So I never saw a satisfying conclusion to the Arm Girth Saga earlier and wanted to investigate to the best of my ability. I believe I now have a handle on the systems at play and wanted to share my findings as well as illustrate their direct and indirect impact on the current Meta, citing Lucifer's build as evidence. I will also more generally tie this back to my proposal that ARMS, by and large, exhibits Good Game Design with a lot of polish.
First of all, a disclaimer:
you do not need to care about this. If you are just trying to enjoy the game, the handful of frames' worth of discrepancy I'm going to be discussing will have little relevance to your life. This is RIDICULOUSLY complicated, and the devs do not expect the vast majority of players to concern themselves with it.
--
So we all saw this image, I think:
Along with the notion that Arm Girth somehow affected punch speed and rush meter gain rate.
First off I want to clarify a few terms that were easily confused during the initial discussion. I myself didn't choose my words that carefully while I was still trying to grasp it.
- Punch Travel Speed/Velocity - this varies wildly between different Arms, but is identical on the same two Arms across all characters.
- Maximum Punch Distance - this is the max range a punch extends to before being retracted. If it does not hit anything along the way, any punch thrown will extend to this distance and then stop.
- Rush Meter Gain Rate - while a punch is traveling through the air, your super meter is building. This is easily demonstrable in a training Vs match. A single punch from a slower Arm (i.e. Megaton) will thus build your super more than a single punch from a faster Arm (i.e. Cracker) because it will have spent longer flying through the air. This is a compromise/balance attempt with regard to Arm selection because faster Arms offer less gain per punch but can be thrown at a higher rate.
So where does Arm Girth come in?
Different characters throwing the same Arms don't have quite the same full thrown-travel-recovery cycle duration, and seem to accrue super at different rates. This doesn't seem possible with an equal max distance and consistent travel speed, but it is- because t
he hitbox of each Arm is different based on the character equipping it, and
the edge of the hitbox is what determines when max distance as reached.
This was difficult to wrap my brain around and I tried forever to come up with an analogy but I honestly think the best thing I can do is create a stupid-ass ASCII diagram. Pretend this shape is an Arm: (##). This shape is that same Arm on another character: (#). This shape is that same arm on another character: (###). The @ symbol is max punch distance (pretend they are exactly aligned vertically please).
(#)---------------------------------------------------@
(###)------------------------------------------------@
(##)--------------------------------------------------@
We're going to throw them all (starting from the left over to the right) at the same time. Throw!
--------(#)-------------------------------------------@
--------(###)----------------------------------------@
--------(##)------------------------------------------@
-------------------------(#)--------------------------@
-------------------------(###)-----------------------@
-------------------------(##)-------------------------@
------------------------------------------------(#)---@
------------------------------------------------(###)@
------------------------------------------------(##)--@
Notice how despite all traveling at the same speed, the largest one hit the target first.
At that point, the Arm retracts, readying it to be thrown again or guard. This slightly increases the rate at which punches can bee thrown continuously over time; it also slightly decreases the time the Arm spent in the air, reducing Rush gain per punch.
So
Arm girth impacts the size of the hitbox of a character's Arms, with higher girth being associated with faster recovery speed and lower super gain rate per punch. This was noted in
this reddit post:
There is one stat that effects meter gain that varies by character though, and that is Arm Girth. Arm girth affects the size of a characters ARMS, but that size also has an effect on the amount of time it takes an ARM to reach max distance and its total speed. Examples of Triblast on Mummy vs Ninjara:
https://gfycat.com/CooperativeCoordinatedBunny
https://gfycat.com/IcyAltruisticAsianpiedstarling
These clips are not + version arms. Both arms are punched on the same frame, but Mummy's increased arm size makes it reach max distance sooner and retract faster. This means Mummy can block sooner, but has the downside of requiring more punches for him to build a full rush meter. Mummy needs 22 uncharged Triblast punches, while Ninjara only needs 19.
But the story is not quite over. If Arm girth simply determined the hitbox size of any given Arm, then two characters with the same girth would have the same results for all the metrics as described above. Well, I knew that couldn't be exactly the case, because Kid Cobra and Master Mummy have the same, maxed Arm girth- Kid Cobra is just obviously faster to toss out barrages of punches and to gain super than Master Mummy.
So what's the deal? The original reddit post, and this thread on the last page, missed one more variable:
Size. Character size impacts the hitbox size of Arms, with larger characters generally having larger hitboxes. Arm girth can be thought of as a way to provide for larger or smaller arms than another character at the same size would otherwise have. Kid Cobra is the ultimate example of this.
I went into training to do some testing.
First of all, the Arms look very similar in size on the two characters:
In fact, when Kid Cobra charges Bubb to buff its size, it honestly looks bigger on him than the bigger (Character Size-wise) Mummy!
However, I think that due to Master Mummy's overall size, the camera is not placed exactly the same behind the two characters, rendering these direct visual comparisons difficult.
I knew that the data wouldn't lie however.
Tossing out uncharged Bubb punches at the same time,
Kid Cobra got his Rush in 20, while Master Mummy needed 21. Despite having identical Arm girth, Master Mummy is recovering slightly faster and charging rush slightly slower.
Even more interesting to witness was Phoenix. Phoenix is a much slower moving Arm, and thus builds Rush a lot more per punch. Cobra and Mummy both got their Rush on the 16th throw, but
Mummy got his substantially (maybe 1.5 seconds) later.
Kid Cobra is two notches in size below Mummy. I was curious then how two characters of the same size but different Arm girth would match up. So I grabbed Mechanica, kept Mummy and popped on Megaton.
For uncharged Megaton punches,
Mechanica got her Rush in 16, while Master Mummy needed 18. So, we can derive that Arm girth is more impactful if size is constant versus size if Arm girth is constant.
And just to finalize everything and remembering the Redditor's original test, I checked Ninjara, who is the same size as Kid Cobra with lower Arm girth. Tossing out uncharged Bubbs, it took 19 punches to gain his super, compared to KC (girthier) with 20 and MM (bigger and girthier) with 21.
So, that mystery solved, let's talk about Lucifer.
---
Check out this match between Lucifer and Jorpen and you'll get what it's all about in a couple of seconds:
https://www.twitch.tv/videos/153496878?t=06m22s
He's running dual Bubb and just going nuts, crowding with curved punches and constant pressure. He is almost never NOT throwing a punch, and builds super extremely quickly, because the whiffs are not
misses, they are an expected aspect of the build that result in Rush charge. (I don't mean that he's missing on purpose, just that he knows whiffs will have this benefit).
Here are the advantages of this build knowing everything we know now:
- A character with the highest arm girth in the game, resulting in a large hitbox for any equipped Arm, has been paired with an Arm that increases the size of its hitbox when charged on both hands. As you can see in the screenshot above, Bubb gets absolutely huge on KC when charged.
- High arm girth means less rush meter per punch, but quicker recovery from each miss- perfect for a build in which you intend to throw relentless nonstop punches, most of which will miss, still building super at an astounding rate.
- Because Kid Cobra is a character with high arm girth for his Size, he still builds Rush per punch at a higher rate than Master Mummy despite having hitboxes comparable to his, while remaining more difficult to hit.
- If a character seeks to escape the onslaught of constant punching by actually putting distance between the two combatants, Cobra's slide dash can easily get him to back within Bubb's effective range.
This is honestly a near-perfect ARMS build, drawing (intentionally or otherwise) from nearly every system the game made available to the player. In watching it in action, you can easily see how unskilled or even highly-skilled players just below his ability would be absolutely massacred by the nonstop aggression.
You can easily understand from watching how this playstyle got him to rank 13, and it's only due to the match being hosted by a rank
15 that it's such a spectacle. Lucifer should be proud of this build... and he is. To the tune of not bringing any other ARMS with his loadout, it's just 3 Bubbs, any fight.
So now maybe you're thinking, "okay, so you've just proven Kid Cobra is OP and this build is ridiculous, this is obviously broken, everyone is going to be running this now, nerf it, etc."
Except... Lucifer lost that fight. Jorpen's Min Min was way too much for him to handle unless he panicked, and Jorpen didn't panic. In fact, that Dragon-arm Ram Ram is pretty much a hard counter to Lucifer's Cobra. If not for "that mistake he makes every fucking time" (noting that it was his own fault, not the game's or unfair behavior on the part of his opponent), Jorpen would barely have lost any health in those two rounds.
From a meta perspective, Lucifer's confidence in his build is understandable and it's worth mentioning that the 3 Bubb thing could be seen as an intimidation mind game; why would he do that if he wasn't sure he at least COULD beat me? At the same time, literally sacrificing the opportunity to do a single counterpick may have cost him the chance to beat a rank 15 with a little creativity. Neogaf's very own Dan beat that same Min Min build on this very day.
Anyway, hope this helped.
TL;DR: Arm girth and character size combine to determine Arms hitbox size. Higher hitbox sizes yield slightly faster recovery and slightly lower Rush meter gain per punch, and can block/deflect more. Lower hitbox sizes yield slightly slower recovery and slightly higher Rush meter gain per punch, and are typically found on smaller/tougher-to-hit characters. 
σσσσσσσσσσσง
