Throw Comparisons

[Nybb]

So, I was bored today and needed to take a break from studying for exams, so I decided to look into something I've been curious about for a while: relative throw strengths in SSB. I'm sure something like this has been done, but I've never seen it since I've been/lurked here.

Basically I just wrote down what percentage Mario had to be at in order for a throw to kill him off the left platform on Dreamland. The numbers are all plus/minus about five percent since you can't just set damage in training mode like in Melee/Brawl. After compiling all the data I decided to post it here in case anybody else was interested.

Throw Strengths
Here's the table with the percents. The numbers in and of themselves are not meaningful, rather it is the numbers relative to each other that should be considered. Also, remember that they are percents, so lower = more powerful.

     | Lu | Ma | DK | Li | Sa | Fa | Ne | Yo | Ki | Fo | Pi | Ji | Avg
Back | 120| 100| 70 | 100| 80 | 100| 70 | 100| 130| 120| 80 | 90 | ~97
Fwd. | 100| 120| 170| 100| 90 | 185| 80 | 130| 115| 155| 130| 160| ~128
Avg. | 110| 110| 120| 100| 85 | 143| 75 | 115| 123| 138| 105| 125| ~112

Some noteworthy results:

- Best Fthrow: Ness
- Worst Fthrow: Falcon
- Best Bthrow: DK, Ness
- Worst Bthrow: Kirby
- Best Average: Ness
- Worst Average: Falcon

- Luigi is the only character whose Fthrow is stronger than their Bthrow (edit: oops, and Kirby -- thanks Blue Yoshi ); Link is the only one for whom both throws are identical.
- Obviously Falcon's, Puff's, and DK's results are a bit skewed as their Fthrows are not really for killing. This would mean that Fox actually has the least powerful throws overall.
- Luigi and Mario are seriously average. lol
- For comparison, Falcon's Up-B "throw" would fit into this chart with a value of 100 using the same test.



Edit: just to clarify a bit more, I didn't really consider trajectory in any way for this data, only the absolute killing potential under identical conditions. From below:

Also, I have to point out that depending on what stage and where you are in the stage, the throw damage to kill values are completely different. Take the top platform of hyrule for example. The worst throws on your list here are Falcon's, Puff's, and Kirby's forward throw, because it sends the opponent upwards, not horrizontally. On the top platform of hyrule, these would be the best throws in the list, because it sends the oppoent upwards.

I was mainly just going for consistency in the test for the sake of comparing the raw launching power of each throw on average. Trajectory is something I didn't really take into account at all. For example, I think (just eyeballing it) that DK's bthrow actually beats Ness' for raw launching power, as it seems to launch at a higher trajectory but still kills at about the same time.

Edit: Grab Ranges.

This test was just me seeing whose grab range beat who. I just made the characters face each other and try to grab, and if neither could reach, move them closer in tiny increments, save-stating each time. As soon as one could grab, I went back to the save state to see if the other could also grab from that distance; if not, then the first person got ranked higher.

Samus
Link

Yoshi
DK

Fox
Kirby
Pikachu

Jigglypuff, Falcon

Mario, Luigi 
Ness

- Some of the differences came down to idle animations. That is, at one distance, both could grab each other consistently; at the next farthest distance, one could not, and the other only could grab upon catching the first in the right pose, i.e. not consistently.
- The spaces in the list very roughly correspond to "tiers" of grab range. These are solely based upon me eyeballing it.



Edit: Grab Speeds.

So I got some recording software going and looked at the grabs frame by frame. Here are the frames between the start of a grab and the actual "hold" occurring.

Samus: 22
Link: 17
Yoshi: 18
Everybody else: 6 - 7

- I would allow for an error of plus or minus one frame, just due to framerates of recording equipment and the actual game maybe not syncing up quite perfectly, and also due to the actual start-up of grab animations possibly being ambiguous. This leads to the next conclusion:
- Every non-tether grab probably takes the same amount of time. Possible error prevents me from concluding this for sure, but it seems to make sense. I would also guess that Link and Yoshi take the same amount of time; however, the difference to Samus seems too big to be an error.

 

[Dajayman]

Interesting. I already knew Ness has the best fthrow, but I always thought DK and Pika had better bthrows. Guess not.

 

[SpongeBathBill]

Oh, man, this is really interesting.

Nybb, what with this and the priority thing, you should really be a Smash Researcher :D

 

[Blue Yoshi]

lol why does it always seem that I'm the one criticizing and correcting your things I should almost just tell you in person instead... since we do see each other quite often lol

Kirby's and Luigi's F-throw are both better than their B-throw, not just Luigi.

Yoshi's throws seem to be most average according to your results:
-Forward throw differs by 3
-Back throw differs by 2
-Average differs by 3

I don't think there is much of a use for "average", as some characters' throws are too different to be averaged well. You did point this out, giving Falcon and Puff as good examples. I would replace this with "best throw", with all the lowest percents in there (everyone's back and luigi and kirby's forward).

Also, I have to point out that depending on what stage and where you are in the stage, the throw damage to kill values are completely different. Take the top platform of hyrule for example. The worst throws on your list here are Falcon's, Puff's, and Kirby's forward throw, because it sends the opponent upwards, not horrizontally. On the top platform of hyrule, these would be the best throws in the list, because it sends the oppoent upwards.

Just a few minor technicalities I had to point out. If you do need any help in any of the research you are doing, I can always help you out. Testing against CPUs can be really frustrating at times (especially when you need to raise their damage level to over 500%). Plus we see each other at least once a week lol

 

[Nybb]

Kirby's and Luigi's F-throw are both better than their B-throw, not just Luigi.

Yoshi's throws seem to be most average according to your results:
-Forward throw differs by 3
-Back throw differs by 2
-Average differs by 3

I don't think there is much of a use for "average", as some characters' throws are too different to be averaged well. You did point this out, giving Falcon and Puff as good examples. I would replace this with "best throw", with all the lowest percents in there (everyone's back and luigi and kirby's forward).

Also, I have to point out that depending on what stage and where you are in the stage, the throw damage to kill values are completely different. Take the top platform of hyrule for example. The worst throws on your list here are Falcon's, Puff's, and Kirby's forward throw, because it sends the opponent upwards, not horrizontally. On the top platform of hyrule, these would be the best throws in the list, because it sends the oppoent upwards.

Thanks for pointing out Kirby, 'twas an oversight on my part.

And I was thinking of averages in a different way. I just literally took the mean of both throws, and this average was what I used for comparison. You're right that Yoshi is indeed the closest to average on a per-throw basis.

And yes, I'll add more clarification on the vertical throws to the OP. I was mainly just going for consistency in the test for the sake of comparing the raw launching power of each throw on average. Trajectory is something I didn't really take into account at all. For example, I think (just eyeballing it) that DK's bthrow actually beats Ness' for raw launching power, as it seems to launch at a higher trajectory but still kills at about the same time.

Nybb, what with this and the priority thing, you should really be a Smash Researcher :D

lol, I'm sure that somebody has done something like this before in all the years the game has been out, so it hardly qualifies as research. I just kind of do stuff like this when I'm bored


I might also do a similar test with the 'turbo' attacks in the game, i.e. to compare Falcon Punch vs. Charge Shot vs. Rest etc. I've been curious about that for a while too.

 

[NixxxoN]

Ness and DKs grabs are too good.
Oh and Falcon's too, considering the long easy combo that awaits after the grab.

 

[Kefit]

Interesting. I wasn't aware that there was such a huge difference between some fthrows and bthrows. I mean, I knew a difference existed, but I had no idea it was THAT large.

However, I'd be more interested in info such as grab speed (in terms of frames) and range. The actual grab is usually far more important than the resultant throw, and would be better basis for comparing throws between characters overall.

 

[ShellFall]

i agree with kefit. i would be interested in throw speed as well. but regardless good info and good post.

 

[Daedatheus]

grab speed (in terms of frames) and range.

GRAB RANGE!
I second this motion. Also any oddities regarding grabs such as Samus' and Link's grabs might be interesting to know.

 

[Dajayman]

Funny how long this game has been out and just now we are researching the game mechanics.

 

[Blue Yoshi]

No, there are a couple of people that have gone to in-depth research of the game, going down to the decimal of character weights, etc. No one has done anything like it for a while. Ant-d was the main person doing the research I believe.

 

[Nybb]

I've just tried Mupen for the first time, and it actually seems to run on Ubuntu fairly easily, which is more than I can say for Project 64. Hopefully I'll be able to look into grab ranges with some detail now. I'm just trying to get rid of a random flashing yellow bar that seems determined to say at the top of the emulator window whenever I am running smash.

Edit: I put up some data on grab ranges.

 

[Daedatheus]

Wow. I'm really surprised Ness has less range than Mario and Luigi. For some reason it feels like his grab has at least mid-level range, maybe just because he's faster and techchases better than Mario or Luigi.

 

[MattNF]

I've just tried Mupen for the first time, and it actually seems to run on Ubuntu fairly easily, which is more than I can say for Project 64. Hopefully I'll be able to look into grab ranges with some detail now. I'm just trying to get rid of a random flashing yellow bar that seems determined to say at the top of the emulator window whenever I am running smash.

Edit: I put up some data on grab ranges.

I think Mupen was made for Linux, that's why it works so well

 

[otg]

Very cool, I like this alot, helps put somethings into perspective.

 

[Blue Yoshi]

Samus
Link

Yoshi
DK

Fox
Kirby
Pikachu

Jigglypuff, Falcon

Mario, Luigi
Ness

Haha and you used to say DK's grab range isn't broken compared to the rest of the non-grapple cast, no matter how much we tried to convince you otherwise. Now, you've proved it yourself

 

[Cryptic C62]

I think the length of the invincble throw animation is also useful, especially in team battles and against Link's boomerang. I think Ness's bthrow has the longest invincibility, yes?

 

[Fireblaster]

Samus
Link

Yoshi
DK

Fox
Kirby
Pikachu

Jigglypuff, Falcon

Mario, Luigi
Ness

Wow, yet people complain so much when you grab spam with these guys.

Yet it's weird that pikachu's grab has such low range, it always seems ridiculously prioritized. Maybe it's because his grab hitbox comes out before everyone else's? You should do a test on that as well.

 

[Nybb]

Look at grab frame data!

I'm trying to look at some frame-by-frame stuff too, but I'm having trouble with Mupen's video recorder. It really seems to like freaking out when I start recording and sometimes it complains about memory, although I can't see how that would be possible. This computer is not the best, but if I can run Civ 4 I would think that I would be able to handle an N64 emulator...

Haha and you used to say DK's grab range isn't broken compared to the rest of the non-grapple cast, no matter how much we tried to convince you otherwise. Now, you've proved it yourself

He also has a minimum grab range though...that is, there is a place where an opponent can be standing and he is too close to grab them. I've never noticed that with anybody else, although maybe I just don't really grabspam enough with other characters that I play.

Wow, yet people complain so much when you grab spam with these guys.

Yet it's weird that pikachu's grab has such low range, it always seems ridiculously prioritized. Maybe it's because his grab hitbox comes out before everyone else's? You should do a test on that as well.

I think the main reason Pika's range seems stupid is because he is so small. Most characters' grab ranges are visible (i.e. arms or tethers), whereas his extends way beyond his body. I would theorize that this relates back to the concept of priority, where the hitbox of his grab is way larger than his hurtbox, leading to many attacks not being able to reach his hurtbox before the attacker gets grabbed.



Edit: I've been playing around with some other video capture software, and have sort of got it working. A trial run seems to show Kirby's grab grabbing on frame six...I'm not sure about framerates and stuff though. I think I am going to try to find antD's move frame data and compare to see if I'm doing this right before I do every character.

Edit 2: So I found antD's tables on jab frames and compared a couple of those to my own observations, and my data usually matched or was off by one. So then I went and counted up some jab frames, and the data is now in the first post.

 

[Fireblaster]

Did you also take into account that samus/link/yoshi could possibly have different grab frame ranges due to the nature of the grapple? If their opponent was right next to them, it would grab earlier. But at least with samus's case, I definitely notice when people are farther away when they get grabbed and it definitely takes longer.

 

[Nybb]

Did you also take into account that samus/link/yoshi could possibly have different grab frame ranges due to the nature of the grapple? If their opponent was right next to them, it would grab earlier. But at least with samus's case, I definitely notice when people are farther away when they get grabbed and it definitely takes longer.

The frames listed are the earliest possible times that a character can be grabbed (i.e. when the tether grabbers are right next to their target). If you're meaning to ask about how late an opponent can be grabbed (like from the maximum distance), then I don't know and I can try to look into that.