The Character Statistics Thread

[mariofanpm12]

Yea I was going to disinclude specials mainly because... well some characters' specials function so exoticly and some characters would have disadvantages... (ex: mario's FLUDD deals no damage so he has 3 specials to go by... ) and (ex: marth's counter)

 

[mariofanpm12]

but yea i may need you guy's help to confirm if all my data is correct. For example, bowser mains could help me if i get th emax damage of one of bowser's moves wrong.

also, these will be done in versus mode, not training mode, with the hack that turns off stale moves negation... because in versus mode moves are at 105% full strength, yeilding more accurate results

 

[xDD-Master]

Tether Grabs (Grabs disjointed from the character)
1.Olimars Blue Pikmin 10.1
2.Olimars White Pikmin 10.0
3.ZSS 9.75
Olimars Red Pikmin 9.75
5.Olimars Yellow Pikmin 9.6
6.Samus 9.2- 4 (I couldnt really tell so I took the range)
7.Link 8.25
8. Olimars Purple Pikmin 7.5
9. Lucas 7.25
10. Toon Link 7
11. Yoshi 6.75
12. Ivysaur 6.1

Non-tether grabs

13. DDD 6
14.Charzard 5.5
15. Zelda 5.3
16. Donkey Kong 5.1
17. Squirtle 5
18. Marth 4.75
Diddy 4.75
Snake 4.75
21. Kirby 4.5
ROB 4.5
23. Peach 4.25
Fox 4.25
Falco 4.25
Ness 4.25
27. Wario 4.1
28.Bowser 4
Pit 4
MK 4
Jigglypuff 4
Ike 4
Sonic 4
34. Pika 3.9
Sheik 3.9
36. Lucario 3.75
37.Wolf 3.5
ICs 3.5
Mario 3.5
Luigi 3.5
G&W 3.5
42. Captain Falcon 3.25
43. Ganon 2.75

It would be sooooo great if you would add the duration of the grabs to them

And startup lag would also be very nice to know

Then we could devide the lenght through the startup up lag and the total number of frames and we could see who has the best grab (Standing Grab and without Throwdamage / -knockback / Pummel and CGs or Follow-Ups...)

For example:
Olimar blue 10.1
Startup Lag: 10 Frames
Total Frames: 29
So it would be: 10.1 / 10 / 29 x 100 (We should multiply by 100 for better numbers ^^)
-> 3.48

Now we take:
Meta Knight 4
Startup Lag: 5 Frames
Total Frames: 29
So: 4 / 5 / 29 x 100 -> 2.76

One bad example:
Samus 9.3
Startup: 16
Total: 91
We get: 9.3 / (16 x 91) x 100 -> 0.63

Just one another Example:
IC 3.5
Startup: 6
Total Frames: 29
3.5 / 6 / 29 x 100 -> 2.01

==> Olimar > MK > IC > Samus

 

[mariofanpm12]

Once all of this information is compiled, we would technically be able to see who is statistically the best character in the game. lol watch it be Ganon

 

[Bellioes]

Well we already have the durations from AA's list.
@Morpheus, Ok so which characters do you want me test. And for smashs, were fully charging them right?

 

[xDD-Master]

Well we already have the durations from AA's list.

plse Ie neede ae linke ;P

5e's

 

[ndayday]

I believe he mis-spoke, as AA has the thread on the first frame the grabs hit, not the actual frame data for all the grabs. To find the duration, just go scouring over the character boards frame data. It should be fairly easy.

 

[Bellioes]

Oh ok Im sorry
Yeah I misunderstood.

 

[Bitter Romantic]

it would be amazing if you guys did the range for pivot grabs and dash/running grabs as well :O

because some characters may have bad/average range with their regular grabs, while their pivot grab/dash grab range [jiggs and yoshi comes to mind] is ridiculous :O

otherwise, very nice project :3


*also, duration would be nice as well :3

 

[Bellioes]

Yeah I think ND3s working on the dash/pivot grabs. As for duration, I really dont have an accurate way of testing it since things like that require very specific frame data

 

[ndayday]

Well, I could go on a frame data hunt. O.O

I'll PM it to Morpheus since it'd take up way too much space.

 

[smashkng]

Please, put the initial jump height and initial jump speed.

 

[Bellioes]

Well how about I help. You take the first half of the character boards (to like MK) and I take the bottom half; that is what a frame data hunt is, right? Ill just PM the characters I could find the data for and you just put the two lists together.
Sound good?

 

[mariofanpm12]

Right now I'm recoriding the max damage each normal attack for each character does. These are Training Mode values (for those who are unaware, Versus Mode moves are at 105% full strength). All moves are sweetspotted, fully charged for Smashes, and then divided by the number of moves recorded. Ganondorf, for example, has his total strenghth of all his moves divided by 20, whereas Snake has his divided by 22, because Snake has more moves (his jab is three moves while Ganon's is only one).

Anyone mind helping me? All I need is for somebody to take, say, 15 charcaters and record the max damage all their moves do.

By the way, I'm taking in the Ledge attack that's performed when you have less than100%... as those are stronger. And trip attacks are not included.

 

[Ray_Kalm]

Right now I'm recoriding the max damage each normal attack for each character does. These are Training Mode values (for those who are unaware, Versus Mode moves are at 105% full strength). All moves are sweetspotted, fully charged for Smashes, and then divided by the number of moves recorded. Ganondorf, for example, has his total strenghth of all his moves divided by 20, whereas Snake has his divided by 22, because Snake has more moves (his jab is three moves while Ganon's is only one).

Anyone mind helping me? All I need is for somebody to take, say, 15 charcaters and record the max damage all their moves do.

By the way, I'm taking in the Ledge attack that's performed when you have less than100%... as those are stronger. And trip attacks are not included.

Since I have nothing better to do right now, I'll help you.

 

[xDD-Master]

LOL this sounds funny xD

The character who does the most damage in avarage xD

LOL MK could possible be last xD

 

[ndayday]

Well how about I help. You take the first half of the character boards (to like MK) and I take the bottom half; that is what a frame data hunt is, right? Ill just PM the characters I could find the data for and you just put the two lists together.
Sound good?

Sounds good.

I have the data for Link if you want it by the way XD, it's not much but it'll save you just that much more time.

 

[Bellioes]

Ok Ill PM you the list tomorrow without Links data since you already have it XD

 

[ndayday]

That was a good choice of a nickname on my part, I meant to talk to xDD Master, haha. And okay.

I'll try to do some of the pivot data tonight and tomorrow, as well as wrap up jump height and crouch height.

 

[mariofanpm12]

Since I have nothing better to do right now, I'll help you.

well thanks...

as for peach, i planned to exclude special moves altogether.

also, for the f-smash, take the one that deals the most damage was what i had in mind?

 

[mountain_tiger]

well thanks...

as for peach, i planned to exclude special moves altogether.

also, for the f-smash, take the one that deals the most damage was what i had in mind?

Wouldn't it be better to get all of their damages and divide it by 3?

 

[mariofanpm12]

Wouldn't it be better to get all of their damages and divide it by 3?

not a bad idea...

 

[Noa.]

This is a very interesting thread. It's nice for some good tidbits of info.

But there's one thing that I found missing.

In the Grab speed frame data section, Olimar isn't listed.

Why is this? Was he forgotten or what?

 

[mountain_tiger]

Maybe I'm starting to over-complicate things here, but I have another suggestion. You said you were just doing sweetspots, but IMO you should do non-sweetspot hits too. For example, Marth has a sweetspot on every attack except counter, and unless you have inhumanely good spacing, there's no way you'll sweetspot every attack, so wouldn't it be better ot take the average of the sweetspot and the non-sweetspot? Same for other attacks with sweetspots...

 

[Gindler]

Ha, wow is peach's airdodge really that horrible?

 

[Bellioes]

I know I said Id PM it to ND3 but then I thought, why not just post it here. So yeah heres what I could from scourging the character boards on grab durations.

Grab Durations *Incomplete*
Credit: The various character boards
Format: Standing grab duration/Dash grab duration/Pivot grab duration

This is incomplete so Im missing things. If its not there, then that means I couldnt find it. Any help would be appreciated. Thanks. So now, on to the data :

Bowser 39/49/39
CF
Diddy
DK 29/45/39
Falco 29/39/35
Fox
G&W
Ganon 29/37/35
ICs
Ike 30/40/36
Jigglypuff 29/39/35
D3
Kirby
Link
Lucario 33
Lucas
Luigi 29/39/
Mario
Marth 29
MK 29/39/34
Ness 35/35/36
Olimar 29
Peach
Pikachu
Pit 29/39/35
PT
ROB
Samus 91/74/91
Sheik 30/46 (39)/
Snake 29/29/29
Sonic 30/40
TL 84/93/93
Wario 29/39/35
Wolf 13 (According to the Wolf boards 0_o)/33/
Yoshi
Zelda 29/39/35
ZSS

 

[fromundaman]

Ummm... I have testedt this myself and know for a fact that MK's grab range is longer but slower than Kirby's, so the exact opposite of what this OP says.

 

[uhmuzing]

I have to admire the effort of this.

 

[SuSa]

I am almost 99% certain a "getup roll" (eg: After a Ganon side-B or a Snake dthrow) are both different then a normal roll. Because Snake rolls REALLY far in a getup roll, but pretty poorly in a regular roll. May want to test this further. If you want a list of furthest/shortest rolls check the Snake boards for our Dthrow Tech chase topic (it's 1st page) and it has the data at the bottom.

 

[Mota]

Amazing thread. Good info

 

[smashkng]

Ha, wow is peach's airdodge really that horrible?

Yes, it is. Invincibility lasts 10 frames shorter than anyone else making her having only 1/3 of the animation with invincibility (only 15 frames instead of 25 frames) so if she air dodges she is still vulnerable to attacks even if she dodged it.

 

[Bellioes]

Fromundman. the grab durations are in alphabetical order so MK isnt necasarily slower than Kirby just because hes lower on the list. I couldnt even find out how long Kirbys grab lasts. All I did was take info from the different character boards.

 

[MorpheusVGX]

I am sorry I have been missing for a while. I am having a very hard week at work. I will come back soon to help an update the OP. Congratulations for the work you are doing. I think KO power needs to be measured in terms of : 1) how many options do I have for killing (moves) , 2 ) how early does options kill
So to do that we should take a Score based on those moves that can actually kill. I suggest we use sweetspoted non-charged smashes for that. In the end is going to be the same. The proportions won't change. As I said before, we won't be considering precision (sweetspot difficulty) or hitbox size on this. That would be another topic I think. An example of how I think we should do it:

character A has X-move that kills at 120% (uncharged sweetspoted and with the considerations I told you before). The score is 999 - 120 = 879.
So for each killing move he gets a score that is higher in relation to how early it kills. In the end, if he has many kill options (moves) and many kill early, his score will be higher. Moves that cannot kill even in 999% get zero score. But, If you think of it, if you are close to the blast zone, it can KO you. So maybe we could give them 1 point. This is why it is better to test in a small stage so we can even consider that some moves that can hardly kill have better knockback than others.

Maybe we can modify those scores by considering sweetspots or hitbox size, but for now, I think we shouldnt.

I will test on weekend these characters: Peach, C. Falcon, Lucas, Wolf, Pk Trainer, Sheik and Zelda, Bowser. See ya.

 

[mountain_tiger]

I am sorry I have been missing for a while. I am having a very hard week at work. I will come back soon to help an update the OP. Congratulations for the work you are doing. I think KO power needs to be measured in terms of : 1) how many options do I have for killing (moves) , 2 ) how early does options kill
So to do that we should take a Score based on those moves that can actually kill. I suggest we use sweetspoted non-charged smashes for that. In the end is going to be the same. The proportions won't change. As I said before, we won't be considering precision (sweetspot difficulty) or hitbox size on this. That would be another topic I think. An example of how I think we should do it:

character A has X-move that kills at 120% (uncharged sweetspoted and with the considerations I told you before). The score is 999 - 120 = 879.
So for each killing move he gets a score that is higher in relation to how early it kills. In the end, if he has many kill options (moves) and many kill early, his score will be higher. Moves that cannot kill even in 999% get zero score. But, If you think of it, if you are close to the blast zone, it can KO you. So maybe we could give them 1 point. This is why it is better to test in a small stage so we can even consider that some moves that can hardly kill have better knockback than others.

Maybe we can modify those scores by considering sweetspots or hitbox size, but for now, I think we shouldnt.

I will test on weekend these characters: Peach, C. Falcon, Lucas, Wolf, Pk Trainer, Sheik and Zelda, Bowser. See ya.

How do you define what a 'killing move' is though? At what percentage must a move kill below to be considered a proper kill move? Take Samus' DSmash for example. It will kill, but not until 150%. Does that count as a kill move. Likewise, someone's UTilt may kill at 200%. Does that count? The term 'kill moves' on its own is rather ambiguous.

 

[MorpheusVGX]

For this measurement, a kill move is any move that can kill at some percentage. There are moves that cannot kill you even if you are at 999% right?. The score for that move would be zero. A character get points for moves that can kill and by how early they kill.

Also, I was thinking that if everyone is using Mario on the middle of final destination and we can just gather that info to set the score for each character, it would be good.

 

[mountain_tiger]

I see. Are you measuring from the centre of FD?

 

[Dekar173]

Awesome, awesome thread. Good compilation of some pretty important information here

 

[Bellioes]

Heres my suggestion for measuring 'killing ability' for each character:
We use Mario in the middle of FD like you said. We also use the same scoring system (999-120= 879 for a move that kills at 120%).
Heres where my method changes a little; you only count moves that actually have an hitbox and do damage (so Marios FLUDD doesnt count) and divide the total by the number of moves that character actually has. So even though a move does cant kill at 999%, we still add that move when dividing so that the characters killing ability value is lower than someone with the same amount as moves but more killing moves. It would be a very accurate way of testing just how good a character is at 'killing'.
What do you think Morpheus?

 

[MorpheusVGX]

In my scoring system you don't ever divide. I will post a better example now:

Character A has 2 moves that can kill. Move 1 kills at 120%, so he gets 999 - 120 = 879 points.
Move 2 kills at 200% so he gets 999 - 200 = 799 points.
Character A total score is: 879 + 799 = 1678

Character B has 3 moves that can kill. Move 1 kills at 90% so he gets 999 - 90 = 909 points.
Move 2 kills at 400% so he gets 999 - 400 = 599 points for that move.
Move 3 kills at 150% so he gets 999 - 150 = 849 points for that move.
Character B total score is: 909 + 599 + 849 = 2357

So, character B has more killing power than character A.

A move refers to a A or B move. I think we should just consider counters at they minimum value (for those who vary). Any other rare move you think of, let's discuss it. Do you see why it is not necessary or convenient to divide anything? We are simply giving a better score based on killing options and by how early they kill. Do you agree?

 

[mountain_tiger]

I think I understand. Well, now that I get how we're scoring this, I'll happily help you out with that. Now that my exams are more or less over, I have a LOT of free time on my hands, so doing these things doesn't both me in the slightest.

I'll happily test Jigglypuff, Samus, Ganondorf, Wario, Falco, Fox and Lucas for you, and if you want any others tested, then tell me and I should be able to do so.

Edit: Actually, I have one last question. Are we doing smash attacks charged or uncharged?