HomeFAQCalendarSearchMemberlistUsergroupsRegisterLog in
Check out the Drawing Contest going on in the Artwork Center! Deadline extended to April 20th! If you have submitted something, you may resubmit something else on the next deadline!

Share | 
 

 Predator's ranking guide (In progress)

View previous topic View next topic Go down 
Go to page : Previous  1, 2
AuthorMessage
predator852
Gingerbreadian [SPECIAL]


Join date : 2012-10-04
Posts : 1326

PostSubject: Re: Predator's ranking guide (In progress)   Fri Oct 19, 2012 9:08 pm

So I've edited my calculations that I use to get the Total number.
Before: the multipliers were
Land*2
Air*1.5
Sea*1

But for very good reasons this was obviously too flawed for this type of ranking. So now I have taken the time to divide all of the units (still only working with 6* cards right now) into one of three groups based on their stats.
Group A is made up of all units that have one stat that is obviously larger than the other two, thus making it a specialty (such as Vlad or Dhimitros). For those units, the multiplier would be:
Higest stat*1.75
Second Stat*1.25
Lowest Stat*1

Group B is made up of units that have two high stats and one considerably lower stat. (Like Charles or Petra) For these units their multipliers are as follows:
Two highest stats*1.5
Lower stat*1

Group C is made up of units who have balanced stats across the board or who have stats that do not fit into group A or B (I put cards like Dorthea and Abel in here.) For these cards I mulitiplied all the stats by 1.33 to get the total score.

Using this method the rest of my formula would be the same, but the initial total score would be more specified to each units stats, unlike my former approach, wich was simply a metaphoric cast of a net.

I am working on the rest of my formula, specializing the critera to account for things like Events, quests, overall DPS, etc. So that in the end I will have mulitple groups. For example, I would have group AL (a ranking of cards who have a large amount of attack in one area {A} and the attack is focused in its land stat {L} so that players know all of the best cards to use to fight the Black Dragon.

Does any of this sound better??
Back to top Go down
View user profile
timlai1209
Gingerbreadian [SPECIAL]


Join date : 2012-09-10
Posts : 34

PostSubject: Re: Predator's ranking guide (In progress)   Sat Oct 20, 2012 5:36 am

Well, I believe this guide is more precise than struedell's tier list since it's based on calculation, instead of personal analysis.

On the other hands, you can present it better by inserting table.

For the skill multipler, it's optional to be changed and possible to be given a range that based on critical factors.

;)Anyway, good work.
Back to top Go down
View user profile
clocksprocket
Commando Chicklet [EPIC]


Join date : 2012-08-17
Posts : 1395

PostSubject: Re: Predator's ranking guide (In progress)   Sat Oct 20, 2012 10:34 am

timlai1209 wrote:
Well, I believe this guide is more precise than struedell's tier list since it's based on calculation, instead of personal analysis.

Code:
=ROUND([@[Manual Adjust]]*(((SUM([@[max.Land]]*LandValue,[@[max.Air]]*AirValue,[@[max.Sea]]*SeaValue)*INDIRECT("Speed"&[@Speed]&"Value"))*(INDIRECT("Range"&[@Range]&"Value"))*INDIRECT([@Type]&"Value")*INDIRECT([@Skill]&"Value"))),0)

Wink

_________________
KiK: clocksprocket | Play DOT! Referral: 7Px4

forum rules | FAQ (ask questions) | find allies | unit list | want a green name?
Back to top Go down
View user profile
freedomchicken



Join date : 2012-10-10
Posts : 9

PostSubject: Re: Predator's ranking guide (In progress)   Mon Oct 22, 2012 1:24 pm

I did something similar, at the link below. Modded a copy of the google doc found elsewhere in the forums.

I didn't give extra points based on type of unit, as monster defenses typically vary from all three pretty evenly it seems, and giving static points for having a skill or not, as it's very debatable if a skill is better than others or not. It's really more about positioning of units than anything. As well different skills are better for different events.

https://docs.google.com/a/mccluremultimedia.com/spreadsheet/ccc?key=0AgV88hQ6OpDqdEZWOE1MNk9oTklvSUtmN3VCLXUzMmc#gid=0

My formula:
1 point for each 10k in a stat at max
1 point for each range above 130
1 point for each rank in speed
1 point for having a skill
1 point for each 5 cost below 30
Edit: Then multiplying that total by rarity.
Back to top Go down
View user profile
predator852
Gingerbreadian [SPECIAL]


Join date : 2012-10-04
Posts : 1326

PostSubject: Re: Predator's ranking guide (In progress)   Wed Nov 14, 2012 9:11 pm

Everyone else are posting their formulas, you should post this ^^^ in it's own thread, it's great Smile

_________________


Wanna Know What I Do In My Free Time? Click here

Back to top Go down
View user profile
digitalsynner



Join date : 2012-10-29
Posts : 92

PostSubject: Re: Predator's ranking guide (In progress)   Wed Nov 14, 2012 11:38 pm

freedomchicken wrote:
I did something similar, at the link below. Modded a copy of the google doc found elsewhere in the forums.

I didn't give extra points based on type of unit, as monster defenses typically vary from all three pretty evenly it seems, and giving static points for having a skill or not, as it's very debatable if a skill is better than others or not. It's really more about positioning of units than anything. As well different skills are better for different events.

https://docs.google.com/a/mccluremultimedia.com/spreadsheet/ccc?key=0AgV88hQ6OpDqdEZWOE1MNk9oTklvSUtmN3VCLXUzMmc#gid=0

My formula:
1 point for each 10k in a stat at max
1 point for each range above 130
1 point for each rank in speed
1 point for having a skill
1 point for each 5 cost below 30
Edit: Then multiplying that total by rarity.

I like your approach FreedomChicken. There is a certain elegance to a point-based approach as opposed to scaling multipliers and modifiers. I will have to spend some time assessing your list.

Nice work!
Back to top Go down
View user profile
BkWiz
☆☆☆☆☆


Join date : 2012-09-24
Posts : 954

PostSubject: Re: Predator's ranking guide (In progress)   Thu Nov 15, 2012 2:28 am

digitalsynner wrote:
freedomchicken wrote:
I did something similar, at the link below. Modded a copy of the google doc found elsewhere in the forums.

I didn't give extra points based on type of unit, as monster defenses typically vary from all three pretty evenly it seems, and giving static points for having a skill or not, as it's very debatable if a skill is better than others or not. It's really more about positioning of units than anything. As well different skills are better for different events.

https://docs.google.com/a/mccluremultimedia.com/spreadsheet/ccc?key=0AgV88hQ6OpDqdEZWOE1MNk9oTklvSUtmN3VCLXUzMmc#gid=0

My formula:
1 point for each 10k in a stat at max
1 point for each range above 130
1 point for each rank in speed
1 point for having a skill
1 point for each 5 cost below 30
Edit: Then multiplying that total by rarity.

I like your approach FreedomChicken. There is a certain elegance to a point-based approach as opposed to scaling multipliers and modifiers. I will have to spend some time assessing your list.

Nice work!

There is a really simple way to assess the usefulness of units which no one has really touched upon as of yet. Suspect

It's incredibly straightforward and almost completely free of bias if you factor out the unit's skills...
Back to top Go down
View user profile
Wheatmidge



Join date : 2012-10-01
Posts : 106

PostSubject: Re: Predator's ranking guide (In progress)   Thu Nov 15, 2012 3:58 pm

BkWiz wrote:


There is a really simple way to assess the usefulness of units which no one has really touched upon as of yet. Suspect

It's incredibly straightforward and almost completely free of bias if you factor out the unit's skills...

I have no idea what you are talking about. Are you saying that skills are the only thing that affects biases?
Back to top Go down
View user profile
predator852
Gingerbreadian [SPECIAL]


Join date : 2012-10-04
Posts : 1326

PostSubject: Re: Predator's ranking guide (In progress)   Thu Nov 15, 2012 4:59 pm

I think he's saying that freedomxhikens approach dosnt favor any one type of unit over another except for units that have skills

_________________


Wanna Know What I Do In My Free Time? Click here

Back to top Go down
View user profile
BkWiz
☆☆☆☆☆


Join date : 2012-09-24
Posts : 954

PostSubject: Re: Predator's ranking guide (In progress)   Thu Nov 15, 2012 8:29 pm

predator852 wrote:
I think he's saying that freedomxhikens approach dosnt favor any one type of unit over another except for units that have skills

Nope. I'm saying that all the threads regarding unit tier rate atm are biased towards the author's views whatever they may be.

Atm there is a very simple way of calculating unit effectiveness that would hold true for even events for the most part, but almost every person so far has yet to use it.
Back to top Go down
View user profile
Dannupa



Join date : 2012-09-18
Posts : 60

PostSubject: Re: Predator's ranking guide (In progress)   Thu Nov 15, 2012 9:28 pm

Bk, don't be so coy, enlighten us.

Personally, I tend to favor units that have high values for any given attack type. Almost all events and battles really require just one thing at a time. A portfolio of things that excel at one thing seems best.

Now, I am far from 6* units and would probably go with one of these tier systems to evaluate there. I'm not likely to get multiple units to use in different scenarios so I'd want a balanced 6* that is better than 5*s all around.
Back to top Go down
View user profile
Wheatmidge



Join date : 2012-10-01
Posts : 106

PostSubject: Re: Predator's ranking guide (In progress)   Thu Nov 15, 2012 11:37 pm

BkWiz wrote:
predator852 wrote:
I think he's saying that freedomxhikens approach dosnt favor any one type of unit over another except for units that have skills

Nope. I'm saying that all the threads regarding unit tier rate atm are biased towards the author's views whatever they may be.

Atm there is a very simple way of calculating unit effectiveness that would hold true for even events for the most part, but almost every person so far has yet to use it.

There are way too many variables for a "simple way of calculating unit effectiveness." I think you are just being facetious. If you aren't; could you please share with us so we could all be a little more enlightened.
Back to top Go down
View user profile
cmud2007
☆☆


Join date : 2012-10-22
Posts : 248

PostSubject: Re: Predator's ranking guide (In progress)   Fri Nov 16, 2012 12:10 am

Using the speed letter directly would be misleading!
Please check this topic
Back to top Go down
View user profile
snappycow
Gingerbreadian [SPECIAL]


Join date : 2012-10-07
Posts : 159

PostSubject: Re: Predator's ranking guide (In progress)   Fri Nov 16, 2012 2:21 am

BkWiz wrote:
predator852 wrote:
I think he's saying that freedomxhikens approach dosnt favor any one type of unit over another except for units that have skills

Nope. I'm saying that all the threads regarding unit tier rate atm are biased towards the author's views whatever they may be.

Atm there is a very simple way of calculating unit effectiveness that would hold true for even events for the most part, but almost every person so far has yet to use it.

You mean checking the event leaderboards and see which is the most common 6* the top 10 are using? Very Happy

edit: in all seriousness though, if we are talking strictly about unit effectiveness with no overall ranking, then perhaps calculating the unit's DPS against different types and different defenses in a matrix would show where the unit would excel / flop.
Back to top Go down
View user profile
dnonamus
War Hero [Rare]


Join date : 2012-10-10
Posts : 89

PostSubject: Re: Predator's ranking guide (In progress)   Fri Nov 16, 2012 7:39 pm

I have a few... questions, first off, I still don't get how the rarity of the unit (or it's commonality) adds to it's usefulness. regardless of the odds of pulling a card what the player pulls is what they have.

Second, why did you use the boss ratios you used? The ratio I used in my spreadsheet (Quest monsters 74% ground, 20% Air 10% sea Bosses 61% Ground, 26% air 13% sea) are based on the number of each in the game, why would any other ratios be used?

Third, after running tests I found that S speed units were approximately 2x as fast as D speed. I used D speed as 1 because more units in the game are D speed than any other (I used the same reasoning for giving 150 range 0 bonus since more units in the game have 150 range than any other) I used 2 for S then spread my speed bonuses evenly between the two and gave a penalty for E speed. I feel this is the most accurate and least subjective way to calculate the bonus.

Lastly, seriously, how could you not have a range bonus?
Back to top Go down
View user profile http://www.FantasicaHelp.com
dnonamus
War Hero [Rare]


Join date : 2012-10-10
Posts : 89

PostSubject: Re: Predator's ranking guide (In progress)   Fri Nov 16, 2012 7:56 pm

freedomchicken wrote:
I did something similar, at the link below. Modded a copy of the google doc found elsewhere in the forums.

I didn't give extra points based on type of unit, as monster defenses typically vary from all three pretty evenly it seems, and giving static points for having a skill or not, as it's very debatable if a skill is better than others or not. It's really more about positioning of units than anything. As well different skills are better for different events.

https://docs.google.com/a/mccluremultimedia.com/spreadsheet/ccc?key=0AgV88hQ6OpDqdEZWOE1MNk9oTklvSUtmN3VCLXUzMmc#gid=0

My formula:
1 point for each 10k in a stat at max
1 point for each range above 130
1 point for each rank in speed
1 point for having a skill
1 point for each 5 cost below 30
Edit: Then multiplying that total by rarity.

I felt that a simple ranking is somewhat misleading. Nina is in no way the best unit to use against a ground boss like Black dragon, nor is she the best to use against an air boss like Faith (second best, yea). A through list like the one I created felt more complete. A player can choose his unit based on it's strength in a particular area which can allow a player to shore up weaknesses.
Back to top Go down
View user profile http://www.FantasicaHelp.com
BkWiz
☆☆☆☆☆


Join date : 2012-09-24
Posts : 954

PostSubject: Re: Predator's ranking guide (In progress)   Fri Nov 16, 2012 8:17 pm

snappycow wrote:
BkWiz wrote:
predator852 wrote:
I think he's saying that freedomxhikens approach dosnt favor any one type of unit over another except for units that have skills

Nope. I'm saying that all the threads regarding unit tier rate atm are biased towards the author's views whatever they may be.

Atm there is a very simple way of calculating unit effectiveness that would hold true for even events for the most part, but almost every person so far has yet to use it.

You mean checking the event leaderboards and see which is the most common 6* the top 10 are using? Very Happy

edit: in all seriousness though, if we are talking strictly about unit effectiveness with no overall ranking, then perhaps calculating the unit's DPS against different types and different defenses in a matrix would show where the unit would excel / flop.

Ding ding ding. We have a winner.

Sometimes I feel like explaining everything in detail, sometimes I don't. It depends on my mood.

On a related side note, I usually leader Charles for events. Does this mean Charles is now a top 6*? Smile
Back to top Go down
View user profile
BkWiz
☆☆☆☆☆


Join date : 2012-09-24
Posts : 954

PostSubject: Re: Predator's ranking guide (In progress)   Fri Nov 16, 2012 8:38 pm

dnonamus wrote:
I have a few... questions, first off, I still don't get how the rarity of the unit (or it's commonality) adds to it's usefulness. regardless of the odds of pulling a card what the player pulls is what they have.

Second, why did you use the boss ratios you used? The ratio I used in my spreadsheet (Quest monsters 74% ground, 20% Air 10% sea Bosses 61% Ground, 26% air 13% sea) are based on the number of each in the game, why would any other ratios be used?

Third, after running tests I found that S speed units were approximately 2x as fast as D speed. I used D speed as 1 because more units in the game are D speed than any other (I used the same reasoning for giving 150 range 0 bonus since more units in the game have 150 range than any other) I used 2 for S then spread my speed bonuses evenly between the two and gave a penalty for E speed. I feel this is the most accurate and least subjective way to calculate the bonus.

Lastly, seriously, how could you not have a range bonus?

All the questions raised here are very good ones. They all highlight the 'subjective' flaws in this kind of data gathering/calculations.

dnonamus you do have a few flaws in this post that are easily corrected. For some D units, S is 2x the speed of D, for other D units it is 2.2x the speed of that D unit speed, OR in other cases only 1.5 times the speed of a D unit speed. (In case that was confusing, there is a LARGE range of speeds in D speed due to animation, missile speed, and the internal numeric number for the unit's speed which all combine to give you the actual attack speed of the unit).

Please see the speed thread in order to fix these figures. Though unfortunately I have already removed most of my own findings and you can only find pieces of it here and there...

And while the quest monster ratio is a great one to look at (Great and easy to test way to figure out land v. sea v. air ratings) considering that the majority of the time you will have no need to calculate these sort of things (Mina's Trials is the only time it becomes somewhat challenging), you're probably better off disregarding it (It's a changing %) and instead focusing on boss /arena DPS ratings since those ratings do not change every time a new unit or quest line is introduced and are pretty stable.
Back to top Go down
View user profile
dnonamus
War Hero [Rare]


Join date : 2012-10-10
Posts : 89

PostSubject: Re: Predator's ranking guide (In progress)   Sat Nov 17, 2012 11:23 am

BkWiz wrote:
dnonamus wrote:
I have a few... questions, first off, I still don't get how the rarity of the unit (or it's commonality) adds to it's usefulness. regardless of the odds of pulling a card what the player pulls is what they have.

Second, why did you use the boss ratios you used? The ratio I used in my spreadsheet (Quest monsters 74% ground, 20% Air 10% sea Bosses 61% Ground, 26% air 13% sea) are based on the number of each in the game, why would any other ratios be used?

Third, after running tests I found that S speed units were approximately 2x as fast as D speed. I used D speed as 1 because more units in the game are D speed than any other (I used the same reasoning for giving 150 range 0 bonus since more units in the game have 150 range than any other) I used 2 for S then spread my speed bonuses evenly between the two and gave a penalty for E speed. I feel this is the most accurate and least subjective way to calculate the bonus.

Lastly, seriously, how could you not have a range bonus?

All the questions raised here are very good ones. They all highlight the 'subjective' flaws in this kind of data gathering/calculations.

dnonamus you do have a few flaws in this post that are easily corrected. For some D units, S is 2x the speed of D, for other D units it is 2.2x the speed of that D unit speed, OR in other cases only 1.5 times the speed of a D unit speed. (In case that was confusing, there is a LARGE range of speeds in D speed due to animation, missile speed, and the internal numeric number for the unit's speed which all combine to give you the actual attack speed of the unit).

Please see the speed thread in order to fix these figures. Though unfortunately I have already removed most of my own findings and you can only find pieces of it here and there...

And while the quest monster ratio is a great one to look at (Great and easy to test way to figure out land v. sea v. air ratings) considering that the majority of the time you will have no need to calculate these sort of things (Mina's Trials is the only time it becomes somewhat challenging), you're probably better off disregarding it (It's a changing %) and instead focusing on boss /arena DPS ratings since those ratings do not change every time a new unit or quest line is introduced and are pretty stable.

I agree, the speed flaw is a big one. Generally lists like these exist for the purpose of identifying what units to pursue. The flaw in the data comes from the inability to run actual DPS tests because no one has all the units, and the fact that no one has the units is the whole reason for the lists. Because I wanted a realistic list that would give me an edge (I don't care about units perceived value unless that value is based solely on the units ability to defeat monsters and bosses efficiently in the game) I made sure my data was as accurate as it possibly could be given what data I did have. I have run countless tests to validate that data and the only flaw in my system is the inability to accurately calculate speed data for it. I however am comfortable with that flaw since the letter grades do an ok job of narrowing down this variable.
Back to top Go down
View user profile http://www.FantasicaHelp.com
predator852
Gingerbreadian [SPECIAL]


Join date : 2012-10-04
Posts : 1326

PostSubject: Re: Predator's ranking guide (In progress)   Sat Nov 17, 2012 12:52 pm

BkWiz wrote:
dnonamus wrote:
I have a few... questions, first off, I still don't get how the rarity of the unit (or it's commonality) adds to it's usefulness. regardless of the odds of pulling a card what the player pulls is what they have.

Second, why did you use the boss ratios you used? The ratio I used in my spreadsheet (Quest monsters 74% ground, 20% Air 10% sea Bosses 61% Ground, 26% air 13% sea) are based on the number of each in the game, why would any other ratios be used?

Third, after running tests I found that S speed units were approximately 2x as fast as D speed. I used D speed as 1 because more units in the game are D speed than any other (I used the same reasoning for giving 150 range 0 bonus since more units in the game have 150 range than any other) I used 2 for S then spread my speed bonuses evenly between the two and gave a penalty for E speed. I feel this is the most accurate and least subjective way to calculate the bonus.

Lastly, seriously, how could you not have a range bonus?

All the questions raised here are very good ones. They all highlight the 'subjective' flaws in this kind of data gathering/calculations.

dnonamus you do have a few flaws in this post that are easily corrected. For some D units, S is 2x the speed of D, for other D units it is 2.2x the speed of that D unit speed, OR in other cases only 1.5 times the speed of a D unit speed. (In case that was confusing, there is a LARGE range of speeds in D speed due to animation, missile speed, and the internal numeric number for the unit's speed which all combine to give you the actual attack speed of the unit).

Please see the speed thread in order to fix these figures. Though unfortunately I have already removed most of my own findings and you can only find pieces of it here and there...

And while the quest monster ratio is a great one to look at (Great and easy to test way to figure out land v. sea v. air ratings) considering that the majority of the time you will have no need to calculate these sort of things (Mina's Trials is the only time it becomes somewhat challenging), you're probably better off disregarding it (It's a changing %) and instead focusing on boss /arena DPS ratings since those ratings do not change every time a new unit or quest line is introduced and are pretty stable.

Let me clarify a couple of things here.

A) I got to tweaking this general outline to what I thought my final ranking guide would come out to look like and ran into things that I thought just made it impossible to complete. Those things are:

1) The speed thing. I really dont have the time or resources to find the true speed for every unit in the game. I decided until a comprenhensive explanation of their actuall speeds was attained, this idea of a ranking guide would be flawed in at least one way.

2) Range. Honestly, IMO, range makes little to no difference, as I stated before. Against bosses:not accounted for. In colleseum:not accounted for. At the front of a quest:only partally effective. At the end of a quest:only partially accounted for. If I have a great unit with a range of 130, I'll just slap some KB and slow units in his general vicinity and he/she can hack away at the monster for a huge amount of time. Range has not ever really been a factor that makes or breaks a card.

3)Although hard numbers would tell me that because there are ~six sea enemies and over 20 land enemies, land is at least three times as nessasary as sea, skewing numbers to show that creates a very biased spreadsheet. Also, if I must be honest, I personally NEVER look at the sea attack. As far as I'm concerned if I have 4-6 units that are great at destroying land and air enemies, and none of their sea attack is 0, I have nothing to worry about. So far I have been correct in this very vauge assumption =)

B) When I recommented on this thread I was just saying that I think Freedomchicken was onto something new here, and that he should post it. Personally, I think Dnonamus made a great, very intuitive spreadsheet that I've referenced almost everyday since he posted it. I often look at Clocks and Strusdells list when trying to decide on which card to use or trade. And although BK's idea is a little...weird...I do sometimes find myself doing that as well (just because I'm nosey)

_________________


Wanna Know What I Do In My Free Time? Click here

Back to top Go down
View user profile
Sponsored content




PostSubject: Re: Predator's ranking guide (In progress)   Today at 5:10 am

Back to top Go down
 
Predator's ranking guide (In progress)
View previous topic View next topic Back to top 
Page 2 of 2Go to page : Previous  1, 2
 Similar topics
-
» Kingdoms in mai otome [Guide]
» Millers Collectables price guide 2009 By Judith Miller & Mark Hill
» halogen oven user guide
» Guide to Blackwings
» 57 Chevy Restoration Guide

Permissions in this forum:You cannot reply to topics in this forum
 :: Fantasica :: Game Discussion-
Jump to: