? you're gonna search for the specific values anyway to get the exact price when you put it on. Its not like you'll see it has 0,6 rating and then just put it on for 5 mil without even checking the market.
@Drez, read the main topic and what I replied to you earlier... it states that this tool is not to be used to see if an item is an upgrade for you as a player, but whether the item has received good rolls for a certain class/build combination. It does not consider what gear that hypothetical player is wearing.
@Biggz: As also mentioned in the main topic, I would like to incorporate different class/build combinations. This would result in a "wet-finger" indicator about the quality of an item for a respective build/class combination.
I appreciate the discussion, but please keep it constructive.
Not to mention... the DiabloFans forum basis is SMALL. Not that big of a source of people to pull information from.
Despite being affiliated with the Curse.com network of sites, this site doesn't see anywhere near as much traffic (evident by the fact if you watch the forums overnight, nothing moves, where for example, on mmo-champion.com, it is constantly moving, 24/7).
Not to mention... the DiabloFans forum basis is SMALL. Not that big of a source of people to pull information from.
Despite being affiliated with the Curse.com network of sites, this site doesn't see anywhere near as much traffic (evident by the fact if you watch the forums overnight, nothing moves, where for example, on mmo-champion.com, it is constantly moving, 24/7).
I will see if I can get it up on MMO and battle.net as well... nonetheless I believe in the knowledge that is circulating on the web and I believe that we can create a great list as a community (also outside of diablofans.com :P)
if at all possible it would be best to list items and the price they were actually sold for and compile that somehow to affix-combinations that are worth more.
but this would require sooooo much data.
and how would we gather that data... that would require everyone keeping their own scores...
Basically I feel that currently prices are all over the place and there is no solid foundation for what is being asked as a price. Sure the market will figure it out in the long run, especially with supply growing as we speak. However i do feel that the market could operate on a little bit more transparent basis, hence this tool.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
I am currently in the process of developing an item valuation tool. The tool is not intended to make suggestion whether an item is an improvement for you... it is there to analyze whether an item has received favorable rolls that could benefit a certain class/build.
This tool will be able to analyze an item based on a combination of the following points:
- Item Type
- Item Level
- Rolled Affixes
- Rolled Affix values (compared to their min/max values)
Based on this input it can give an estimate on the quality of the item. Currently I have the tool working and my intention is to share this with the community as I think it will make life a whole lot easier when you're doubting to vendor/ah an item.
Let's for instance look at some random item that I can plug into the tool:
Item Type Belt, Ilvl 62
Strength: 150
Vitality : 186
Gold find: 15%
All resistances: 50
Once plugged in the tool it will tell me that it has a 58,64% quality indicator for these affixes, based on the “completion percentage” per affix (str: 37,47%, vit: 82,61%, GF: 50% and AR 63,77%). There is some more logic as to the number of affixes on the item but I’ll leave that out to keep the discussion focused J. This in itself is a pretty good indicator, however I would like to take this a step further.
So what?
This is where you (the community) come in. I would like to ask every class/build to provide a rating to the below affix list in order to determine weights for each affix per class combination. Since each class/build will weight affixes and their respective values differently I would like the tool to give a quality indicator on a class/build basis. Since it will take a lot of work determining these weights I thought it would be a better and more reliable way to “crowdsource” this to the community.
If you would be so kind and awesome to fill in the list of all the different affixes and give a rating between zero (0) and one (1) where 0 is not important and 1 is the most important then we’d be well underway in making all our lives a little bit easier . I would like to ask you to fill in from which build perspective you have filled in this list. Thanks for the feedback!!!
The list can be found in the following google spreadsheet:
https://docs.google.com/spreadsheet/ccc?key=0AgW_oxdCKqchdDZlSmREUkJSLU90cUs2SnZHWnZLT3c
Mind the following things:
- please fill in your ratings in the appropriate class sheet (found at the bottom of the docs) since we don't want barbarians preferring int for instance
- please fill in the name of the build for which you filled in the ratings
Hoping we can make somethings awesome together!
Kind regards,
Staalie
First of all there are other tools out there that can indicate whether an item is considered an upgrade to you; this tool looks at items decoupled from any other contextual values that may be present for your character/build . It is to identify which stats in general on a class to class basis are more important.
Secondly please do not start discussions about the statistical relevance of this excercise; I know it's not a perfect represenation/sample, however it's better then nothing and would greatly help the quality of the "quality indicator" (sorry for that one :))
TLDR: Consider this a tool where your character is not important (I know that's hard for some of you, lol) and your gear is not being considered. It's purely looking at an item on it's own and the respective quality of the roll... interdependency between the affixes will be incorporated with the stat weights; this will be done on a class to class basis
Thanks
Wow... the internet never seizes to amaze me. You haven't seen the tool nor do you understand what it is intended for and yet your already are and condemning me for creating it. Let me try to explain one more time.
It looks at what is possible on a certain ilvl/item combination (both the affixes and their respective possible values) and checks how well the item rolled compared to what it could have rolled. The weights come in to identify stat importance (which can be used to develop interdependencies with other stats) and this will be based on information gathered from the community. I understand that this information will be far from perfect, however it's a start and we can work on it from there.
Considering the fact that a lot of information regarding affix/value rolls on items is not available to the majority of the people in a quick and easy way I believe this will not be a worthless tool. If you don't get it or you don't believe in it that is fine, however please refrain from flaming. Instead try a constructive approach, I am more then happy to listen to suggestions.
Thanks
This is where it goes wrong. For me, attack speed is not important at all, i actually try not to get any at all because it makes me run out of arcane power since i dont have any signature spell in my build. For other people with different builds, attack speed is very important. You see, there is no such thing as a general quality indicator based on class.
We cant give a stat a rating without taking our spec into consideration, it just doesn't work like that.
If people want to see the quality of their item, they just need to type in the stats in the AH and see what similar items sell for; thats their value right there (value = quality for obvious reasons).
Again thanks for the feedback, I know it's still work in progress, but I do believe this could be a good tool to use.
It seems to me this would make the tool much more useful and flexible in many more situations. IE "Currently I am very low on X, so even though Y is usually better for me, I should increase the value of X to help me decide on this piece of gear."
If this were one day paired with a simulation tool (like simcraft for WoW) the tool would be able to automatically set stat weights for people with a reasonable degree of success- until then I'm not so certain you can.
The tool is not intended to make suggestion whether an item is an improvement for you... it is there to analyze whether an item has received favorable rolls that could benefit a certain class/build.
@ Zubacz: thanks for clarifying this. mad props to someone who can shine a light in dark cornes (the internet)
The point is, the values of the stats are wrong since people value them differently. An item with attack speed might be worth alot on AH because its good for some people, but the value in this tool is average because people valued it low (because its bad for them). And people cant just say, "alright, i'll rate this based on no spec" because that doesn't make sense; how do i rate 2 stats against each other based on no spec, it cant be done.
Another thing about the AH value of items, is that its more the combination of stats on an item that makes it worth alot. An item could easily have alot of "high quality" stats based on this tool, but if the combination of those are bad, then it will not sell, even if all the stats on their own are great.
Thirdly, why would you need a tool to determine AH value ? you're gonna search for the specific values anyway to get the exact price when you put it on. Its not like you'll see it has 0,6 rating and then just put it on for 5 mil without even checking the market.
@Biggz: As also mentioned in the main topic, I would like to incorporate different class/build combinations. This would result in a "wet-finger" indicator about the quality of an item for a respective build/class combination.
I appreciate the discussion, but please keep it constructive.
Despite being affiliated with the Curse.com network of sites, this site doesn't see anywhere near as much traffic (evident by the fact if you watch the forums overnight, nothing moves, where for example, on mmo-champion.com, it is constantly moving, 24/7).
I will see if I can get it up on MMO and battle.net as well... nonetheless I believe in the knowledge that is circulating on the web and I believe that we can create a great list as a community (also outside of diablofans.com :P)
Secondly let's try to get this back on track with the people who are willing to contribute. This would mean the following.
Thanks for contributing once again!
Kr,
Staalie
and how would we gather that data... that would require everyone keeping their own scores...
Basically I feel that currently prices are all over the place and there is no solid foundation for what is being asked as a price. Sure the market will figure it out in the long run, especially with supply growing as we speak. However i do feel that the market could operate on a little bit more transparent basis, hence this tool.