Important Untechable Reel Animations

[Jams.]

I assume this does not apply to moves with the 361 angle?

ROB's dtilt actually uses this angle, so I can confirm it does work with moves using the Sakurai angle.

 

[thehard]

Hey fellow labbers!
We covered this topic in our newest video! We also labbed it on our own to confirm it but BIG shoutouts and thanks to all of you!
Watch here if you want!

https://www.youtube.com/watch?v=4i6-mldGJog

really unfortunate that this exists, but at least we're informed now ^^

It is being said that you can DI up and jump out of the spinning animation earlier than other hitstun animations. Could you possibly cover this in a follow up vid? This could potentially make all "untechables" avoidable on reaction (or just mash jump and prepare to tech).

 

[C3CC]

I won't lie: finding out about this is really disappointing. Teching is super important and the fact that it's random (not 100% of the time, but you get it) is just ridiculous. We should make an outcry as a community to see if it gets patched or something. I mean, it worked with Bayonetta, right?

... Right?

 

[Lomogoto]

I tested this a bunch using ROB's dtilt on various characters, and I can confirm that on every character I've tested, they were able to jump out of the reeling animation at a lower percent than the 'standard' launch animation.

Thats interesting! To clarify, you are aaying they can jump sooner if they get this animation? Not just that because they are more horizontal in this animation they do not hit the ground as soon, and are more likely to be able to jump before lamding?

 

[Funen1]

It is being said that you can DI up and jump out of the spinning animation earlier than other hitstun animations. Could you possibly cover this in a follow up vid? This could potentially make all "untechables" avoidable on reaction (or just mash jump and prepare to tech).

My first impression is it's not so much being able to jump out earlier than usual as it is preventing yourself from hitting the ground during the untechable period, thus allowing you to jump properly anyway (or tech the landing if you so choose). But yeah, that is the general strategy to avoid this. Same as with moves that would otherwise force people to tech at low percents as well. Presumably though, with both cases you wouldn't be DI'ing up so much on reaction unless the move has a lot of hitlag or something - you're more likely to do it from anticipation when you realize you've left yourself open.

 

[A10theHero]

Hi everyone! c:
I was curious what the probability of this animation occurring is, especially since in the BSD video, it was said to be ~25% and in that Tweet that Shaya posted, it said 30%. Since I don't have class today, I ended up Thunder Jolting Captain Falcon 250 times at 999% in Training Mode with Pikachu. Here is the data and results: https://docs.google.com/spreadsheets/d/1ic4ae5IpjuDdZXBhXCEBH1ExQSBZ8vL7BQLgjIPjTu4/edit?usp=sharing

The experimental probability from these close-up Thunder Jolts was 0.28 (28%). Keep in mind that this number is the experimental probability. The actual/theoretical probability of this animation occurring could very well be .30, .25, .28 or even some other number. At the very least, we can say that it certainly is in this general area.

The assumption is that as the number of trials increases, the result should approach the theoretical probability. Hoping that a graph plotting the relative frequency vs the number of trials would hint at the actual probability, I made this:

Spoiler: Relative Frequency vs Number of Trials

Spoiler: What is Relative Frequency and What is the Point of this Graph?

Relative frequency is exactly what it sounds like: It is how often a certain event occurs (its frequency) relative to the total number of trials. Mathematically: Relative Frequency = # of Times Event Occurs / # of Trials. This graph shows the relative frequency after each trial. For example, in the first attempt, Captain Falcon underwent that reeling animation. So the relative frequency at that point was: 1 Success / 1 Trial = 1. However, the probability of this animation occurring is very clearly not 100%. As I mentioned before, as the number of trials increases, the relative frequency approaches the theoretical probability. So the purpose of this graph is to see if there's a certain value, like 0.3 or 0.25, that the relative frequency seems to hover around as the number of trials increases. That number would ideally be the theoretical probability. Here is an example of this using flipping a coin as an example (remember that the theoretical probability is .5 / 50% for both possible outcomes, "heads" and "tails"). If you still have any questions about this, I'm happy to help! ^-^

Here's the graph again but with lines at 0.3 and 0.25:

The data seems to suggest that the actual value is not actually 0.3 or .25, but something between them. A close-up on the last 50 trials confirms this:

(A line is drawn at 0.28)

Although all data for the last 50 trials is within this range between 0.27 and 0.29, I wouldn't be comfortable calling 0.28 the actual probability and leaving it at that. Maybe after having done 1,000 trials or so I would. But I don't have the time for that right now, unfortunately.

Some other questions I have are: Does the probability change depending on the victim's percent? Do different moves have different probabilities of causing this animation? Is the probability character-dependent?
I have a feeling that the answer to all of these is "no", but it'd be nice to have some confirmation on this. I likely will try to answer these myself when I have the time if nobody else knows for certain.

Thanks for reading!

 

[Jams.]

Thats interesting! To clarify, you are aaying they can jump sooner if they get this animation? Not just that because they are more horizontal in this animation they do not hit the ground as soon, and are more likely to be able to jump before lamding?

I didn't go in-depth with testing, so I'm only confident in saying that they can avoid hitting the stage earlier if they are launched with this animation, but I'm not sure of the exact reason this is the case. I personally believe that the likely cause for this phenomenon is a difference in launch trajectory, launch speed, or collision hitbox position; however, it could possibly be because you can jump out of the reeling state earlier than the normal launch state.

 

[Lomogoto]

I didn't go in-depth with testing, so I'm only confident in saying that they can avoid hitting the stage earlier if they are launched with this animation, but I'm not sure of the exact reason this is the case. I personally believe that the likely cause for this phenomenon is a difference in launch trajectory, launch speed, or collision hitbox position; however, it could possibly be because you can jump out of the reeling state earlier than the normal launch state.

Ok i also think the collision box would be more likely, but it sounded like you were more sure it was timing. Glad i clarified.

 

[Pazx]

Related to ongoing "does it change launch trajectory" discussion, the untechable reeling animation causes characters to die earlier horizontally from high knockback moves. I initially suspected that it was do to with hurtboxes being altered but I only did very basic tests on this over a year ago so it's very possible I've missed a slightly altered launch angle. If someone with recording equipment wanted to create a large custom stage and test whether the trajectory is the same or not with a fixed camera that would be wonderful.

Initial documentation, proof and discussion of the reeling animation causing you to die earlier (or later for certain characters) can be found HERE.

 

[ぱみゅ]

Related to ongoing "does it change launch trajectory" discussion, the untechable reeling animation causes characters to die earlier horizontally from high knockback moves. I initially suspected that it was do to with hurtboxes being altered but I only did very basic tests on this over a year ago so it's very possible I've missed a slightly altered launch angle. If someone with recording equipment wanted to create a large custom stage and test whether the trajectory is the same or not with a fixed camera that would be wonderful.

Initial documentation, proof and discussion of the reeling animation causing you to die earlier (or later for certain characters) can be found HERE.

This might be very important.
The animation may somewhat shift collision boxes causing both phenomenons.

 

[~ Gheb ~]

Related to ongoing "does it change launch trajectory" discussion, the untechable reeling animation causes characters to die earlier horizontally from high knockback moves.

This may have already been the case in previous smash iterations. I specifically remember not being able to reduce momentum during the reeling animation in Brawl.

 

[Darklink401]

https://www.reddit.com/r/smashbros/comments/4si1tw/tested_the_new_untechable_situation_100_times/

Basic testing I did, 100 times, got 35/100 times. Though other numbers are likely more accurate, but I am starting to think it's not only 25%...

 

[Lomogoto]

I have gone through to test if the chance changes with percentage by hitting a DK 100 times so his final percent was 100%, 120%, 140%, 160%, 180% 200% (one hundred hits at each percent)
I got this reel animation as follows:
100%: 30/100
120%: 30/100
140%: 22/100
160%: 29/100
180%: 31/100
200%: 20/100
I kept thinking there was some strangeness around 130-140% but i doubt there is
This makes me think the chance is 30/100 until 200% where is changes again, like how the chance is 0 before 100%

But if anyone else gets more odd results in the early percents let me know

Edit:
At 999% i got 29/100 so maybe its just always 30% and 140% and 200% are outlyers

 

[teluoborg]

Alright so people have been too busy researching the core mechanics and conducting rigorous scientific experiments to think of a decent name.

I propose AIR TRIPPING but I've left other alternatives

What should that thing when you can't tech because you're spinning when you're launched at >100% be called ?

— LC|Telu (@teluoborg) July 14, 2016

 

[Stryker95]

Alright so people have been too busy researching the core mechanics and conducting rigorous scientific experiments to think of a decent name.

I propose AIR TRIPPING but I've left other alternatives

What should that thing when you can't tech because you're spinning when you're launched at >100% be called ?

— LC|Telu (@teluoborg) July 14, 2016

Well, when discussing this with Fox Is Openly Deceptive we just referred to it as "critical hit knockback".

 

[Lomogoto]

I think "critical" is inportant to have in the name since its random so i would think "critical reel animation" or "critical hit stun"
Im sure there is some corny acronim here somewhere
"Critical Reel Interupted Tech" a CRIT?
Hopefully not
"Critical Untechable-State Stun" or CUSS (i think this one could grow on people as this is used on them)

 

[epicnights]

According to SSBWiki, the state is called "reeling", which is different from your regular "tumbling". No need to make a new name when the name has existed before.

 

[Lomogoto]

So how would you make a disticntion from reeling where you can tech? That seems to be the question thats being asked

 

[Ulevo]

Alright so people have been too busy researching the core mechanics and conducting rigorous scientific experiments to think of a decent name.

I propose AIR TRIPPING but I've left other alternatives

What should that thing when you can't tech because you're spinning when you're launched at >100% be called ?

— LC|Telu (@teluoborg) July 14, 2016

Call it reeling and leave it be. No need for silly names.

 

[Fox Is Openly Deceptive]

I haven't had much spare time to test stuff recently but I did manage to look into these claims about there being less hitstun during the critical-hit reel animation as opposed to normal.

I tested Link's soft-hit Nair against CF on various percents between 94 and 300% using airdodge, DJ, aerials and specials to act out of hitstun.

What I concluded was that there is in fact less hitstun, but it's only 1 frame less exactly regardless of the percents and regardless of what move is used to exit hitstun, with one exception.
Airdodge began by acting just like the others, letting CF act 1 frame earlier than normal during the critical hit reel, but then at 121% (in this specific scenario against Link's soft-hit Nair), the amount of hitstun that is endured before CF can airdodge evens out, and from that point it doesn't matter whether you receive a normal knockback animation or otherwise, airdodge will work on the same frame.

 

[Stryker95]

Related to ongoing "does it change launch trajectory" discussion, the untechable reeling animation causes characters to die earlier horizontally from high knockback moves. I initially suspected that it was do to with hurtboxes being altered but I only did very basic tests on this over a year ago so it's very possible I've missed a slightly altered launch angle. If someone with recording equipment wanted to create a large custom stage and test whether the trajectory is the same or not with a fixed camera that would be wonderful.

In regards to this, here is the video requested (bad audio). It appears that the critical hit reeling animation causes one to die earlier not because of a change of angle but rather a change of character head position.

 

[Yikarur]

LOL Air Tripping
10/10

 

[teluoborg]

Call it reeling and leave it be. No need for silly names.

Wrong. Air tripping fullfills the prophecy that has been around since the Brawl days. I really like critical hit too though.

 

[Ulevo]

Wrong. Air tripping fullfills the prophecy that has been around since the Brawl days. I really like critical hit too though.

It's appropriate that you bring up the Brawl days, because that era was particularly bad for creating ridiculous acronyms and terms for tactics, mechanics and techniques that rather than alluding to what they were in a simple way, sought to just give people credit for a discovery.

Case in point: reverse aerial rush.

 

[Radical Larry]

Well I've been using this thing for nearly two years. You guys just found it out, lol?

 

[Dr. Tuen]

Hey all, there's a detail missing, I believe. I may have only found it because I've been labbing reel applications for a very weak nice (ZSS Zair). This detail (in bold) adds to requirements needed to cause the reel animation:

1. Victim end at 100% or greater.
2. 70 degree launch small or lesser.
3. Causes 100 units of knockback or greater.

I found this because ZSS' Zair doesn't work on Mario into 135% under training mode conditions. This also means rage effects the starting reel prevent for low damage moves like this.

So, there's that, which is researched and known. Next, I have a bit that's purely anecdotal, but had been observed by some players I chatted with at EVO... even so, take with salt.

I have run into a scenario where the reel animation happened 100% of the time. This was in training mode, versus Mario at 135, using ZSS' zair. I performed 50 zairs and he reeled 50 times across 50 training mode resets.

This was never replicated again after I quit out of training mode. While this could be a training mode quirk, it may not be. Unfortunately, I cannot think of any other factors at play. If anyone else has experienced this, comment or contact me with your lab conditions!

 

[Ulevo]

Hey all, there's a detail missing, I believe. I may have only found it because I've been labbing reel applications for a very weak nice (ZSS Zair). This detail (in bold) adds to requirements needed to cause the reel animation:

1. Victim end at 100% or greater.
2. 70 degree launch small or lesser.
3. Causes 100 units of knockback or greater.

I found this because ZSS' Zair doesn't work on Mario into 135% under training mode conditions. This also means rage effects the starting reel prevent for low damage moves like this.

So, there's that, which is researched and known. Next, I have a bit that's purely anecdotal, but had been observed by some players I chatted with at EVO... even so, take with salt.

I have run into a scenario where the reel animation happened 100% of the time. This was in training mode, versus Mario at 135, using ZSS' zair. I performed 50 zairs and he reeled 50 times across 50 training mode resets.

This was never replicated again after I quit out of training mode. While this could be a training mode quirk, it may not be. Unfortunately, I cannot think of any other factors at play. If anyone else has experienced this, comment or contact me with your lab conditions!

I too have experienced the 100% reel animation rate using Meta Knight's down tilt, although I do not know how it happened or how to replicate it. I was able to reel Mario beginning at 100% around 10-13 times in a row. The probability that this was 'luck' is pretty low, so I am assuming that there is a specific mechanic at work here I am not yet understanding.

 

[Kenturo]

I too have experienced the 100% reel animation rate using Meta Knight's down tilt, although I do not know how it happened or how to replicate it. I was able to reel Mario beginning at 100% around 10-13 times in a row. The probability that this was 'luck' is pretty low, so I am assuming that there is a specific mechanic at work here I am not yet understanding.

I also saw a post on reddit yesterday of someone saying he was getting 100% reel with metaknight but the post got deleted. I remember he said he was doing FF back dair to dtilt and was getting it so I think that there is something there.

 

[~ Gheb ~]

Air Tripping is way too ridiculous a name to not at least become a meme.

 

[Darklink401]

Air tripping <3

Also this 100% thing sounds scary....what do ZSS zair and MK dtilt have in common, that something like Link nair doesn't have?

 

[Emblem Lord]

I really need to test this **** with Ryu.

 

[Dr. Tuen]

Air tripping <3

Also this 100% thing sounds scary....what do ZSS zair and MK dtilt have in common, that something like Link nair doesn't have?

I wish I knew. I haven't replicated it yet :-(. But if there's one more condition that makes the reel guaranteed, that's incredible valuable.

 

[Darklink401]

They're literally common in....nothing

Uhh...

maybe the % chance is character specific?

 

[Lomogoto]

When i was labbing percents i posted above, at the higher percents i got a couple times with 25+ times in a row being critical, but then by 100 total times the same chances.
With flipping a coin its pretty easy to think youve gotten a trend when you havent. So i would like to hear exacly how many trials were done with any claims (especially with 100% claims).

 

[Xygonn]

Hey all, there's a detail missing, I believe. I may have only found it because I've been labbing reel applications for a very weak nice (ZSS Zair). This detail (in bold) adds to requirements needed to cause the reel animation:

1. Victim end at 100% or greater.
2. 70 degree launch small or lesser.
3. Causes 100 units of knockback or greater.

I found this because ZSS' Zair doesn't work on Mario into 135% under training mode conditions. This also means rage effects the starting reel prevent for low damage moves like this.

So, there's that, which is researched and known. Next, I have a bit that's purely anecdotal, but had been observed by some players I chatted with at EVO... even so, take with salt.

I have run into a scenario where the reel animation happened 100% of the time. This was in training mode, versus Mario at 135, using ZSS' zair. I performed 50 zairs and he reeled 50 times across 50 training mode resets.

This was never replicated again after I quit out of training mode. While this could be a training mode quirk, it may not be. Unfortunately, I cannot think of any other factors at play. If anyone else has experienced this, comment or contact me with your lab conditions!

For 361 angle moves at least you only have to do the 80 units needed for tumble. I'm testing di up plus jump vs Sakurai angle moves. Most people can jump out.

EDIT: The results are back. For Sakurai angle moves, DI up always gives jump out. The same can't be said for lower angle moves.

 

[DavemanCozy]

I tested this a bunch using ROB's dtilt on various characters, and I can confirm that on every character I've tested, they were able to jump out of the reeling animation at a lower percent than the 'standard' launch animation.

This is a nice find.

In Fox's case, this wouldn't matter: he can cover both a missed tech and a jump with a running U-smash. No wonder Fox has such an easier time getting a KO than he should be.

 

[Fox Is Openly Deceptive]

Semi-related, but I just noticed that it's impossible to FF directly out of an untechable spin. See you can't FF directly out of the tumble state, but normally there is a period in between being launched and going into tumble where you can FF without e.g. needing to use an aerial. But the untechable spin makes you go immediately into tumble after being launched, so there's no chance to FF.

 

[Luigi player]

Hey all, there's a detail missing, I believe. I may have only found it because I've been labbing reel applications for a very weak nice (ZSS Zair). This detail (in bold) adds to requirements needed to cause the reel animation:

1. Victim end at 100% or greater.
2. 70 degree launch small or lesser.
3. Causes 100 units of knockback or greater.

What does the 2nd requirement mean? The angle the opponent is sent at has to be 0-70? Then maybe it's possible to influence things like Fox' dair with DI to get an angle that's high enough for it not to happen?

From a few of my testings it seems you're able to influence the angle for upwards hitting moves enough to use vertical vectoring if they're ~78° or less or something like that [resulting angle needs to be at ~67° or lower to be able to use vertical vectoring], but it might not be the same influence for all angles?

Well, nevermind. Fox' nair has 361 and dair (landing) has 45, so it likely wouldn't be enough (still gonna keep the above paragraph for possible other moves people might want to test).

Semi-related, but I just noticed that it's impossible to FF directly out of an untechable spin. See you can't FF directly out of the tumble state, but normally there is a period in between being launched and going into tumble where you can FF without e.g. needing to use an aerial. But the untechable spin makes you go immediately into tumble after being launched, so there's no chance to FF.

That's very useful information, thanks for sharing! It happened quite a bit lately that I wanted to fastfall after getting hit and it just never happened and I didn't understand (cuz I sometimes even mash down to get it). That explains it.

 

[Fox Is Openly Deceptive]

What does the 2nd requirement mean?

Oh well see I had actually worked that out already on the previous page https://smashboards.com/threads/untechable-reel-animations.439039/#post-21305448
but essentially it just means that moves that hit you upwards at an angle greater than 70 degrees cannot induce the reel animation regardless of DI. https://smashboards.com/threads/untechable-reel-animations.439039/#post-21306336

That's very useful information, thanks for sharing! It happened quite a bit lately that I wanted to fastfall after getting hit and it just never happened and I didn't understand (cuz I sometimes even mash down to get it). That explains it.

I know exactly what you mean. It does explain a lot.
I'm interested to know if there's anything else that this affects. For example tether characters are able to z-drop items without using their zair directly out of tumble, so it's useful for them to be able to do this much much sooner than they are normally able to after being launched.

 

[Fox Is Openly Deceptive]

Random update, but it seems this does in fact work if you spike grounded opponents. I tested with Falcon just then. I'm assuming the misunderstanding came from using a move that works differently on grounded vs aerial opponents.