• Welcome to Smashboards, the world's largest Super Smash Brothers community! Over 250,000 Smash Bros. fans from around the world have come to discuss these great games in over 19 million posts!

    You are currently viewing our boards as a visitor. Click here to sign up right now and start on your path in the Smash community!

Applying Frame Data to Create A Strong Mixup Game

Nefarious B

Smash Champion
Joined
Dec 13, 2008
Messages
2,002
Location
Frisco you know
I expect this to go very in depth, so if you are a new player or new to the character some of these concepts may not be helpful until you understand more basic things like spacing and hitboxes.

First I recommend reading TKD on Baiting, Interruptions, and Mixup Chains. It's an excellent essay that explores the general concepts that I will address.

Unlike many other high tier characters, ZSS does not have extremely abusable traits on most of the cast, nor does she have exceptional weaknesses in most matchups. Because of this, her approach to each matchup changes drastically. For this reason, an in depth understanding of frame data, and the ability to apply those details to matchups can help us understand what actually works.

So let's go into an example situation that as a ZSS player you will face very often.

ZSS is in the air facing her opponent at the peak of her SH, opponent is grounded, let's say Snake is the opponent.

--ZSS fast falls a uair at max range.
--Snake anticipates the aerial, sheilds, and after contact shield grabs
--ZSS dsmashes Snake during whiff frames

Another example from the same scenario

--ZSS fast fall airdodges, landing behind Snake
--Snake shields, anticipating an aerial
--Snake sees the airdodge and drops shield to punish landing with an ftilt
--ZSS utilts Snake during shield drop frames

Both of these mixups worked because the ZSS was one step ahead, and baited a likely action from Snake and was able to punish because of understanding frame data and spacing.

To quote TKD:
What I refer to as "baiting" is the action of provoking -or even just expecting- an opponent's action, be it a specific one -since you're expecting a specific action, it should either a very good one, one highly abused by the opponent, or one you anticipate thanks to "mindgames"- or a range of very likely options with a common counter (by one of your character's abilities, by avoiding and punishing, or whatever).
Snake has slow aerials and slow jumpstart, so he will generally not use those to punish out of shield. However, he makes up for this weakness through incredibly powerful and fast tilts, as well as a very damaging grab that he can use out of shield.

In the first example, you have made yourself appear vulnerable, baiting a grab which you can punish for tons of damage. In the second example, you freeze Snake in his shield with the potential of an aerial long enough to get back to the ground, and then interrupt his attempted punish with a fast ground attack. So we can see that the main concept behind baiting is to punish the opponent's punish.

So where does frame data come into play when analyzing how good these mixups actually are? As a ZSS player, you should know all of your basic frame data. When analyzing mixups for specific matchups, it helps to know things like the speed of your opponent's favorite OOS moves, and the areas they cover. This is what will allow you to bait and punish, or simply avoid those punishes.

Let's look at the two examples I gave from a frame data perspective.

In the first example:

I uair Snake's shield, which is -8 advantage. Snake has a number of options here. He could roll away, roll behind us, attempt a punish. This is where reading your opponent comes into play, giving you a better idea of what exactly they will do.

However, we know that Snake as a character tends to want to punish bad spacing on shield. Let's look at his typical punishes:

--Grab: 8 frames OOS, decent range, good reward
--Ftilt: 11 frames OOS, good range, very good reward
--Dash Attack: 12 frames OOS, very good range, medium reward

Because of its lower reward and slower speed, Snakes don't use DA nearly as much as the other two. So even though Snake could have beaten my uair to dsmash mixup easily by dash attacking, you can bet that he most likely wont choose this option.

Our dsmash is very slow on startup at 21 frames making it usable mostly only on reads and baits, and has amazing range. However, to get in range to hit Snake's shield with the uair, we cannot max range the dsmash. So we are gambling that Snake will go for the grab and not the ftilt, because through frame data we can easily see that ftilt will outspeed dsmash.

So even though this mixup worked in the example I gave, when thinking about it as a whole you can break it down simply:

If I uair then dsmash a shielding Snake, the mostly likely two outcomes are:
1. Snake whiffs a shield grab, I get the dsmash for upwards of 30%
2. Snake ftilts, I eat 21%

Knowing that Snake has a hefty weight and killing advantage on us in the matchup, we can conclude that this mixup should not become a staple because the risk/reward ratio is not high enough. If you notice that the Snake tends to shield grab a lot however, this mixup becomes much more valuable.

Lets do the same for the other example:

If I airdodge behind a shielding Snake and buffer a utilt the most likely outcomes are:
1. Snake drops shield, but I interrupt his punish with utilt for 11% and a possible juggle
2. Snake holds shield, I hit his shield with utilt, Snake can punish me heavily
3. Snake reacts quickly enough to drop shield and ftilt me before my utilt comes out, I eat 21%

So here we can see that this mixup has a pretty terrible risk/reward ratio, and thus should be used very rarely, or only if you spot a habit that you can punish consistently. In addition, knowing that ZSS's utilt is not safe on hit at low percents, you also could not use it at low percents. You have to take all of this into account when evaluating a mixup to see how effective it is.

If you look at a situation as a whole, like ZSS being at SH height against a Snake, and decide that most options from this position have a good or bad risk/reward ratio, that should tell you whether you want to be in that position in the first place.

So I know what you're thinking. "Man all this Nef guy does is give us examples of how bad our character is wtf". Here's an example of what I would consider a good mixup:

You're outside of ftilt range, Snake has just pulled a grenade.

--ZSS dsmashes, Snake shields
--Snake attempts to punish with ftilt OOS
--ZSS dtilts and interrupts the ftilt

Why is this mixup good? For starters, when Snake pulls a nade he enters a special state where he can only shield, jump, roll, sidestep, grab or toss the nade.

So we see that by pulling a nade, Snake has limited his own options. Our response should attempt to limit them further:

--Snake will not likely jump because ZSS will have an easy juggle attempt

--Snake might roll behind us. If we choose dsmash and he rolls behind, dsmash will end before his roll ends, allowing us to escape or attempt to punish his lag. Your dsmash has enough range that if your spacing is at max range he will be right in front or behind you after his roll, making for an easy utilt punish

--Snake might spotdodge. Dsmash will hit the dropped nade and explode it, which may or may not outlast Snake's spotdodge. At the very least, if we see the spotdodge we should have time to dtilt him in its end lag.

--Snake might roll away, in which case the situation is reset with ZSS gaining stage

--Snake cannot grab at this range

--Snake may shield. Seeing our dsmash startup and knowing that it will set off the nade, Snake is pretty likely to choose this option. However, because you know your frame data you know that dsmash on shield is -5, which is very good. Your dtilt is frame 5, which means that chaining from dsmash to dtilt is 10 frames, which means if Snake ftilts OOS immediately you will interrupt him, if he dash attacks you will interrupt him, and if he grabs he will whiff and be hit.

So we have Snake's most immediate punishment options all covered. If the Snake has little to no matchup experience (which can be very likely depending on your region) you can bet he will probably not know how safe dsmash is on shield, and thus your bait will probably work.

The possibility of punishment exists in almost every scenario however. If Snake holds shield after the dsmash he may get shield poked from the dtilt, but if not he may be able to punish you. However, when you look at all the options available in this situation and take into account that you probably know Snake's options better than he does, you can usually make a judgement call on how likely you will be to actually get punished.

Through this analysis of frame data and spacing you can determine that this is a very solid mixup. Looking at the situation itself, Snake pulling a nade at mid range, you can see that this situation is going to be pretty common. Thus you have a good answer to the situation that Snake can't do much about, which may limit his grenade usage to only at distances where you are far away enough to not dsmash him for it. This limits a strong part of his game to long range, where you can see it coming better.

After some discussion and feed back I'd like to get to the meat of what I want to discuss with this topic(there's more wtf?). If a situation like above occurs often and Snake is repeatedly not getting what he wants (dealing damage), you can be sure he will try to adapt to find a better answer to the scenario. So next we'll get to mixup chains, which becomes an everchanging game of adaption and counter adaptation
 

culexus・wau

Purchased premium only to change name ><
Joined
May 13, 2008
Messages
4,636
Location
Irvine CA
We need to talk about things like this more when we see eachother.

We both seem to be the people that value frame data and such very much ><

excellent read however.
 

#HBC | ZoZo

Shocodoro Blagshidect
Joined
Jan 12, 2009
Messages
9,800
Location
Land of Nether
This is pointless as brawl has way to many options.
What if I uair to shield? to spotdodge? to roll? to downB? blablabla
 

Nefarious B

Smash Champion
Joined
Dec 13, 2008
Messages
2,002
Location
Frisco you know
I think it'd be useful to know 2 or 3 of our best mixup strings for common situations, and not only that but know why they are good. Having that understanding of why strings are good is what allows you to be creative with your game IMO, but if people don't see the value in it I'll just write it down for myself
 

culexus・wau

Purchased premium only to change name ><
Joined
May 13, 2008
Messages
4,636
Location
Irvine CA
This is pointless as brawl has way to many options.
What if I uair to shield? to spotdodge? to roll? to downB? blablabla
you can easilly divide it into 5

shield

spotdodge

roll

[assume best punishment] attack

grab
 

culexus・wau

Purchased premium only to change name ><
Joined
May 13, 2008
Messages
4,636
Location
Irvine CA
***** thats 5.

thats almost the same as in a normal fighting game

have you ever fought a good millia?

I know you play guilty gear :|
 

Nefarious B

Smash Champion
Joined
Dec 13, 2008
Messages
2,002
Location
Frisco you know
Yeah WGF is like a 10 minute walk lol. If I can get a ride i'd like to go to the others so I won't be horribly rusty for WGF

And Xonar this game really doesn't have that many options. Like FWK said, you have your basic defensive options, and then you have attacks. The way you determine which attack would be best for the situation is based on startup, hitbox, safety, and number of hits, as well as movement options to space them.

The way I would have this is people would choose a scenario on their own, and analyze it thoroughly to determine what our best options are to counter the most or best of the opponent's options, and this thread would be a sort of database for how to deal with individual situations.

The thing is I don't think that is so unrealistic. I've been looking at competitive chess recently, and the thing is they have soooo many more options than this game yet they have analyzed that game down to where they know the best first 10 moves or more for any given opening. There's no way Brawl is even close to as complex of a game as that imo
 
Top Bottom