Ranking VF moves

I'm tossing this in general so that new ppl don't confuse it with something useful or instructional. It's just for fun. The goal is to come up with a numerical rating for each VF move. Once we can nail down a formula, it'd be easy to put it into a spreadsheet and get a number value for any move in the game just by plugging in some stats.

Categories -

Exe
Rec (block)
Rec (whiff)?
Damage (potential)
Frame adv.
Circularity
Hit level
Range
Priority?

I think the best way to go about this is to assign each category a weight, i.e.
recovery is more important than hit level. Then assign a point value to each based on
arbitary tiers. Multiply the point value by weight to get the move's rating in
a particular field and then add all those together to get the move value.

exe - pretty self explanatory, faster is better. No bonus for 10 frame moves.
11 frames = 6.5 pts.
12 frames or less = 6 pts.
13-14 frames = 5 pts.
15 frames = 4 pts.
16 frames = 3 pts.
17 frames = 2 pts.
18+ frames = 1 pt.

rec (block) - harder to do. It's pretty evenly spaced but moves that give better than
plus 2 are considered pretty valuable, and moves that are -8 drop a full notch from moves
that are -7 since we're now in the guaranteed throw range.

+3 or more = 9 pts
+2 = 8 pts
+1 = 7.5 pts
0 = 7 pts
-1 = 6.5 pts
-2 = 6 pts
-3/4 = 5.5 pts
-5/6/7 = 5 pts
-8/9/10 = 4 pts
-11 = 3 pts
-12 = 2.5 pts
-13 = 2 pts
-14 = 1.5 pts
-15/16 = 1 pt
-17+ = .5 pts

rec(whiff) - should it be there? It's hard to NOT include it. Moves like jeffry's BAB are
always + on block, but that doesn't tell the whole story. I can dodge and knee something
like jeff's f,f+K easily, whereas dodging and kneeing a -4 elbow is hard/impossible.
I don't know the stats for whiff off the top of my head, I don't even have a sense for how
many frames are typical for eg an elbow. I'd have to rely on only one source, 1/60th, for
whiff data.

Damage potential - also flaky. First, I'm assuming that I will rate a move's damage by the
amount of damage that is guaranteed afterwards if it MC's. If a move gives different
ideal combos based on stance and weight and other factors, damage is averaged from all
potential followups - i.e. if lau's bf+P combo does 55 in one stance and 65 in another, I
would rate it at 60 points.

I need to come up with different values for moves that give increased damage potential on
normal hit vs MC only vs mC. I'm tempted to say a move's final damage potential is its
damage potential on MC + 20 percent of minor counter potential + 10 percent of normal damage
potential? OR maybe add a fixed bonus for moves that work on mC and a slightly larger fixed
bonus for moves that work on normal - for example +1 pt for mC and +2 pts for normal.

100+ pts = 8 pts
85~100 = 7 pts
75~85 = 6 pts
65~75 = 5 pts
50~65 = 4 pts
40~50 = 3 pts
25~40 = 2 pts
< 25 = 1 pt

Frame adv. - straightforward, if a move staggers I figure out what the frame advantage is
if the opponent struggles well. I don't need to worry about what kind of stagger
animation it cuases. I do need to figure out something for moves that always stagger vs
those that only do it vs crouchers or on MC. Maybe I should simply average the frame
advantage vs standing opponents and the frame advantage vs crouchers and call it good.
I'm wondering if guaranteed throw/low throw attacks should be bonus'ed in this section or as part of their damage potential section.

knockdown = 8 pts
+10 and up = 7 pts
+6~10 = 6 pts
+5 = 5 pts
+4 = 4 pts
+3 = 3 pts
+2 = 2 pts
+1 = 1 pt

Circularity -
Linear = 0
Half circular = 1 pt
Full circular = 3 pts

Hit level - I believe good low attacks are rare and valuable in this game, they help to
defeat fuzzy guarding and therefore get a small bonus. High attacks are, for obvious
reasons, not as good as mid.

High = 1
Special High = 1.5
Special Mid = 1.5 (i.e. shun's chouwan)
Mid = 2
Special low = 2
Low = 2.5

Range - Pretty straightforward.
Sidekick+ = 5
Sidekick = 4
elbow+ = 3 (this is between elbow and sidekick range, i.e. beat knuckle)
elbow = 2
< elbow = 1 (less than elbow range)

Priority - ?Maybe add an arbitrary number if a move has weird properties that don't show up
in the frames - for example kage's df+P ducking under punches, akira's low backfist cutting
through elbows sometimes, shun di's low kick beating nearly anything.
======

Now the hard part, weighting each category. I also wonder if ALL attacks shouldn't be put
onto a 5 pt or 10 point scale, it'll screw up the ratings maybe if circularity is deemed
important but only goes up to 3, while damage potential goes as high as 8.

Here's how a typical rating might be added up.

akira SDE:
Exe - 5 pts * exe_weight (let's say 10) = 50
Rec (block - 5 pts * rec_weight (let's say 10) = 50
Rec whiff = let's assume 4 pts * rec_weight (let's say 5) = 20
Damage potential = 5 on MC, but also 3 on normal/MC. Let's call it 6 * dmg_weight(9) = 54
Frame adv = 8 * adv_weight (let's say 7) = 56
Circularity = 0
Hit level = 2 * hit_level_weight (let's say 3) = 6
Range = 4 * range_weight (let's say 5) = 20
Priority = 0 (afaict it doesn't have special priority, just speed).
-------
50+50+20+63+56+0+6+20+0 = 256 rating.

Using identical numbers for "weight" and a little subjective stuff, lau's df+P+K is
10+15+10+63+56+0+6+20+10 = 190 rating.

Some opinions I'd like to solicit -
On a scale of 1-10, how would you rate each category's importance in overall gameplay?
Are there categories I don't have but should?
How can I put each category on a 1-10 scale? Give hard numbers. I want to avoid fractions.
What about a move's interaction with walls or pushback effect or evasive properties?
Lastly, is a character's value really the sum of ALL his move moves or

 

crap
you lost me on the first sentence(well.. more like second paragraph) /versus/images/graemlins/tongue.gif

 

[ QUOTE ]
Myke said:

Did you really have to post a reply like that?

OK, so the content was long, and maybe you don't have the attention span, or patience, or whatever, to sit through and digest the information -- FAIR ENOUGH! But, you don't have to let everyone know that you were lost on the second sentence/paragraph. Really, we did not need to know.

If a thread doesn't interest you, or you find it difficult to relate to, then just let it go.

Sounds easy in theory.....

[/ QUOTE ]

I wasn't displaying my boredom of the subject. I guess it could be looked at it that way, but I was "literally" lost on frames and all that technical shit.
I am actually getting more interested on frame stuff on each characters' move now that I am mastering evade throw escapes, but I was just saying I was lost /versus/images/graemlins/tongue.gif and maybe there could be extra explanations for the noobs?
sorry if I displeased you with my wording, Myke

 

I think you're attempting an insurmountable task, but I applaud you for trying!


A couple thoughts that may save you time down the road:

- Ratings should be indexed on the same scale so that when you assign weightings in categories it is clear which category is more important than others, and by how much. So, for example, you might choose 1-10 as the scale for all categories, such that when you decide recovery is 1.5 times more important than frame execution, it would be a simple 3:2 ratio.

- You may have to also unitize categories to the number of moves per character. For example, if you are going to reward points in the damage category (with no negative numbers), the character with the most moves (though actually many of those moves are low damage) may simply end up with the most points. To compensate, you could have number of moves as a category.


Actually, I think you need to take one step back. Before thinking about points and weightings, you should first decide what the categories are. A couple unobvious ones off the top of my head:

- Some attacks allow the character to avoid certain things (by ducking, swaying, dodging, hopping, or perhaps even sabakis fall under this category)...this feels like an important category to me.

- I also argue that having one move with multiple advantages (e.g. circular, middle, fast, etc.) is better than having multiple moves with specific advantages. Makes the character easier to use.

 

Throw in a few ideas:
- What about move properties: stagger on guard, crumble, sabaki, hit-throw etc ?. Rating the value is a little difficult I think the simplest way would be for something like a sabaki is give it an average rating based on how quickly the sabaki is executed and the amount of moves it can sabaki, done in the same way as the sabaki, reversal, inashi chart. So the ratings would be something like:
Timing -
A - Instant - 10pts
B - Throw Counterable - 8 pts
C - Large Disadvantage - 6 pts
D - Small Disadvantage - 4 pts
E - Even - 2 pts
F - Advantage - 1 pts

and attack classes that the move will successfully sabaki would be done in that same sort of way. The problem would be that for sabakis (out of every sabaki in the game) there are a maximum of 8 attack classes per sabaki which have to be divided between 10 pts. Maybe just spread it out so it reads:
Amount of attack classes that attack can successfully sabaki:
1 - 1pts
2 - 2pts
3 - 3pts
4 - 4pts
5 - 5pts
6 - 7pts
7 - 9pts
8 - 10pts

Ordered roughly in those groupings to give the best representation of how useful it is to sabaki 1 attack class compared to 7 classes etc.

Take a move such as Lion's meteor punch (d,db,b+P) which is in the "A - Instant" catagory making that 10pts and sabakis 6 attack classes giving it 7pts so average that at (roughly) 8pts and that would be your sabaki rating.

The other question would be whether an attack having sabaki properties would grant it another 1-10 pts for its final sum. Maybe bunch sabakis together with other hit properties to make another 10pt average but then again the more hit properties you use in the final sum the more representative you make it so *shrug*.

[ QUOTE ]
Now the hard part, weighting each category. I also wonder if ALL attacks shouldn't be put
onto a 5 pt or 10 point scale, it'll screw up the ratings maybe if circularity is deemed
important but only goes up to 3, while damage potential goes as high as 8.

[/ QUOTE ]

I agree with the idea that they should be all put on a 10pt scale (assuming that they all have the same worth of course). A good example as far as circular vs. damage potential goes is Shun's b+P. It's very useful as it's fully circular but its damage potential is nil.

[ QUOTE ]
How can I put each category on a 1-10 scale? Give hard numbers. I want to avoid fractions.

[/ QUOTE ]

I'll have a crack at the ones I can - for all cases I've automatically put the "best" option as 10 and the "worst" as 1 and worked from there. For the catagories that don't have enough options to have all the pts from 1to 10 ( like the execution one) I've just put my own assesment:

exe - pretty self explanatory, faster is better. No bonus for 10 frame moves.
11 frames = 10 pts.
12 frames or less = 9 pts.
13-14 frames = 8 pts.
15 frames = 6 pts.
16 frames = 4 pts.
17 frames = 2 pts.
18+ frames = 1 pt.

rec (block) -
+3 and +1 (or more) = 10 pts [+1 to +3 is only 2 frames. Even all things being equal doesn't make much diff. in my opinion]
-0 = 9pts
-1/2 = 8 pts
-3/4 =7 pts
-5/6/7 = 6 pts
-8 = 5 pts [8 by itself as it's throw counterable]
-9/10/11 = 4 pts
-12 = 3 pts [punch counterable]
-13 to -14 = 2 pts
-15/16/17 = 1 pts

Damage potential -


100+ pts = 10 pts [I think the damage increments are right - don't know of a way to get all 1-10pt values without making the increments smaller]
85~100 = 9 pts
75~85 = 8 pts
65~75 = 6 pts
50~65 = 5 pts
40~50 = 3 pts
25~40 = 2 pts
< 25 = 1 pt

Circularity - [simple but meh. All moves get 1 auto and another 6 for half circ. and another 10 for full circ.]
Linear = 1
Half circular = 7 pt
Full circular = 10 pts

hit level - [agree that a good low attack will be best (10). In reality most low attacks will be slow but the advantage of the attack being low gives it its usefullness.]
High = 1
Special High = 3
Special Mid = 5 (i.e. shun's chouwan)
Mid = 7
Special low = 5
Low = 10

Range -
Sidekick+ = 10
Sidekick = 8
elbow+ = 5 (this is between elbow and sidekick range, i.e. beat knuckle)
elbow = 3
< elbow = 1 (less than elbow range)

[ QUOTE ]
What about a move's interaction with walls or pushback effect or evasive properties?

[/ QUOTE ]
I think it should be counted. A simple rating of how much 'knockback' the attack gives on hit (half the score if it only works on MC or vice versa) should do. Maybe also a modifier if the attack does more than 21 damage (maybe +2 to the score). I don't know if there is really a true way to measure 'knockback' but maybe you could just do it for each individual case and give an estimate. The wall advantage might be more relevent for throws though.

That's all for now.

 

I see circularity as VERY important, and should be weighted that way. As an example, take jacky's b,f+K+G...it's not that fast, has crappy recovery and not that much damage potential, so according to Kreed's original rating system it would receive a rather low rating. But because of being full circular and only because of that jacky's b,f+K+G is a good move. Again consider this...two b,f+K+G's one being full circular and one being linear with all other properties being the same...wouldn't you say the full circular version is 1000 times better? What I'm saying is, determine the rating of a move disregarding the circularity and then multiply the move's whole rating by say 3 times if the move is full circular and 2 times if it's half cicular...so say b,f+K+G gets a rating of 80 then you would multiply by 3 since it's full circular to get 240...of course i took 2X and 3X out of my ass, but i just feel that a full circular move is AT LEAST 3X better than the same non-circular version.

 

re: full circularity - under drunkencat's revised system this is addressed, a fully circular move is worth 10x more than a linear one.

Drunkencat: nice work on the scaling. I liked it. A quick correction about shun's b+P - I think a major counter guarantees stuff like df+P,P,K. So it's not low as far as damage potential goes. It just doesn't guarantee much if it i.e. catches a dodger for a normal hit. It does get bonus pts for frame advantage though. Also I think it may be dodgeable to shun's back but I might be on crack.

 

- Ratings should be indexed on the same scale

Agreed, and 10 points seems like the scale to use.

You may have to also unitize categories to the number of moves per character

If we're to look at this as an individual move rating system, I don't need to, but if I want to use it to rank characters, maybe the best solution is to simply average the value of all their moves and the char. with the highest average wins. That way vanessa isn't going to come out on top. I honestly think this could produce realistic results - it's always been my opinion that akira and lau get very few useless moves, and that the ones they do get are only useless relative to their other attacks. i.e. pai would kill to get some of akira's more obscure, seldom used moves.

Some attacks allow the character to avoid certain things (by ducking, swaying, dodging, hopping, or perhaps even sabakis fall under this category)...this feels like an important category to me.

This will go into the priority category I think - I will treat a move that reliably ducks under high punches the same as I'd treat a move that sabaki's them. I have no idea what I'd do with e.g. shun's low kick except wing it.

- I also argue that having one move with multiple advantages (e.g. circular, middle, fast, etc.) is better than having multiple moves with specific advantages. Makes the character easier to use.

I -think- this will work itself out by the system I'm using, whereby a move's circularity score, speed score, and hit level score are simply added together along with all the other categories. I agree that a killer app can make all the difference. If akira's SDE were full circular, crumpled on normal hit, executed 2 frames faster, etc... the score would reflect that it was better than any other move in the game, but it wouldn't reflect just how gamebreaking such changes would be.

 

Revision 1, ranking VF moves.

Clarifications -
all staggers will be put in the "advantage" category.
special hit effects like slam/crumple/flop are just another form of combo starter, and therefore will go under damage potential. In other words wolf's b+P isn't treated as a 20 point attack with a special 30 point "slams and sets up a combo" bonus, it's just treated as a 56 point attack since that's what you can reliably get every time it hits.
Sabakis are going under the catchall "priority" category.
Some stuff doesn't fit under priority and is at the end, awaiting suggestions.
-----
Categories -

Exe
Rec (block)
Damage (potential)
Frame advantage
Circularity
Hit level
Range
Priority
Bonus properties

EXECUTION - There's only a small difference in points for moves in the 11/12/13/14 frame range, because after a guarded jab you are typically +1 and can get away with any of
these attacks if the opponent hesitates even a tiny bit on a 12 frame jab or d+P.

The gap between the 14 frames group and 15 frames group is larger because many moves
recover at exactly -13 or -14, allowing for minor counters, especially useful dodge attacks.
Comparatively fewer recover at -15 or worse, allowing for minor counters with slow attacks.
For example, lau's basic movelist shows 11 moves at -13/14 and only 3 at -15/16.

The decrease in value for each successive drop in frames is large, as it becomes relatively
harder and harder to stuff quick moves like d+P with 15 and 16 and 17 frame moves.

It might seem unfair to get 0 pts for 18+ frame attacks when you know some slow attacks like
wolf's short shoulder still are useful, but think of execution in terms of a "speed bonus".
A knee is given a small speed bonus because a +5 situation isn't that uncommon or difficult
to create - a low punch MC is +5. Creating +6~8 situations where you can use an 18~20 frame
move to stuff followups is more difficult - for example lau can create a +8 situation with
a jab MC, but it's still not enough to guarantee that a df+P+K will stuff the opponent's
low punch because df+P+K is 21 frames. Therefore, having an 18+ frame move doesn't really
deserve a bonus for speed.


11 frames = 10
12 frames = 9 pts.
13~14 frames = 8 pts.
15 frames = 6 pts.
16 frames = 4 pts.
17 frames = 2 pts.
18+ frames = 0 pts.

RECOVERY - I don't like this exactly but I think it's the best I can do on a ten
point scale with no fractions. +2 is significant compared to +1 because when you create
+2, you are instantly at an elbow-or-throw guessing game. +2 is the point where fuzzy
guard can no longer avoid attack or throw (motion cut can, but it's pretty hard).
On the other hand, +1 is relatively common (guarded jab) and so is classed differently.

The rest is a pretty standard scale. At first I wanted to differentiate between -15,16,
and 17 moves. But, thinking about it, kage gets a 15 frame upper minor counter for OK
damage, jeffry gets the double upper, shun gets chouwan, jacky gets P+K,P for -16, etc.
Knee counterable is not so special that it needs a separate entry anymore, as hardly anyone
gets a 17 frame knee that combos on minor counter now. With the removal of the knee, many
no-counter combo starters are higher than 17 frames.

+2 or more = 10 pts
+1 = 9 pts
-0 = 8pts
-1~2 = 7 pts
-3~4 = 6 pts
-5~7 = 5 pts
-8 = 4 pts [8 by itself as it's throw counterable]
-9~11 = 3 pts
-12 = 2 pts [punch counterable]
-13~14 = 1 pts
-15+ = 0 pts

DAMAGE - Stuff that causes over 85 pts is rare and therefore tops. 75~85 is only
slightly lower because you still need only three of them to kill the opponent. 65~75 is also
just a little lower for the same reason. 50~65 is considered much less valuable
because you definitely won't be killing the opponent with three moves of this type.

40~50 is close to 50~65 in that they're all in typical throw damage range and all need
at least four and probably 5 to kill an opponent.
Below 40 moves get a major drop because you can't kill your opponent with even 5 of 'em. 25~40 pts isn't that scary and is typical damage for MC from just about any non-jab
attack, e.g. an elbow MC or P+G throw.

Less than that nets only 1 point because the damage
isn't going to change the course of the match much - which isn't to say the crap damage
from a d+P major counter makes it useless, just that the damage isn't what's scary about
the move. By the same token if a move does no damage but sets up further damage (i.e. an
inashi reversal), it isn't getting any award in the damage department, but it'll get
some points elsewhere for frame advantage.

85~100 = 10 pts
75~85 = 9 pts
65~75 = 8 pts
50~65 = 6 pts
40~50 = 5 pts
25~40 = 3 pts
1~25 = 1 pt
0 = 0 pts

ADVANTAGE - Still straightforward. I'm putting knockdowns and +12 or more in the same
category, since they amount to the same thing when you think about it: If you're knocked
down, I can stuff a wakeup d+P (or rising attack) with lau's 21 frame df+P+K the same way I
could if I somehow got a miraculous +12 bonus against a standing opponent.

The reason for a distinction between +6~11 and +12 and up is because there are a few key
power combo starters that will beat a low punch at +12 or more, but fail at e.g. +9 or so.
These include kage's df+P, wolf's short shoulder, goh's b+P+K, and lau's superknife.
Also +12 is good for guaranteed damage from a PPx combo or d+P.

There's a large drop between +5 and +4 because +5 sets up some of the truly damaging
combos in the game if you stuff a 12 frame jab (wolf/jeff/goh knee for example).
There are a few 16 frame moves that lead to big damage but the damage is usually not on
the same scale as a knee-type move. Successive drops are also large because they're
common and require immediate action to be useful, and even then the reward isn't much.

+12 or more/knockdown = 10 pts
+6~11 = 9 pts
+5 = 8 pts
+4 = 6 pts
+3 = 4 pts
+2 = 3 pts
+1 = 2 pts
+0 = 0 pts

Circularity - For consistency's sake, there's no bonus awarded to a move just for existing.
Therefore linearity = 0 pts bonus.

Linear = 0
Half circular = 7 pts
Full circular = 10 pts

Hit level - Again, a 0-10 pt scale demands high gets no bonus at all. If you hit them
with a high attack, it wasn't because the attack was high (as opposed to mid or low). It
would have worked if it were some other hit level, usually (always?). Therefore having a
high hit level isn't worth any bonus whatsoever.

High = 0
Special High = 4
Special Mid = 4 (i.e. shun's chouwan)
Mid = 7
Low = 10

Range - Pretty straightforward. Again a 0-10 pt scale. I decided that if a move has less
range than an elbow/low punch, then it deserves no bonus at all. The only moves with
that kind of range pretty much move you backwards. Therefore I'm giving backwards
moving attacks their own bonus under the 'priority' category, since it allows one to
avoid attacks and possibly counterattack.

Sidekick+ = 10
Sidekick = 7
elbow+ = 5 (this is between elbow and sidekick range, i.e. beat knuckle)
elbow = 4
< elbow = 0 (less than elbow range)

Priority - needs more work than I can do right this second, so I'm leaving this open.
I'm thinking of a flat (unweighted) point bonus for various categories like:
sabakis
avoids certain mids or faster high attacks
evades or has canned backdash/dodge

Special bonuses -
slams in floats?
causes difficulty when counterattacked? (i.e. shun's sweep sometimes is too low to elbow).
guarantees a low throw on hit/MC?
has added value when used vs on staggers? (wolf's f+P or b+P)

 

i haven't read ALL of this but i've read most. Will moves get more points for being double-limbed? which can only really be reversed by Aoi, or how bout the Chouwan how it forces you to stand if you're crouching, will any of that effect it's rating? i'm not trying to make things more difficult just curious /versus/images/graemlins/smile.gif

 

but um.. seriously..
any help for the noobs to understand what the hell is going on?

 

hey pghacker, I'd be glad to clear it up.

My goal is to make a system which will let you use numbers to determine how good or bad a single attack is. You can take any attack, plug in some numbers related to that attack (like its speed, its damage, its recovery) ... and through a mathematical formula, you will end up with one big number. That big number is meant to measure the move's overall value based on its speed, damage potential, recovery, and a bunch of other things. The higher the "big number" is, the better the move is (if I did my math right).

The frames stuff -
Frames are a measurement of attack speed and attack recovery. Slower attacks take a lot of frames to execute, fast attacks don't take so many frames.
Once an attack is blocked, you can measure how slowly it recovers by frames also - a move that is -5 is uncounterable, a move that is -8 gives the opponent guaranteed throws, a move that is -17 gives the opponent a free knee combo.

A short example of what I was hoping to do -

Akira SDE - fast, safe, gives combo on any hit. Hits mid.
Total value = 260 pts.

Shun chouwan - a little slower, very safe, gives a combo on any hit, hits special mid.
Total value = 220 pts

Goh f,f+P - slower than chouwan, still pretty safe, good damage only with MC, hits mid.
Total value = 170 pts

So I'd call akira's SDE better than shun's chouwan and goh's f,f+P worse than either of them. That's a super simple version with made up numbers.

 

Hey - double limbed will be under the special bonuses since it's rare and not a huge deal.

Shun's chouwan standing you up is covered under hit level - a mid move that doesn't damage crouchers but merely stands them up is called special mid, I have a category for it. It's a little worse than true mid.

 

Re: Revision 1, ranking VF moves.

I like the new revision /versus/images/graemlins/grin.gif . I think everyone would agree that you know the frames game and I'm interested in the way you have scored them depending on guaranteed moves and advantage. Things like "lau's basic movelist shows 11 moves at -13/14 and only 3 at -15/16." certainly add to the weight of this 'formula' (if it is a formula /versus/images/graemlins/smile.gif).

[ QUOTE ]
Priority - needs more work than I can do right this second, so I'm leaving this open.
I'm thinking of a flat (unweighted) point bonus for various categories like:
sabakis
avoids certain mids or faster high attacks
evades or has canned backdash/dodge

[/ QUOTE ]

Unwieghted sounds good. As far as sabakis go if the system is set up so that the better the actual sabaki is, the more points it gets, then I think unwieghted is good. The reason I say this is that Lei-Fei's sabakis will obviously have alot more use than say Shun's and should be scored accordingly, but I don't think sabaki properties are enough to give it any weight. I'm of the same opinion when it comes to the 'evading' moves.

[ QUOTE ]
Special bonuses -
slams in floats?
causes difficulty when counterattacked? (i.e. shun's sweep sometimes is too low to elbow).
guarantees a low throw on hit/MC?
has added value when used vs on staggers? (wolf's f+P or b+P)

[/ QUOTE ]

Slams in float should get some bonus, although with the fact that they can be TRed now I can't imagine it being much. The other factor to maybe consider is that some characters have a slow down attack (lei-fei comes to mind) and it might not be as useful if the follow up isn't guaranteed on a successful slam.

Difficulty on counter attack seems to be in the same category as moves that evade other attacks. I don't think it's major enough to add.

Guaranteed low throw should definitely be in there in my opinion. Jeff's db+P+K comes to mind - it's slow, and doesn't do much damage. Its real purpose in my mind is to give you a low throw (or guessing game), and without it included in the score I don't think the move would be represented as well as it should be.

Keep it up /versus/images/graemlins/grin.gif.

 

Re: Revision 1, ranking VF moves.

As far as sabakis go if the system is set up so that the better the actual sabaki is, the more points it gets, then I think unwieghted is good.

Sounds good. I'm thinking something like this (I'll need to scale these later) - if it sabakis high attacks or attacks that are relatively uncommon (crescents, sweeps, low kicks) then it gets a small bonus.
If it sabakis mid attacks it's a little larger. If it sabakis elbows or low punches the bonus is larger yet, since those are very common.

+5 - sabakis high elbows
+5 - sabakis crescent kicks (does this exist?!)
+5 - sabakis sweeps
+10 - sabakis high punches
+10 - sabakis knees
+10 - sabakis high kicks
+10 - sabakis low kicks
+20 - sabakis sidekicks and mid kicks
+20 - sabakis mid punches (not including elbows)
+20 - sabakis elbows

So goh's b,f+K is 10+20 = 30 pt bonus.
Lei fei's Tiger stance P is 5+10+10+10+20+20+20 = 95 pt bonus. At first this looks a little fuxxored but consider that lei's move gets 0 points in the speed category, a low score in rec., 0 points for circularity, etc.
Goh's b,f+K will get more points in recovery and range and might end up being rated higher overall.

Slams in float should get some bonus, although with the fact that they can be TRed now I can't imagine it being much. The other factor to maybe consider is that some characters have a slow down attack (lei-fei comes to mind) and it might not be as useful if the follow up isn't guaranteed on a successful slam.

So maybe apply a bonus only for moves that can conceivably hit in floats and where the stomp followup is likely, and a larger bonus if a move slams enough for a ground throw and higher yet for a pounce. I.e. akira's qcb+P gets no slam bonus, who the hell uses that in floats? But AS3 would receive a small one, and wolf's b+P (which creates a free pounce if not TRed) would get at least double the bonus points.

Difficulty on counter attack seems to be in the same category as moves that evade other attacks. I don't think it's major enough to add.

Maybe it shouldn't be bothered with, yah, it's an issue for like... 10 out of 1,700 moves.

Guaranteed low throw - I'm thinking I could put this in damage potential, but triple low throw escapes sort of fucks that. It deserves points just for forcing the opponent to work and enter those escapes though.

I was thinking for guaranteed throw moves, I'll use this formula under damage potential:

Attack MC damage + damage from the character's "third throw" - the third-most damaging throw direction and the one people aren't usually going to escape with a typical double throw escape. For low throws though, this might give an attack an unfairly high damage potential rating.

How to handle normal hit vs major counter vs minor counter damage potential is still fucking with me.

My feeling is - get the point rating for normal/mC/MC ... and average them.
Round up.
I.e. akira's SDE is (6+6+8)/3 = 20/3 = 7.
Lau's upkn gives (3+9+10)/3 = 22/3 = 8.
Lion's FC upper gives (1+3+9)/3 = 13/3 = 5.

 

Re: Revision 1, ranking VF moves.

Creed how did you get 6pts for normal and mC hit on the SDE?

 

Re: Revision 1, ranking VF moves.

off the top of my head. Seriously, six was a bad figure and if I played akira any, I'd know better. I knew that in some situations AS3 is a possible followup, and AS3+ground punch is 57 damage - right in the 6 pt range. However in the other stance you only get 45 damage for SDE-m-dbpm ... and that's 5 pts.
Lastly, these combos only work on the lightweights/girls. Vs. others, I think you'd need the tightest m-dbpm in the world to do SDE-mdbpm. I'm not even sure it works vs midweights. Against lion I cannot even get a low kick after any SDE (he TR's).

And to top it all off, you really need to be watching stance etc closely to get the AS3 or the m-dbpm and even good japanese players don't try it that often and let a lot of normal hit SDE's slide even vs lightweights.

So real numbers for SDE are a little trickier.
I'd say it's 1/1/6, averages to 3, and I'd just mentally upgrade that to 4 because there's a chance of good damage on normal/minor counter hit in a few situations, and the MC combo is easy and doesn't need stance watching.

Thanks for paying attention /versus/images/graemlins/laugh.gif

 

Re: Revision 1, ranking VF moves.

ok here comes my curiousity again to confuse things (if they haven't already been stated) what about stances? those are considered moves right? i know some stances are obviously different then others and because i'm an Aoi player i'm thinking more of her stance and how it's a move at the same time.

 

Re: Revision 1, ranking VF moves.

Stance moves will be in there too and just treated the same as normal moves. Technically there's a catch (that you have to enter a stance before being able to use a move) ... but it isn't that big a deal, most players can find a neutral spot or a string that can put them where they need to be.

Maybe I can do a fixed bonus (like a subtraction of points) if a move must be used from stance only. That automatically limits when and where you can do it, limits your abilities when the move is accessible (i.e. you cannot guard if you're in a position to kage's jm K+G), and generally being in the stance 'telegraphs' the move. So maybe subtracting a few points from a move for being a stance attack is the right way to handle it.