Monday, November 30, 2015


I've been thinking a bit about the way that jigglypuff is played. I guess in spirit this is a followup post to the Questions for Puff Mains to Consider post having had more time to consider and experiment.

1st posit:
By the data, Jigglypuff can barely be considered viable.
As is, this character sucks (or at least Jigglypuff players do, and what is the difference?) Disregarding public opinion and looking instead at Jigglypuff’s broader tournament success she is borderline unviable. While it’s easy to dismiss the claim with appeals to a small sample size that shrinks dramatically the more stringently you judge by skill, I think that such an appeal would be backwards. Melee has a wonderful built-in and objective feedback mechanic called the results screen. Every win and loss is reflective of two players’ decisions. Every result is earned. If you don’t make it out of pools it’s because your holistic gameplan wasn’t good enough to make it out of pools. If you get 9th at a regional that means that your holistic gameplan was not good enough to get top 8. If you can’t break into top 100 then your holistic gameplan is not good enough to get top 100. Jigglypuff players have very underwhelming results that reflect very underwhelming gameplans. Her “meta” as widely understood simply hasn’t been able to support widespread success.

So what's holding (your) puff back?

In order to dramatically change (your) Jigglypuff’s results, you must dramatically change your gameplan.

For many Puff players, the first and easiest methodology is to focus on execution. Every missed input is a missed opportunity. Jigglypuff’s success depends on taking advantage of small opportunities. Missing them forfeits success. In the same vein, every dropped or suboptimal punish is a missed opportunity. Jigglypuff has a number of weaknesses, but her punish game is not among them. Yet Jigglypuff players, with few exceptions, have shitty ass punish games.

* Where are your flowcharts? Do you know what to do vs what DI in even the most common scenarios?
Falcons have been doing this shit since before Scar was considered a good player. How can you expect yourself to execute punishes if you are counting on blind luck to even identify what they are? How are you using your playtime? Are you playing smart or are you playing hard? One of those is going to cause improvement, the other will cause you to stagnate. Figure out your character’s flowcharts. 20xx exists in addition to youtube, there’s no excuse for ignorance. If this is too hard for you then you've located the essence of your limitation.

* Where do your punishes start? What hits are feasible to land in what situations? Are you playing in a way to enable your punish game or are you simply waiting for your opponent to hand them to you? They might not be so nice. What then?

This brings about a second line of thinking that’s a little bit less straightforward.

Root Optimization

I like 20GX not because they are optimizing falcon's punishes (that can be taken for granted) but because at the very beginning Gravy looked at Captain Falcon's unique properties and rebuilt his whole gameplan/attitude from the ground up. It's similar because the stuff that works still works but they make different decisions than Darkrain does because they’ve looked for situational and character-specific properties (ex: full analog control from any position via pivots) then started rebuilding everything from there. The result is a nuanced playstyle rich in tactics and strategies that all of the best falcon players are pulling from.

If we operate under the assumption that more of the same will only produce more of the same results then it’s obvious that Jigglypuff can benefit from a similar kind of rebuilding. In order to arrive at the best gameplan, Puff’s strategies need to some degree to be rebuilt after examining her core advantages, many of which are unique and unintuitive to the rest of the cast.

Off the top of my head puff's biggest strength (other than being immune to some things like combos or stage control) is her insane aerial mobility. You have enough control to make a decision, change your mind, then change your mind again. That's ridiculous.
A problem is that your horizontal mobility is sort of on a timer. You are only relevant for as long as you are a) in the air and b) at the correct height, meaning that you have to jump at the right time and depending on the situation might not be actively threatening anything in the time between jumps. Most often this problem is softened (but not eliminated) by using the threat of bair to purchase enough space to land/double jump. In the worst cases this strat results in predictable/exploitable patterns that give up stage control/relevancy. It loses harshly to run up shield and thus depends heavily on (largely reactable) empty land grabs to remain even decent. Alternatively, I propose that the timer problem can be converted into a strength by thinking about and practicing a larger variety of methods to prolong/reset range effectiveness (SH bair DJ, ledgecancels,wavelands, etc) and by thinking about puff's threat range as diagonal instead of horizontal. No one has perfected this because it’s very complicated and any single technique has drawbacks if done predictably or at a bad time but the sheer wealth of angles available to puff at any given time is incredible and deserves fine attention. Additionally, in many situations puff's mobility can allow her to cover multiple options but only if she jumps preemptively at a strict timing (ex: covering an in place option with an early crossup bair, then grabbing/resting should they roll with your jump). Sometimes Puffs will do this intuitively but it can be better taken advantage of if examined consciously.

Earning Your Openings

PPMD says that puff's biggest problem is that she can't approach. I agree. Puff's punish game can be improved (a lotttttt) but that’s just a matter of course. Neutral is not so easy and is a more fundamental problem. Any time I see a puff fall apart it's because they're being forced to approach and just can't do it. She just can't pin fox down, can't adjust her spacing horizontally or vertically from no commitment as well as the fox can (note: while this issue exists in other MUs vs fox is by far the worst. I don’t think it’d be all that problematic if fox didn’t exist or didn’t have a gun). Over the course of his career Hungrybox has devoted much of his effort into identifying ways to avoid approaching altogether with mixed results. In every case it requires a lead. In many it requires a stupid opponent. While avoiding the issue entirely in this way is clearly beneficial, I’m unconvinced that it’s the the singular solution that it gets treated like.

(This is mostly verbally. People have a hard time navigating their own cognitive biases. Your perception of your records on stages is much different than the reality in numbers. Etc. It’s difficult to accurately track the success of any tactic without watching vids back with a notebook. In this way I think that the inadequacy of puff’s neutral and dependence on camping is greatttly exaggerated by virtue of ease of perception. It is easy to identify planking. It is easy to talk about planking. It is ultimately easy to solve planking. It is much more difficult to perceive and to work out options from different positions and tempos on stage. If Puff players aren’t willing to do that then it’s no wonder that they struggle and fall back to jumping away. If Puff players don't recognize their neutral wins via selective abstraction or by ignorance of followups then it's no wonder they don't abuse their frequency.)

I think that the same problem can be solved better with what has been called "Incremental Spacing" and with FOOTSIES. At it's simplest, FGC footsies is the triangle of options Whiff Punish beats Poke beats Walk Forward beats Whiff Punish. With the primacy of pokes and the sheer strength of puff's whiff punish game it is easy to neglect a walk forward game or to write it off as too complicated/difficult. This is to your Puff's severe disadvantage.

I think that the majority of Puff’s problems disappear with strong footsies (i.e. identifying, solving, predicting, and executing mixup scenarios). In this way I think that Jigglypuff is in all honesty a relatively weak character choice UNTIL you are able and willing to be the smarter player, at which point she is S-tier. Her outstanding punish and unique positional options create mixup opportunities that when played correctly frequently result in insta-kills. In order to take full advantage of these mixups you must use a) clearly defined setups (ex: Puff on Shield) and b) active prediction. I don't think that anyone has committed much to either. They're too busy floating around waiting for their instincts to tell them what to do. You can and should literally kill someone for predicting when they will sit in shield. You can and you should kill or combo someone for predicting how they will move. Stop playing a 1-player game of "zone with bair until they walk into me or I feel like nairing" and and acknowledge the opponent making decisions in front of you.

In order to whiff punish, a Street Fighter player has to have tested (and practiced) which normals are useful to punish which normals with at which ranges. Do Puff players know any of that? Heck no, we guess from moment to moment. So how can we expect anyone to practice whiff punishing (and subsequently following things up) in different scenarios? Eventually I want to have notes like I have for on shield vs all common options. I boiled grounded shieldpokes down to two simple pokes that compliment each other and grab. Similar, simple systems exist all over the whole game. They just need identified then practiced.

Footsies start where tactics end. When we don't have or know of a tactic to win a situation (i.e. low tiers automatically lose to bair, falcon automatically loses to crouch, etc) then we HAVE to predict and punish (much much much beefier and easier punishes if you're waiting for something. Not jumping habitually after a fireball but actively looking for the fireball, hearing him go "HADO—" and jumping. Gotcha bitch. Jumping after the "KEN" might be too late. Maybe puff is slow enough that you need to jump before they do. With a prediction that's ok!). Luckily for us jigglypuff is a freakin god at it. Perfect mobility. Perfect options.

Bair is an AMAZINGGGGG tool because generally speaking you risk nothing meaningful by committing to it but they have to give you information every time you throw it out. You need information to make predictions and EVERY BAIR IS FREE INFORMATION. It's not a poke, it's not for zoning, it's for information. Isn't that crazy? Even empty jumps can be treated like that because if you get nervous you can just fill it with bair or jump away and start creeping back to center over again. No big deal. Because you don't care about stage control (via being able to use the air off stage as part of your “aerial dash dance”) you have as long as you like to figure your guy out (exception being fox because you're potentially taking damage the whole time from lasers. However that just puts you on a shorter timer, doesn't make this less true). But I think that you should be willing to actively figure him out, then blow his shit up off of predictions. I think that’s Jigglypuff’s strongest methodology but it takes a smart and dedicated player, not the pussy bitch cowards that get attracted to her because upthrow rest is easy and bair saves them from having to think.

Look at your character. Look at her tools. Change your gameplan. Change your results.

Friday, November 20, 2015

Teams for FoxPuff

I have a static fox teammate. I wanted to think about how to better play to long-term and effective strategies so I wrote this in order to brainstorm and distill a bit as well as identify things to talk with him about. Will edit/expand/correct over time.

- - -


Teams is really really cool because there’s so much active problem solving. Little situations can make big differences and well executed, holistic goals can dominate.

One of the biggest differences between teams and singles is that advantages are temporary. Everything that you do is on a timer i.e. what can you get away with before you need to split your attention? You rarely have the luxury of stretching an edgeguard out over 10 seconds. That’s way too long to assume their partner won’t interrupt you. But with good positioning you can frequently get away with whatever you want in small intervals. Fox puff might be the best team in the game at literal murder under 3 seconds but they have to work together.

Hypothesis: Better synergy > Better players.
At root, synergy is: you’re both working on the same or compatible plans.

I think that talking a lot is very helpful. Hopefully you and your partner have good synergy naturally or through practice but you can’t expect him to be a mind-reader. If you want him to do something different, tell him what and why, at least in between matches. Make a habit of asking what you can do to make things easier for him.

When there’s an ambiguous edgeguard etc make a habit of verbalizing “I have ledge” or “I have high.”
When a clear structure doesn’t work because someone doesn’t play their role (example, fox has ledge, puff covers the on stage recovery and fsmashes their landing lag but fox gets up and gets hit with it too), quickly say “That was my job.” It may not have been obvious to your partner in the heat of the moment.
Defining situations where you each have a simple and clear “job” is a huge benefit. The more that you can work out and internalize the stronger your synergy and the stronger your team.


Pros: Super fast
Super good at killing the exact characters that puff has trouble killing
Super fast
Super amazing at getting grabs etc for puff to play off of

Cons: Super 2v1able
Super fragile if puff doesn’t babysit him

Pros: Perpetual threat of instant death off on a laundry list of setups
Auto stage control/safe zone around her
Usually lives forever
Can't be gimped or 2v1ed so she can liberally edgeguard etc

Cons: Soooo Sloooowwww
Super easy to ignore
Can die very early to knee etc

Fox can and should switch targets a LOT. Because puff’s mobility is limited, there are times where he has to pull himself out of a potentially bad situation before it gets bad and fall back to where she can help him. This defensive movement can instantly be turned into an offensive advantage should he catch someone between him and puff for a quick 2v1 rally/kill (Lucky in particular is really good at that).
Puff can and should go on a killing rampage. In teams she’s even more polarized than she is in singles. She can slow the game down and keep fox safe to keep being fox and she can use fox to fish for devastating rests/edgeguards.

I think the simplest way to think about puff fox is that they’re both ridiculous killing machines so you have to use each other to blow shit up but they’re also fragile so you have to constantly take care of each other. Assuming you are watchful and keep your options open every time someone tries to kill one of you they’re leaving themselves open to a quick gimp/grab/upsmash/rest/etc from the other.


A staple teams strategy is “double team the fast faller.” This works well because a) double teaming the fast character means that he is getting hit and thus can’t move freely, b) presumably their partner is a slower character that can’t rush in to help as well, c) fast fallers can die quickly.

Looking at teams, I think that you can translate that 3 part advantage to any team via prioritizing when to fight where. Quick judgements create sudden opportunities.

a) Rather than double team the fast faller you should double team the situationally more mobile character. This is often the fast faller but may be the better player or the one lower in %. It depends on the situation, hence, situationally.

b) look for/create scenarios where one player cannot assist their partner. If an edgeguard is lengthy, abandon it for a quick 2v1 on their teammate, even if it’s only for 2 seconds. You can always turn around and come back to finish the fight with heavy stage control behind you. Consider putting off killing a high% floaty for a while if their repeatedly getting hit away pays off in time spent 2v1ing their teammate. Etc.

c) Every character in the game can die quickly to foxpuff. Identify and practice setups to kill efficiently.


Your goal is to establish a position to pincer and incapacitate the more mobile character while maintaining control. This is not your only strategy but I think it is the primary one for foxpuff.

Just like in singles, while damage is valuable it is NOT an end-goal. Wherever possible volleys should have a purpose beyond damage, largely to link into a strong kill setup i.e. to set up one of several specific hits for your partner (dair to grab to rest, bair to uptilt to bair to rest, bair to upsmash, upair to upair, bair to edgeguard, etc). Maybe it’s not always possible but a melee 2v1 is sooo crazy that I think if you’re creative enough you’ll find something.
Many exchanges are not strictly sandwiches but offer similar opportunities to slip in and mess someone up. Puff can jump into fox’s fight w/ nair rest on a distracted opponent etc.
Similarly, you can often force their hand in some way that helps your partner a lot. Puff’s aerials (on hit or on shield) translate very often into a free grab for fox. A puff in the air cuts off a large number of escape options, making his neutral easier. If fox can force out a double jump from below with an upsmash threat then puff can WoP them to oblivion.

Personally I think that while the Support/Carry model might not be a false paradigm it isn’t a sufficient one. FoxPuff in particular benefit immensely from making instantaneous changes in position in order to create/exploit a setup to rest/edgeguard/etc. Positioning and strategy is not so simple as one player attacks a lot and the other supports him from behind.

Puff in front
Puff’s wide vertical range of effectiveness commands space and funnels the opponent for fox to pick up. Fox can then be more surgical with his attacks as they are forced down under the approaching puff.

Fox in front
Fox's neutral game is godlike. Puff can interfere whenever fox slips up, making fox’s biggest weakness (gets heavily punished) negligible.

Neither of these takes full advantage of puff/fox’s joint ranges.
I think as a rule of thumb puff should stay over fox at ~45 degrees. Fox shouldn’t stray much farther than in reach of 45 degrees. This is of course a little lenient because puff just needs enough time to get to 45 degrees and then descend by the time anything goes down. If she’s free to move then the angle is more forgiving than if she’s occupied. This maintains the advantages of both fox and puff in front over a wide range of situations.
If you picture puff’s range as a big X shape (diagonals being her potential jump/fall path) and fox’s as a sort of V with a line under it (like a K turned 90 degrees counterclockwise) then puff over fox covers an absurd, dense, and complimentary range. They should both constantly try to maximize this.


__FP___XX_ (these seem neutralish but because of threat ranges foxpuff is better at it than most other teams).
_____P_FXX (this is amazing, one of the best positions)
__F_XX_P__ (good but can go bad fast. ___FXXP___ is great but _FXX_P____ is bad)

 ____F_XPX_ is better than ____P_XFX_ because puff can’t be gimped or 2v1ed very well. They are like a happy neutral.
I think  ____F_XPX_ is generally ok but ____P_XFX_ is horrible and fox should just jump to top platform/roll toward puff/something dramatic to get out.



Basically in any position where there’s lots of space between puff and fox (puff can’t move fast to save or to take advantage, even worse when fox doesn’t have horizontal space to stall for time in) foxpuff is losing.
The closer fox gets to the ledge the bigger the risk. Jigglypuff doesn’t need much stage to play with but fox does. He’s allowed to pursue edgeguards etc but they should be plays, not positional stances.

When stuck in a bad spot fox needs to prioritize changing the game by running through or over to arrive back at center/in reach of puff. Puff’s job is to establish a safe zone without getting hit back out. If either are too desperate and pick a bad moment then the bad position can get worse or get one of them killed.


You've got 3 seconds to mess someone up. Where do you want to be?
I thought about it and don’t think that they vary based on who they are teaming with. A 2v1 is a 2v1 and the only difference should be how frequently they come up/how long you can stay in them on average, no?

(sandwich with goal being push him off or up into bair into off
puff keeps ledge safe for friendly fox to use)

(force him off into gimp, puff as barrier to prevent autocombos on fox out, fox whiff punishes hits on puff into edgeguard)

(fox has stronger edgeguard vs low, puff backs him up/kills falcon that goes high)

(high % combo/pin into upsmash, low % catch her slipping and rest
rare to want to engage in a 2v1 vs peach. Position isn’t as strong as move-based setup. Maximize fox’s usable stage and puff cuts off her float/pins her for grab/upsmash. Puff is better in front than fox because she can’t screw up as hard. 2 deep might be better than sandwich when peach is free to move. Peach is more likely to die off of pickups that fox has to keep an eye out for but not obsess over.)

(either force grab/shine or damage by forcing her up so that fox can upsmash later (sheik and peach die to upsmash at ~ same range because identical weight).
sort of weird to double team, mostly just gonna try to bair her into something juicy.
Where is sheik weakest? Away from the thick. Strongest at center stage or gimping. So if you keep her out of center then she is not so able to play her game unless you give her what she wants.
The trouble with sheik/peach is that they live for a while regardless of position etc. And we can’t rely on cheese or edgeguards that will get interrupted. So we have to accept that and a) leave them be when we can be doing better on their partner and b) develop team combos when possible via move choice. When we are just trading hits/exchanges then they are winning handily.)

(puff eliminates fadeback option so fox wins neutral for even more free than normal. Like peach, no reason to bother most of the time.)

P_X_F___ (I think)
(marth seems like an easier version of sheik to me. He can surprise us but doesn’t have much to his name. He’s very very very vulnerable when forced up, so prioritize nudging him onto/above the platforms when it isn’t better to ignore him.)

(seems like peach but slower so potentially easier to lock into rest setup.
a problem is that if you ignore her she gets a charge shot and then puff can die at 40. So whoever is closer has to baby sit her a little bit but it just takes bairs from either of fox/puff.)


At highish but not killable%s in a 2v1 be aggressive because if you get hit and don’t die then fox gets a free grab upthrow upair upair or something.


in sandwiches if they DI up then we combo to upair(if it will kill)/WoP. If they are sent low then we combo to fox’s grab.

if they tech on a platform they die. If puff is close let her cover every option. Fox just immediately runs to other guy. If puff is not close enough to cover everything then let her cover tech toward her.
I think there are lots of places in this team where while fox does have a combo it’s better to let puff take it and to go open up the other guy. If puff converts to a rest it’s better than what fox could have done. If puff drops it no big deal because you get some damage/more opportunities on both.

a 2v1 that kills is an immediate 2v1 on the other dude.

Don’t stagger invincibility if you both die, use your double invincibility to kill someone

Attack through each other’s shield. You break up the situation.

Fox can go to platforms vs double floaties because his gameplan changes to “bair a lot.”

vs lots of characters’ recoveries fox on ledge puff rest landing lag on stage

vs other characters fox covers low puff jumps over his head with fair then keeps fairing if it hits

when puff aerials low to ground, fox can grab for free (beats sheild and hit) and then puff is right there for grab rest and after at 27s

bair rest

shinegrab rest

drill grab
drill shine rest (see him in position then run in on reaction to grab?)

running shine is very strong vs light characters because it ejects them and commits them to a tech as well as recovery time.

look for these shines near the ledge (leffen and hbox both preying on the other getting attacked. Not all of them work but leffen does a lot of move choices that he can’t follow up on but that hbox can kill from) (look at how lucky runs through a lot and that enables mango)

Friday, November 13, 2015

Hbox vs Mango The Big House 5 Analysis

Hbox vs Mango Big House 5

I don’t think that mango played particularly well (relatively passive and didn’t abuse his advantages) but I do think that hbox played the best that I’ve ever seen him. I want to pick up on what contributed to that.

Things to look for in the MU
Fox gets most of his work done by using lasers to force puff to make punishable all-in kinds of decisions. Puff is most vulnerable when landing predictably OR when she lets fox occupy the space directly above or below her. The matchup is difficult for her largely because puff does not have clear or very reliable ways to threaten fox’s positioning. His high mobility and preference to just run to the other side and start up the gun negates traditional notions of stage control.
The MU is very tense because while this is definitely a bad MU puff has a lot of death touches on fox. If he consistently avoids her pokes and doesn’t overextend himself (i.e. then puff is going to have a tough time getting them but they are always 2 steps away at most, which makes it stressful on both players. As you’re taking laser damage it is very tempting to press too hard for unreal pokes, opening yourself up for fox to pick apart with grabs or 15% bairs.

This is going to be somewhat exhaustive and descriptive. Not abstracted yet because I want to be able to confidently identify all of the patterns before I think about how they can be improved on.

Time notation by in game clock

Game 1

8:00 non-neutral start mango lasers and hbox goes high to see what the other will do behind the GO!
hbox takes stage, mango halts for laser stance so the not full momentum bair whiffs then runs under the endlag
hbox goes high to close space outside of laser range,  mango commits to shs so he can’t run and has to shield.
7:54 smart of mango to avoid WL off top fair, that’s not common
hbox fair after rising nair on shield was late and bad FH. Probably ideally shielddrop upair. Gets punished w/ 15% bair.
Don’t like useless WDes vs fox, that’s a lot of frames to commit to nothing in this MU.
7:51 identical fair vs laser situation, I think hbox misjudged and thought fair would connect.
Mango does the same FH out of shield. Probably to avoid a grab. Because it’s a repeated situation you can identify this as a key moment. Gets another bair.
hbox goes high after ledge. Initial fair is unsafe. Mango goes to bair it but is late so he leaves the jump empty. Falls through platform so hbox’s 2nd fair whiffs. That’s a difficult fox jump to punish.
7:47 Mango appears to decide not to press the issue and resets to lasers but bairs under when hbox goes high (3rd time). Obviously pattern recognition from mango.
7:44 interesting position, hbox waits to confirm a probable hit before dropping through platform. No the best spacing from mango. Mango immediately predicts hbox retreating vertically again, then waits for hbox to go high again and gets a bair. More pattern recognition.
7:37 going top platform level allows you to react to any rising fox aerials. Hbox gets a punish on a weird fsmash from mango. No idea what that was supposed to be. But crossup so no grab. Mango FHs anyway. Looks like he uses FH at low % to option select a get out of grab/followup vs puff’s approaches.
7:34 unfortunate
7:30 this time hbox is far enough horizontally to protect his going to the side platform but lands too far in so gets clipped with a bair while fixing his spacing. Ideally shield then shielddrop upair.
bad habitual bair to get out of hitstun. Mango thought about punishing. Mango goes to side plat to pressure ledge then gambles with fsmash but hbox spaced out enough to avoid and gets a nair. Ideally dair grab. Another FH from mango, then decides to initiate shield pressure. Upsmash might have been read on spotdodge but was close to poking. Shielded upsmash should have definitely been rested. Very easy. Stock trade at worst. Would not have missed the stage, toph is incorrect.
Hbox’s few hits have been % inappropriate so far.
7:18 Mango has established that hbox likes to reset to corner of the platform so he can just shark it.
7:14 mango drops shield too late and gets baired
obvious tomahawk grab spot so mango spotdodges but hbox doesn’t take it and forces a shield.
Another FH out of the spot so hbox finally DJ upair rest. Looks like a read, not sure. Anyway good adjustment.
This is jigglypuff at a high level. If you notice a pattern you just need one read and they will die.
7:05 hbox retreats to the side of the side platform then takes stage when mango goes high (but avoids straight vertical spacing). But then he gives it up for no reason.
6:59 WD to avoid grab goes under the jab lol. I don’t know why he chooses to roll. Probably out of fear of grab/upsmash. Not really a fan but that’s a weird spot. Jab after the roll is a “this situation is weird I think I’ll get away with this” decision. Pretty terrible. Gets baired for misjudging spacing, then rolls to reset the situation twice more.
6:52 ahhhh that was the primmmme grab but input too early so it whiffs. Fox whiffed a FH aerial he had no choice but to land. Mango FHs out but DJs this time. Not sure if adjustment to getting upaired earlier (unlikely) or if wants to establish vertical spacing since hbox is close to center.
Hbox gets punished for trying to take to side platform while fox has tempo. Misstimed sdi on upthrow upair.
Bad WDs get him upaired. Fairs to get out of hitstun are iffy but hbox DIs away so it’s alright.
6:42 Spammy fairs to catch if mango decides to FH at hbox for no reason/to procrastinate landing. Mango guesses return to ledge but hbox goes up instead and gets a bair. Empty bair covers if mango shields and maybe a late jump but mango just FHs early again, gets grabbed empty landing. Good.
6:37 backthrow
50/50 dj to ledge or dj back
hbox guesses back then reads an instant sideB/upB probably because the dj was too high for dj to ledge so it telegraphs an intent to do something high/fast.
Hbox grabs ledge to set up for falling bair on low firefox. Unexpected trade but hbox makes it back
6:30 mango uses invulnerability to take ledge and pressure hbox to come back to stage. Hbox stalls over dj aerial range until invulnerability runs out. Getup inv forces a shield. I think that mango likes shielding next to shields to bait a reaction and look for your first impulse. He’s been doing that for a long time.
Mango knows that hbox likes to double roll so he does an extra DD before the grab. No second roll means it’s free.
6:20 that was actually a sick shine calling out the fadeback, surprising it didn’t land. Could have been the stock.
Hbox is demonstrating intent to plank ledge and choosing low risk options.
same situation but the getup is earlier so mango is the one that has to shield. Nothing to do after bair on shied so hbox takes center then pressures the vertical escape. Annnnnnd runs to other ledge.
Note that mango DJs around a potential grab on landing.
6:09 upair from ledge oddly timed so he ends up getting stuffed with jab > bair
takes ledge when mango commits to laggy uptilt
6:04 these off stage engagements are sort of about balancing risk reward via how far you space and expect them to space. Lots of near misses because while you can predict an angle it’s harder to predict the amount of extension and if you’re off you can die. For both characters but fox has it harder.
5:56 lol that’s a funny upair. This is mostly damage to get fox out of CC range and closer to juicy combo % before he gets you.
Mango delays his timing slightly and gets a bair.
5:48 uptilt to catch any eager movement.
Another FH. Those are hard to punish because they are so fast you have to predict the timing and if you are wrong you can get baired. It’s better to punish the fall (like with the grab/backthrow) but fox has a mixup where he can either aerial necessitating a shield/WD, empty land to grab the shield, or double jump over your grab. I don’t think there’s a way to cheat that so you either have guess or take center stage. Lowest risk is probably WD then instant bair toward stage.
5:43 as soon as hbox goes to side plat mango runs forward and shields to protect from dropthrough aerial and force hbox’s hand. Ideally puff stays still and has a tight reaction to fox’s movement out of shield.
I think hbox predicted a pursuit when he went high. He was wrong and mango tries to punish the land but is a little slow.
5:39 has to shield vs vertical spacing, rolls again then reestablishes side platform occupancy but gets spooked into jumping out. Mango misspaces rising bair punish on return but hbox reels in so he gets vertical spacing and confirms a falling bair.
Good CC grab. Good recognition of upair on DI forward. Turnaround bair is good too. No DJ so grab ledge and roll.
5:27 on return hbox goes high to wait out inv. Gets opened up from below because he DJed then faded back in. Either would have been fine but not both when fox runs forward
5:23 I think this nair is horrrrible. Sooo easy to CC shine/grab/upsmash. Mango FHs out.
5:22 I think mango accidentally buffered a roll instead of fallthrough bair but could be wrong. SH away and hbox is in position to pound the fall. See hbox do that to marth a lot. Missed tech, mango combo.

* upair vertical spacing > rest
* grab on tomahawk > backthrow > bair > read dj back bair gimp
* CC grab on bair > upthrow > upair > bair edgeguard
* pound on empty fall > missed tech > mango combo

Stock losses
* fox runs under a puff bair and bairs
* baired while planking

Interesting to note that lots of situations repeat but the winner depends on the timing more than the spacing.

There is a lot of pattern recognition that allows mango to safely pressure hbox’s habitual positioning. Next step to remove variability and to ensure not having to rely on clutch moments as much would be to look for patterns in mango’s positioning (in addition to scuffle habits like the FHs) to exploit similarly. Difficult and might be tempo dependent but that’s the way fox vs puff works.

Game 2

CP presumably because mango wants earlier kills and to be closer to puff. Based on how game one played out it’s a good CP. Puff should die more frequently and earlier while fox’s rate should remain similar.
PS is better if fox is getting grabs, which he isn’t.

8:00 non-neutral start
mango take center hbox runs to side plat
mango goes to punish because this is obvious but hbox reacts, jumpsaway, and pressures land. Too slow on second bair.
Good jump away from fox’s attempt to whiff punish.
Generally speaking I think puff should wait for fox to jump but jump away on reaction to fox’s jumps in.
7:57 tomahawk uptilt barely shieldpokes because of the slant? wow. Reads the roll but upair is input too late. I think not jab resetting was because DI down was unexpected so he jumped for an upair then moved back on recognizing no tech.
Mango could have shined but spotdodges followup then shines. He’s late on the techchase, hbox resets with roll.
7:52 It’s actually crazy that the grab here whiffs wtf melee fox is right there, hbox played that right.
7:52 funny little WD forward under fox jump. Fair could have been better.
I think hbox is trying to bait a spotdodge but mango FHs instead.
This time hbox’s WD was too short so he gets baired.
really slow on the CC grab attempt so eats another and gets grabbed
Missed sdi, comes back from the side, mango sits on platform and gets shieldpoked from below. All puffs should be doing that it’s easy. Iffy fsmash assuming that mango will stay on platform and do nothing. If anything grab would be better.
Mango whiffs his upair punish.
SICK powershield > walk forward and do nothing from mango lmao
7:40 mango takes vertical position but lands empty and DDs when he recognizes that hbox is spaced well for it.
Because hbox committed to an early bair that whiffed he feels pressured to back off so mango takes stage.
7:39 mango makes good calls vs hbox’s off stage jumps but misplaces them and gets faired. Hbox lands on stage so he doesn’t have time to stop the dj sweetspot. Hbox spaces out to avoid punishes but is burning hella jumps. Mango is tripped up by final pound shieldstun and lets hbox back.
7:30 neat little fake fallthrough to bait a bair from hbox.
poweshield pushback messes both players up.
Mango outplays hbox in vertical position and earns a bair for the kill.
7:21 mango shields vs inv because there’s not much stage but hbox recognizes and land uptilts. Hard to tell if the uptilt pokes (barely possible) or if mango goes to jump. Weird decision anyway I would have grabbed, It’s just as fast.
Platform techchase, hbox goes for mindless aerial so he gets nothing and actually gets naired out of the aftermath positions.
Mango chooses to space outside of hbox’s attempts to poke an approach, then fullhop nairs to poke the repeated jump. Roll was probably meant to be a bair OoS. Note hbox WDs away from potential WD grab OoS.
Mango slips up and fairs breaking the rhythm so hbox dash attacks into the dash dance. Reads and fairs the dj sweetspot (it looked like mango tried to DI to ledge and barely missed so it makes sense for mango to do that).
7:07 hbox goes high to avoid inv, both space very carefully and mango falls back to center.
7:03 hbox uses mango shielding as a chance to land. Clips mango’s advancing DD with nair but nair doesn’t do shit in this MU so mango FHs out and  ends up with verticle spacing. Mango leaves it empty and gets hit with a space-filler upair. DIs out so no grab. Hbox barely avoids whiff punish with a jump away. They both whiff a counter attack looking for closer spacing than what was given but hbox committed to pound so its significant endlag gets baired.
6:56 mango feigns letting hbox back when hbox goes to ledge but stays close enough to nair ledgejump.
holy goodness mango almost rocks hbox with well placed smashes but barely whiffs both times via hbox returning to off stage. But those jumps while evasive aren’t very productive so they eventually get whiff punished with grab.
Successful sdi.
6:47 spotdodge is likely failed shielddrop but mango flubs the punish.
Hbox goes for an all in pound but an attack there was somewhat predictable.
6:44 easy rest OoS
6:43 mango FHs over slightly late upair and back to center but spaces too close and hbox is ready for land with a fair
hbox takes the stage then gives it up on a high risk high reward read on roll away. Mango rolls in but whiffs his whiff punish and gets whiff punished lol.
6:36 hbox stuffs the FH aeiral with a dj bair. This is pretty much exactly how he beats m2k 50% of the time lol. I don’t know why fox would want to do that when he could run forward and maybe FH bair after but they do.
Missed tech dsmash. Ideally jab reset fsmash.
6:32 I think hbox tried to bait a horizontal firefox but mango didn’t fall for it.
6:30 good spacing by mango to arrive back at center and hit on shield. Wow and to bair the fair.
6:25 Hbox’s jumps getting predictable.
6:23 lol high risk juicy % on both characters so they are inching into position, no sudden commitments
mango nairs on reaction to shield, assumes advantages, initiates pressure on more shield, takes stage when hbox rolls out but overcommits and gets baired.
Hbox anticipates returning to stage via platforms because that’s what mango’s been doing but doesn’t drift far enough.
AC nair prevents movement (even though as fox I would gladly trade with bair myself) and triggers a kill attempt by both that hbox loses because FH upair is faster than WD fsmash.
6:08 more FHs from mango out of situation, gets a bair and shine OoS’s obvious retaliation. Surprised at that DI, normally that’s a techchase.
Bad nair gets punished followed by another punish on hbox pushing his luck with aggressive spacing after getting hit with an aerial. These aerials piled on damage fast. Links to grab, upthrow good DI, mango doesn’t dash back so he bairs for damage.
5:58 hbox slips over committed jumps from fox but decides he’d rather go to ledge lol. Burns all of his jumps then gets super lucky with the tech and makes it back. Barely misses edgeguard but that was tight anyway.
5:49 lol they want the other to come forward a litttle more so badly. Mango eventually backs off and attempts to punish a land but upsmash is too slow so he gets killed.
5:41 interesting that mango read the fall through with an uptilt although it’s been a bit since hbox stuck to a platform seeing as he was getting punished for it on BF.
5:36 misses dash and gets baired
5:30 the rest of this mathc is pretty ugly, lots of weird commitments at each other and questionable spacing.
5:10 they’ve both landed a few whiff punishes but this one leads to a platform techchase, hbox bairs rather than rest because of stock deficit. Actually predicts the jump back sideB perfectly but needed to turn around fair for extra mobility to hit it. Instead he’s left in a bad position.
5:00 shielddrop dj upair after that bair, not grab.
4:52 mango is starting to give himself time to react again, allowing a bair on this FH nair
4:39 potential shielddrop uapair again. Much better rewards.
4:30 positions vertically rather than continuing to guess at a jumps, hbox guesses and gets whiff punished.

* dash attack into DD > read dj sweetspot w/ fair > fair
* fair on missed upsmash

Stock Losses
* fumbling around each other’s bairs at ledge at high %
* commits to WD upsmash and gets FH upaired
* missed dash and baired at high %
* bairs underneath fox on a platform and gets baired

Hbox’s aerial spacing around the side platforms makes it more difficult for fox to abuse his strengths as long as he chooses to engage. The rewards are still relatively high for either player. Additonally, hbox retreating to the off-stage side of the platform without committing to landing on it greatly reduces the number of angles that fox can use to approach.  However, hbox committing to more aerials and less stage control makes it hard to arrive at the low% kill setups that won him game 1. Damage for damage fox wins, especially on this stage.

Game 3

Much better than FoD for hbox because he can afford to take more damage to fish for his kill setups. This stage buffs puff in the damage for damage to the point that fox likely has to commit to a lot of lasers in order to win it, which is exploitable with good execution.

8:00 non neural start, hbox goes high until mango gives him a route down. Almost gives him a grab but hbox WDs back instead.
Uses FH aerials to close space. Mango whiffs a FH aerial retort.
7:55 good reaction with dj away to rising bair
mango demonstrates intent to reset to lasers after whiffs goading hbox to approach
7:53 looks like he flubs a running shine
hbox leaves this jump empty suspecting no FH aeiral, which lets him upair mango’s bair OoS after the crossup
easy grab
pound covers no tech, tech in place, tech roll in. Tech away was impossible to cover at all so this was great. Also puts in position for following reaction techchase.
Really amazing sequence of reactions. That’s a great lead.
7:44 goes high after rest punish until mango lets him down.
bad decision to land directly above fox, gets baired.
7:37 means to clip fox running through with nair but started too high, has to shield bair
rolls to ledge, fairs fox’s approach.
7:37 I’m assuming that these fairs are supposed to be nairs because they’ve done nothing but get punished.
Fox is too low in % to crossup after hit. Punished heavily with poor di/sdi.
 7:26 hbox demonstrates intent to stay on ledge at least until fox is out of CC range but he’s extending too far. Mango’s in a good spot to bait aerials that don’t really have to be there.
7:17 should have been shined but mango is tunnelvisioning on bair.
7:14 good poke but really really slow on the grab. Mango FHs out. Doesn’t take first punish, tries to get hbox to spotdodge grab, hbox jumps away instead. Hbox is slow on the punish. Dunno why the upsmash was being charged even. Hbox dies.
7:05 Mango’s been running to the side then through so the land sh upair is good. Tries to uptilt instead of grab who knows why. Also dash attacks the missed tech instead of jab please tell me that was a flub.
Reads a FH out of the corner but I don’t think that was likely because hbox’s spacing wasn’t actually pressuring the shield. Gets run under baired.
6:59 tomahawk uptilt attempt was good there, beats shield and jump, but whispy makes iffy spacing horrible spacing.
Immediately whiffs a bair, I think mango wasn’t prepared to be that close so he naired instead of a grounded option to punish. Misses.
Whiffing in front of one another, mango does a bad FH fair, shieldgrab upthrow rest.
6:50 good DI out of the upair to reduce punish. DJ fair a bit early but makes mango jump to center.
Hbox takes ledge rather than contest dash dance.
6:45 hbox goes high (initial bairs are upairable, I think) but mango thinks he’s going to stay offstage so hbox gets to slip in with a nair that should have been CC grab/upsmashed for free. Mango FHs out. Hbox is starting to fill the space above those little engagements but it’s just badly executed so all it does is put puff around fox on a platform.
6:41 mango gives up the direct punish on the upair so that he can more safely aerial what comes after (and so he doesn’t have to shielddrop lol) which in this case is hbox WDing on a platform for no reason again.
hbox pounds nothing. Would have had good reward but mango wasn’t there so hbox gets upsmashed.
6:36 vertical spacing on the platform, hbox retreating to offstage position.
6:35 mango FHs out, hbox snaps to ledge and waits for some kind of endlag.
mango tries the armada downtilt but it whiffs. Very tight spacing from hbox.
6:30 dair uptilt was actually a pretty sick option but mango pulled back at the last moment rather than challenge with a bair so hbox gets grabbed. Upthrow 2nd hit upair. Hbox burns a lottt of jumps coming back then goes straight to ledge. I think he’s trying to get mango away from the ledge so that he can’t get baired on his way back there.
6:10 FH upair that we saw on yoshis but this time hbox doesn’t WD fsmash. Mango lands and shields then bair upsmashes, as he has before. Seeing this repeat, I think it’s designed to get the upsmash to shieldpoke but hbox just keeps rolling.
Hbox jumps away from potential followup after the roll. Mango run up shields looking for an upsmash OoS. Hbox takes ledge rather than challenge, although I think just because he happened to get there fast.
6:03 mango takes ledge but gets up too slowly to punish airdodge. Should shined or acted earlier (maybe he expected a fadeback after the airdodge and planned on hogging) at least ledgedash.
Shield jab shinegrab. Weird string. Maybe to stuff an OoS input. Backthrow likely unintentional.
6:58 hbox does a string of questionable aerials and is then baited into acting OoS and upaired.
Mango is starting to do more and more indirect punishes. Setting up or recognizing a direct punish but then punishing the immediate “quick, cover my ass!” reaction instead.
5:48 hbox is really slow to punish the run-through so he burns his inv on an uptilt
they’re both starting to look for timing attacks over positional ones again.
5:39 I would have eaten the nair and CC grabbed.
a lot of slightly mispaced/mistimed aerials, erring on the side of safe, so it looks like they’re just attacking around one another.
5:35 interesting grab because it covers an extra shl or a too close DD. I don’t think hbox anticipate the landing timing of the shl correctly but good to point out.
gets grabbed out of jumpsquat after roll
good sdi
5:30 catches mango moving to early OoS but not ready to follow up directly
5:26 pretty shitty nair OoS by mango
hbox could have CC grabbed the following nair but was busy holding away, dj fairs too early to punish grab whiff but is in position to punish retreat to platform
 looks like hbox habitually falling upairs after most rising fairs. In many cases I think they’d be better left empty. Could have punished fox’s fall/land.
5:22 mango moves as if to camp then doesn’t camp so gets a run under bair
punishes the waveland on platform. That waveland is assumed now.
Immediately prioritizes vertical spacing after hbox commits to highish djs, gets an upair but good sdi
hbox does the habitual WD on platform before falling through with an aerial, giving mango time to run up shield. Spacing is ambiguous enough to avoid OoS punish.
5:14 dair grabs a whiffed upsmash > bthrow > reads the DJ again, this time prepared for the side B land lag. Chooses dsmash over rest which is good in this situation. Has the edgeguard but barely misses. Perfect angle. 
Mango doesn’t refresh inv again so he gives up the counter edgeguard.
fishing fishing fishing
4:56 if that nair was reactive it was super good
4:54 weird weird djs by hbox but doesn’t get punished
lots of land shields by mango lots of floating away by hbox
mango ends up putting himself in a spot where he has to get baired and he dies.
4:29 mango takes ledge to force hbox on stage, roll through shines when his tempo is awkward 
4:16 after confirming a million of the same let go fair regrabs mango mistimes the shine to punish it. Hbox upairs from ledge to punish the dj from shine. free grab. Should have been upthrow rest but he’s afraid of platform tech so he upairs. Still unsure what will happen so he bairs and mango jumps out. Good reaction to upair the jump by hbox, links to bair but mango has good sdi and manages to grab ledge. Squeezes through and gets a bair.
4:02 that was an amazing grab ledge bair from mango and an amazing airdodge from hbox but mango is ready for fsmash, which is also amazing.
 3:56 mango runs through inv again
3:55 hbox somehow buffers a shielddrop, trying to crouch grab?
3:54 of cool because hbox got hit with the dair while in the air he can grab before fox. Good awareness.
Upthrow, ambiguous DI, good upB
3:48 horizontal spacing converted to vertical spacing to bair
every time mango aerials hbox at low %s he lands and shields because hbox does an aerial out of hitstun every time.
Hbox floating around just outside of FH range looking for a punish.
3:33 lol that fsmash at the WD forward. high tension so drops edgeguard.
3:21 rather than find a mixup mango just keeps giving hbox opportunities to mess up and hbox doesn’t.

* empty FH crossup > upair OoS movement > upthrow > techchase with pound > dj techchase rest
* space out > shieldgrab FH fair flub > upthrow rest
* high % bair
*extended edgeguard vs stubbord upBs

Stocks Lost:
* late pound is shielded > Upsmash OoS
* baited into bair OoS and dj upaired
* ledge stolen, airdodges bair but gets fsmashed

In this game hbox opened the game with an amazing conversion that gave him a full stock lead at which point dreamland being dreamland allowed him to play exchange for exchange without falling behind as long as fox didn’t get an equally amazing conversion or start outplaying him completely, neither of which happened partly because hbox played around the ledge to slow the % building down/scout for drillgrabs and because mango was so willing to meet every challenge head on and play for mostly tempo attacks rather than positional ones.

Game 4

Mango already won on YS. He’s still looking for quick kills and PS is the next closest.
I personally think he should have gone FD. Hbox has been using the side platforms as obstacles pretty consistently. Additionally it’s easier for fox to take his time on FD because there’s only so much else that he can do.

8:00 non-neutral start, mango jumps to center, hbox jumps to side platform as soon as it’s vacant
mango goes to punish but it’s telegraphed and hbox jumps up and pincers mango to the corner really well
WD grab (and missed turnaround) wastes opportunity to just land grab
spotdodges the whiffpunish then grab, backthrow, but mango techs to ledge. Not sure if upthrow would be better. Obviously upair techchase doesn’t work but you could just techchase rest. Under the impression it would kill if sent left.
7:54 the normal getup was pretty predictable I think it’s all he’s done for the whole set.
hbox empty land grabs. Grabs are effective now because of frequency of aerials throughout the set. Hbox does this pattern soooo much in his long sets.
upthrow, full DI behind, pound was never going to hit. Would prefer run after and upair falling fox.
7:49 avoids bair again, forces shield, avoids SH aerial. Same pattern. lmao he even misses his turnaround grab again.
7:46 ideally you CC grab the nair, not shield it.
7:45 interesting, hbox goes to fair running fox but mango recognizes the situation (probably because he used to do it himself) and WDs back.
mango recognizes that he’s in a losing position for scuffles and commits to lasers
7:41 good bair vs hbox coming through the platform but had hbox sheilded it would have been an easy punish.
 ah, looks like hbox recognizes this because that’s what he does right away. Upthrow upair should have been techchase rest but he tries for more upairs until mango upBs out.
7:31 really tricky waveland off esp because he’s been staying on the platform so far.
7:28 goes for it again but mango reacts. Jump away bair is faster than mango anticipated.
7:24 whiff punish attempts until mango converts to dash attack
7:20 mango’s started to sit under platforms and shield, hbox gets a fair by spacing outside of shieldgrab
7:16 mango’s falling bair covered a late punish and a fallthrough, links to upsmash
7:09 fishing super hard with jab fsmash
mango crosses up before inv is done so he gets baired
good attempts to edgeguard firefox on reaction. It’s ok if fox gets ledge esp at this % and in position to punish on stage sideB.
7:02 woah ledgedash where did that come from. Good jump away.
6:58 mango waits so long to jump and DIs in so hard it’s obvious to cover dj sweetspot
6:55 hbox goes to platform then high to avoid fox inv
mango trades with nair when going to vertical bair, not good enough jump angle
6:51 not sure why FH bair here, almost gets grabbed, then he does another FH aerial in the same place. Maybe to fall on top of a late-ish jump from puff.
Ok he does it again so I think that’s it.
But gets CC grabbed, upthrow rest. That’s a bad stock to lose.
6:41 comes down with bair to cut off mango’s route to stage. With that angle it would have been difficult for fox to bair first. WD away from potential CC option.
6:40 lands in front of fox doing reverse fair, dunno why. Jab upsmash.
6:32 mango puts himself in the corner and empty jumps as puff comes in, so puff has time to uptilt before he can cross her up safely. links to grab, upthrow, I think the upair input was late so fox djs out. Should be able to hit his fall with 2nd upair but jump was too shallow. fox jumps out again.
6:25 hbox lands on a platform in front of fox, fox is ready and bairs the landing lag.
6:22 mango finally jumps out of the scuffle situation that’s been getting him in trouble after hbox crosses up.
6:20 looks like he misses a analog jump so he shls in place and gets upaired. I think hbox anticipated another run up shield and was looking to cross up with upair/catch a jump out. Wonky transformation, missed tech, mango combo.
6:15 good double upair, no sdi but that would be tough in that situation.
6:04 hbox doesn’t want to lose lol. Note that he goes high as mango runs out.
gets called out for repeated rhythmic jumps with shine. Just takes ledge.
5:54 gets pounded out of sakurai getup option. Mango sucks from ledge holy crap.
Free rest on upsmash OoS but maybe wouldn’t have killed. A little odd that the grab whiffed because that was super deep.
tight WDs right to get out of kill setups
5:38 fsmash after high jumps forces puff to cross up which he then uptilts, interesting
shitty shitty FH fair whiffs, grab upthrow rest.

* extended edgeguard enabled by high %
* CC grab upthrow rest after repeated FH aerials
* fox SHLs in front of puff, Upair > jab reset rest
* grab fair whiff, upthrow rest.

Stocks Lost:
* falls under fox, bair > upsmash
* whiffs in front of fox, jab upsmash

Mango stopped looking for and jumping out of most scuffles, which coupled with hbox being much much more aggressive and grounded the first 2 stocks this game got him tripped up and grabbed. A couple good plays off the platforms turned into a strong lead. Once up, hbox committed really hard to negating the significance of fox’s stage control and mango was once more A-OK with that all the way to the end. I don’t think he’s accustomed to not knowing when hbox will commit to something or (via hbox making a home on the side of the stage) that knowledge not being as valuable in the long run. Needs to update his tactics in a huge way.

- - - - -

Pretty interesting set to see how hbox softens fox’s advantages here. Punish game can still be improved which is cool because as shown in BF game esp it’s already really good. Neutral is interesting because I’d have to look more specifically and think hard to find lasting improvements. Not that they don’t exist, it’s just that they’re very subtle. It’d be nice to better work out WHEN to jump. A 3rd of a second is the difference between being a stock up and a stock down sometimes. Would be good to make those spots more lenient via identification.
Hbox’s positioning in general should be looked it carefully. Made the biggest difference in this set other than his solid as steel followups. I want to watch the set where leffen bopppppps him and see what hbox does differently from here and esp what leffen does differently from mango. Some of this is clearly fallible but not all of it.