Important Carefully Ask PPMD about the Tiara Guy

[Dr Peepee]

I was wondering if you'd be willing to go into deeper explanation of the relationship between compound movement and intention theory. Or, perhaps provide some useful links that allow me to research the two topics myself.

My second question is this;
We've had very short discussions about practicing before, and you've delved into how you practice in short on your stream. I'm not sure if this has been brought up here yet, but I was wondering if you could elaborate some on "Going back to basics."

I understand that this entails things like practicing wavedashing, L-Canceling, Wavelands, generally using your moves and hitboxes. However, there is something to be said about the structure of one's practice and analysis that I'm not sure has been touched upon, and I would literally pay to learn some more about how you structure your personal practice and analysis.

Mmmm....last time I taught someone that I messed up their understanding of the game. Suffice to say compound movement is what you use to mask intent and also give you reactable setups.

So, for practice, I often go for effectiveness rates. If I'm not doing great wavedashes basically 97% of the time then I'm much more likely to mess it up in tournament. So I do WDs for 2-5 minutes and note my effectiveness rate in a notepad file with any extra info I want to include. That way I know what to think about going into the next practice session and don't repeat mistakes.

Don't waste time doing mindless actions. It makes everything worse.

You can also do stuff for like edgeguards, where you practice doing certain movements to trick the opponent(like in neutral) and end up covering the option as you know the timing in your head. Then you try it in matches and get feedback on it through the game. That type of stuff is pretty useful for a lot of areas.

Examples that come to my mind are:

Fox v Jigglypuff:
I think since Hungrybox's "ascension" a lot of development has happened here, maybe the matchup is the one which developed the most in the last 2 years (it is still relatively unexplored compared to many others, but it was way worse before that). We see multiple different approaches from different Foxes, and several of them do very well (after all, it's a good matchup for Fox). The concepts that looked most convincing to me abused Fox's clearly superior vertical movement, not trying to beat the bair horizontally any more because it is a mixup that is often disadvantageous. Also there are a lot of different forms of positional advantages, and seemingly neutral looking gamestates can be favoured for one side, and players have become better at recognizing this. To give an example, Armada often allowed Jigglypuff to short hop or full jump towards him from across the stage in the past, which leads to Jigglypuff being able to force Fox into the corner with the bair threat zone, and even this is not always safe if Jigglypuff goes for the read with pound. In more recent sets he will use fj nair to challenge the jump by using it preemptively, and if you fall down on Jiggs with it you beat bair. Alternatively, he will only retreat as much as needed in this specific moment and full jumps out after the landing, so bair can't hit anymore and grab not yet.

Plup:
Plup innovated a lot recently. He created a few blueprint sets for Sheik vs Jigglypuff which has never been played at a this level before (although Shroomed had some close sets against Hungrybox). In every matchup, his skirmishing is very refined, to the point where I think it should be considered a big leap, but I haven't looked into it enough to be able to tell what exactly makes it so good.

ASDI down/Crouch-cancelling:
Just in general, but especially Falcon players, because the character is good at it, and he needs it. Not sure if all of this is new, but at least it is applied much more. I've heard of players holding c-stick down during dash dancing when at low %s to option select it every time, which might be new. I think there is also a lot of potential for CC even when the resulting frame advantage isn't enough for a punish, because they will either be at risk of getting hit or will have to commit to something that could get whiffpunished, but I haven't really seen that enough to count it.

Defensive maneouvres:
next to SDI which you mentioned, slide-off DI has improved a lot (against upward attacks at low %s, in platform techchases, against low angle attacks at higher %s), as has teching when recovering.

Gameplans:
In terms of gameplans, in a few matchups (non-dittos) where it was common to see both players dashdancing and mixing in some moves of their neutral game arsenal here and there, this kind of neutral game gameplay can be advantageous for one character, often the faster one at higher range and the one with the more fearsome neutral game tools (like Marth, Sheik with grab, grab/dash attack) at mid-range. To avoid being in mid-range too long, fast characters have shifted away from DD+SHFFL gameplans in many cases, instead trying to close the distance asap. Fox players now often just dash in completely to avoid standing in these threat zones for too long, while using some other option (like a dash in->crossup full jump) to be able to punish a defensive hitbox. I'm not sure how much of an innovation this is, because maneouvres like running shine definitely exist for ages, but from my point of view the understanding of the neutral game has changed in some way. The same thing can be seen with Captain Falcon, although it is usually combined with aerials and often the hitbox s just "sprayed" to cover one or multiple dash patterns, preferably dash away, which often acted as a get-out-of-jail-free card in the past.

I have to admit though that Falco and Marth development definitely suffered from your absence

Please to correct me on anything if anything is wrong, I'm not 100% sure about a lot of this as they are mostly own observations that could fall prey to all kinds of errors and biases...

Fox Vs Jigglypuff: Hungrybox is only now beginning to do some of the weaving Mango did a long time ago and it's much worse overall. I guess it's an overall advancement of the meta but I don't think anything new or him pushing his boundaries are happening. I do agree the Fox meta has advanced here but I believe it primarily did so in 2014/2015. The rest of what you describe is the Hbox vs Armada matchup and minor adjustments imo.

Plup: I think he did a ton in the past couple years, but I also think that this advancement happened earlier on in that time. He was doing well against Hbox a while ago if you recall, so his recent success is more refinement/him getting better/other stuff than new things from what I've seen. Refinement can obviously be something new to add and worthy of saying it advanced the meta, but I don't think that's what has happened for it since much of Plup's initial explosion.

ASDI down/CC: Holding down on platforms happened early last year or the year before I don't remember. Since then it's just people getting marginally better at CC and such which hasn't seemed to make a huge change at the top level yet. You mention Falcon as your example, but I don't really see S2J or None abuse CC much at all, and even Wizzy who is someone who would abuse it seems to not opt for it/gets surprised when he could have CC'd fairly often. This is because it's hard for people to want to come in on Falcon due to DD grab and they can just beat out his aerial hitbox if they space anyway.

Slide-Offs: I don't think I've seen that too much but it's definitely an area that could have a lot of advancement. If that becomes more frequent at the top then I'd agree.

Those neutral changes you mentioned also occurred in 2015 at the top level, and for the Fox part in particular it has been solved in ways like that and many others for a decade. Yes of course new approaches have occurred, but they're borne out of small refinements or "let me change this one detail to beat this habit of my opponent."

 

[Chesstiger2612]

Fox Vs Jigglypuff: Hungrybox is only now beginning to do some of the weaving Mango did a long time ago and it's much worse overall. I guess it's an overall advancement of the meta but I don't think anything new or him pushing his boundaries are happening. I do agree the Fox meta has advanced here but I believe it primarily did so in 2014/2015. The rest of what you describe is the Hbox vs Armada matchup and minor adjustments imo.

Plup: I think he did a ton in the past couple years, but I also think that this advancement happened earlier on in that time. He was doing well against Hbox a while ago if you recall, so his recent success is more refinement/him getting better/other stuff than new things from what I've seen. Refinement can obviously be something new to add and worthy of saying it advanced the meta, but I don't think that's what has happened for it since much of Plup's initial explosion.

ASDI down/CC: Holding down on platforms happened early last year or the year before I don't remember. Since then it's just people getting marginally better at CC and such which hasn't seemed to make a huge change at the top level yet. You mention Falcon as your example, but I don't really see S2J or None abuse CC much at all, and even Wizzy who is someone who would abuse it seems to not opt for it/gets surprised when he could have CC'd fairly often. This is because it's hard for people to want to come in on Falcon due to DD grab and they can just beat out his aerial hitbox if they space anyway.

Slide-Offs: I don't think I've seen that too much but it's definitely an area that could have a lot of advancement. If that becomes more frequent at the top then I'd agree.

Those neutral changes you mentioned also occurred in 2015 at the top level, and for the Fox part in particular it has been solved in ways like that and many others for a decade. Yes of course new approaches have occurred, but they're borne out of small refinements or "let me change this one detail to beat this habit of my opponent."

While I can agree with most of this, I disagree on Hungrybox. While the drift patterns he uses aren't new, many other aspects of his game are. Drill hasn't been used nearly as effectively before, he uses it to combo into grab or up-smash, and even as shieldpoke or in edgeguards from time to time.
The platform neutral game has been an advancement, keeping his location ambiguous with WDs on platforms and spending enough time on platforms to make anti-air attacks to catch him leaving the platform risky. The platform neutral game allows him to be relatively safe (compared to jumping from the ground) during the peak of Jigglypuff's jump during which she is most vulnerable, leaving interactions with the opponent for the stronger window of the jump, the low height descending part.
The wavedashs on ground are strong for similar reasons, lowering the ratio of jumping in that situation and therefore making a challenge (which would need a commitment by then) less profitable for Fox.

I am also always careful about differentiating between refinement and innovation, as it is often a matter of perspective. Obviously, there are cases where known theory previously hindered by tech skill consistency limitations is set into action, like the waveshine across the stage.
In many other cases though, new metagame needs new concepts in which it has to be interpreted, and looking through it with other lenses makes it look like an accumulation of improvements in random micro-situation all across the game.
To give an exaggerated example, one could say how dashdancing is much of an innovation, using dash forward, a known tool which was already used to move closer to the opponent and attack them, and dash back, also known and previously used for escaping attacks, so doing these in succession is just one application of techniques known since the release of the game, albeit a useful one. The simile is just to point out the principle, obviously dashdancing is way more important than anything developed in the last 2 years, and isn't a very "concealed" advancement.
It might be that theorycrafters just haven't managed to condense recent improvements in micro-situations into more general concepts which can then be seen as metagame advancements. So to me any top player improving (apart from techskill and imitation) is considered a potential metagame advancement, which, depending on the case, I might not understand at this point of time.

 

[Dr Peepee]

If Hbox hadn't started doing the platform and WD stuff a couple years ago I would agree with that part. Drill he used further back than that. He does use it a tiny bit for edgeguards but he doesn't do it enough/well enough for it to be a strong advancement I believe.

And I think for your last part I see top players improving and worsening at different things over the past couple years, so even if you see some progress they make you may not be noticing their mistakes.

 

[Chesstiger2612]

Yeah I guess I underrated how much top players knew/did 2 years ago, it's sometimes difficult to judge from watching a few tournament sets only, per player-matchup combination. I'll have to trust you on this one.

With worsening, do you mainly refer to players falling out of shape and unlearning matchups by not playing them often enough, or do you think that there is also worsening happening as a result of misguided improvement attempts?

 

[Dr Peepee]

Ah I really would rather not say. I'm sorry for not answering your question since I like to do that, but it's not a good idea in this case.

 

[Chesstiger2612]

I respect that, I can see why that question would put you into a weird spot.

 

[Andu]

Since we last spoke, I've been investing some time into researching practice methods from other fields of competition. One of the methods that caught my eye the most was Shadow Boxing, in its various forms and applications.

I remember you mentioning this being something you do, or at the very least something similar that you've applied to your practice for Melee, and I was wondering how you use it and for what purposes.

You see. The thing I'm struggling with the most is learning how to learn. In fact, I may be a little too hung up on it, but nonetheless. I'm not looking for the be all end all answer, but the fact that I've been playing this game for about 3-4 years at this point, and I'm only just now opening my eyes, and understanding some of what I need to do to get better, is straight up terrifying.

I love this game. I love what it makes me do for myself, but I'm tired of wasting my time, and want to really work toward efficiently getting better.

Man, did that ever take a different direction than I thought. Oh well.

 

[Blatant J]

Is Marth's F-tilt more useful in certain matchups?

 

[DCW]

PP, I thought you might be interested in this article on the difference between two kinds of play in Melee:

http://alexspuffstuff.blogspot.com/2017/01/gto-and-exploitive-play.html

The author calls these two approaches Game Theory Optimal (GTO) play and exploitative play. His description of GTO play reminds me of Umbreon Mow's emphasis on not taking any unnecessary risks and refusing to play the player vs player approach, but I don't know whether they're really that similar... Anyway, I thought you (and others) might find it interesting.

 

[Chesstiger2612]

Game theory applications in Melee are very interesting. I think the distinction between GTO and exploitative play is an important one and you can make good arguments for both. I suggest to divide exploitative play into 2 subcategories, non-gimmicky and gimmicky.

As example, say both players have the options A-E. A>B>C>A (like RPS), D<A,B,C, E<A,B,C and E>>D (high reward). Choosing D is obviously a bad option (called dominated strategy in game theory), and choosing E would only make sense if the opponent chooses D, so it is banking on a mistake of the opponent, and therefore is gimmicky play.

Not going 1/3 A, 1/3 B, 1/3 C like GTO would suggest, but instead maybe 60% A, 20% B, 20% C scores even against GTO play (like every strategy that only contains A, B and C) and is therefore non-gimmicky.

My philosophy of play would always advise against gimmicky play, but non-gimmicky exploitative play is a different thing. In order to exploit it, you have to put yourself at a risk of getting read, so I don't think it is bad. Using the same non-gimmicky exploitative strategies over and over, however, is bad because players will be able to adapt to them, or prepare for them with match analysis.

As we enter the realm of reads, psychology comes into play. I think it could be a reasonable model to think of a player as having an initial gameplan containing the strategies used for the respective situations, and also having some form of adaptation, so results of previous interactions influence the strategy in future interactions.

From a gametheoretical perspective, making a read means making an assumption about the opponent's mixed strategy in that situation and determining the option that does best against it. Let's take a simple RPS as example: From experience, player 1 assumes that player 2 has a slight habit towards scissors, and estimates the probabilities of player 2 to be 30% R, 30% P, 40% S. Thus it would be, according to his hypothesis, best to use 100% R (note that making a read by making a hypothesis about the opponent's mixed strategy always leads to one single option being best, or at least not worse than mixed strategies, if the hypothesis is correct).

This doesn't mean player 1 will from now on only use rock, but it means that him, having made a read, will use rock this time for sure. Let's say player 2 also chooses rock. Now player 1 has to figure out what to do next time. He might think that his assumption was initially correct, and player 2 has no reason to adapt because the last result was a draw, in which case he would use rock again. He could also think that he was wrong and that player 2 uses rock more often, in which case he would use paper next. Another possible hypothesis would be that the paper probability increases since it would have won last time, which might lead to a scissors choice.

Either way, player 1, if he attempts to read, will create a new hypothesis about the mixed strategy of player 2 and choose his option accordingly.

Alternatively he could go choose randomly between rock, paper and scissors, which would be the GTO way.

There are several factors that play into whether attempt to read or not, like how adaptive the opponent is or how much attention the creation of the hypothesis of the opponent's strategy could take away from your execution, but I will not go into detail here.

About creating the hypothesis, from a gametheoretical standpoint the last result has no effect on the player's next choice, so gaining knowledge about this part has to be done with psychological and empirical means. For that, one has to figure out patterns in the opponent's choice that have a predictive nature. Again, I will not go into detail here, but a simple example for such a pattern would be repetition (favouring one option in a specific scenario independent of other factors), which I think holds true for most if not all players.

 

[Dr Peepee]

Since we last spoke, I've been investing some time into researching practice methods from other fields of competition. One of the methods that caught my eye the most was Shadow Boxing, in its various forms and applications.

I remember you mentioning this being something you do, or at the very least something similar that you've applied to your practice for Melee, and I was wondering how you use it and for what purposes.

You see. The thing I'm struggling with the most is learning how to learn. In fact, I may be a little too hung up on it, but nonetheless. I'm not looking for the be all end all answer, but the fact that I've been playing this game for about 3-4 years at this point, and I'm only just now opening my eyes, and understanding some of what I need to do to get better, is straight up terrifying.

I love this game. I love what it makes me do for myself, but I'm tired of wasting my time, and want to really work toward efficiently getting better.

Man, did that ever take a different direction than I thought. Oh well.

I use shadowboxing to make my training include fighting another person by myself. Once you have an idea of how your actions influence people(analysis and that intent theory I discussed recently go a long way here) you can start developing some setups that give you some choice reactions you can get faster at. When you shadowbox you speed up these choice reactions and also create new decisions entirely, if you're flowing, by imagining an opponent responding to every action you make and you having to adjust on the spot as well as the next interaction. In other words it is adaptation practice, but it only works with understanding.

Is Marth's F-tilt more useful in certain matchups?

I like Ftilt vs jumping Sheik sometimes(wd ftilt) and kinda vs Puff.

PP, I thought you might be interested in this article on the difference between two kinds of play in Melee:

http://alexspuffstuff.blogspot.com/2017/01/gto-and-exploitive-play.html

The author calls these two approaches Game Theory Optimal (GTO) play and exploitative play. His description of GTO play reminds me of Umbreon Mow's emphasis on not taking any unnecessary risks and refusing to play the player vs player approach, but I don't know whether they're really that similar... Anyway, I thought you (and others) might find it interesting.

So my first thought is realizing that this is another example of an age-old dichotomy. The dichotomy is composed of two groups. The first one is a perfectionist school of thought(which the writer seems to belong to) that says there is ALWAYS a best option and best strategy and best answer in every situation. The other school of thought is for more intuitive players, and they are the ones who want to try unconventional things and come up with quirky options that have niche effective uses and also believe there is inherent nuance in every or most every situation that simply can't account for all human factors involved. These two groups tend to not like each other.

I used to be firmly in the perfectionist camp but broke rank occasionally to explore nuance. Now I'm more of a hybrid where I think everyone needs a solid fundamental understanding of correct plays in certain situations and general game sense, but I also realize that optimal is different for different people after that foundation is made. People are going to be better at different things, and forcing them all into one mold, as perfectionists do, will only stifle those players' growths.

 

[LSSH | fish]

Are there any downsides to Marth's neutral game that Samus can use to put the game in her favour? I'm particularily struggling dealing with well spaced fair and uair juggles and I'd like to hear what a professional thinks about the neutral game for that matchup in particular.

 

[capusa27]

Hi Dr Peepee

Recently, I was watching DruggedFox's analysis of you and Axe at Summit. One of the most talked about tactics was crouch cancelling and how Axe does a lot of unsafe attacks which Marths don't punish because they do not know that they can punish Pikachu in those instances. DruggedFox made it seem like you don't need to use F-air and D-tilt hardly at all in the matchup.

What do you see in the matchup that makes you think F-air and D-tilt are necessities? Is it just easier to hit Pikachu out of his attacks, or does Pikachu not have any way to attack at certain F-air and D-tilt timings and fadebacks besides being in shield vs down tilt?

Do you think that Marth should move closer to Pikachu at certain ranges since you have the threat of D-Tilt and you can shield grab/crouch cancel down tilt and shield grab down smash? Also, moving forward takes away space so that when Pikachu N-airs, a wavedash out of shield easily puts you outside of N-air range.

Thanks. Here is the video link that I mentioned: https://www.youtube.com/watch?v=2b5h02YAJw8&t=6904s

 

[Dr Peepee]

Are there any downsides to Marth's neutral game that Samus can use to put the game in her favour? I'm particularily struggling dealing with well spaced fair and uair juggles and I'd like to hear what a professional thinks about the neutral game for that matchup in particular.

If you can get him to swing early or force a late swing and then cc/beat his next option you can be in good shape.

It's a fairly bad matchup overall though so you'll have to focus on outplays.

Hi Dr Peepee

Recently, I was watching DruggedFox's analysis of you and Axe at Summit. One of the most talked about tactics was crouch cancelling and how Axe does a lot of unsafe attacks which Marths don't punish because they do not know that they can punish Pikachu in those instances. DruggedFox made it seem like you don't need to use F-air and D-tilt hardly at all in the matchup.

What do you see in the matchup that makes you think F-air and D-tilt are necessities? Is it just easier to hit Pikachu out of his attacks, or does Pikachu not have any way to attack at certain F-air and D-tilt timings and fadebacks besides being in shield vs down tilt?

Do you think that Marth should move closer to Pikachu at certain ranges since you have the threat of D-Tilt and you can shield grab/crouch cancel down tilt and shield grab down smash? Also, moving forward takes away space so that when Pikachu N-airs, a wavedash out of shield easily puts you outside of N-air range.

Thanks. Here is the video link that I mentioned: https://www.youtube.com/watch?v=2b5h02YAJw8&t=6904s

Dtilt forces Pikachu to jump and Fair beats pretty much all of Pika's options. I don't think I would disagree with Druggedfox about the necessity of CC in the matchup, but I would say that Pikachu can fake a ton and if you happen to hold down during one of those times then Pikachu can gain a temporary mobility advantage which can still be helpful. Really though I think Dtilt/Fair and CC together are the answers here.

 

[Taytertot]

PP, I thought you might be interested in this article on the difference between two kinds of play in Melee:

http://alexspuffstuff.blogspot.com/2017/01/gto-and-exploitive-play.html

The author calls these two approaches Game Theory Optimal (GTO) play and exploitative play. His description of GTO play reminds me of Umbreon Mow's emphasis on not taking any unnecessary risks and refusing to play the player vs player approach, but I don't know whether they're really that similar... Anyway, I thought you (and others) might find it interesting.

i found this quite interesting so i thought id add my two cents about it.

i think the first thing to acknowledge here is that the GTO examples given are going to break even using GTO if the odds are 50/50. now if we look solely at neutral game, then we have to realize in smash that neutral game isnt always even between any two characters. some characters theoretically win neutral more so in a given mu (an example being marth vs peach). disregarding whether or not the other character has a better punish game or anything else like that, If we assume that the player who's using a GTO strategy has the game knowledge to know what tools gives their character an advantage in neutral and randomly picking between the best ones (i.e. if marth's best neutral tools in an mu are fair, dtilt and grab and picks randomly between those three instead of randomly picking dair in neutral) then id lean towards assuming that they'd probably win the vast majority of the time in that mu. this is because the GTO isnt going to statistically break even amongst neutral game interactions. I think this is somewhat where we see umbreon's opinion stand in smash.

On the other side of the spectrum, if we again disregard all aspects of gameplay that isnt neutral game, then a character who should theoretically lose neutral more often then not (by that i mean doesnt have as many effective neutral game tools to deal with a specific mu) is going to continue to statistically lose more often then not if they stick to a GTO strategy. In which case, it seems to me ideal that this given player attempts to play with at least a slight amount of focus on exploitative strategies if they hope to statistically win in a specific mu.

 

[Chesstiger2612]

i found this quite interesting so i thought id add my two cents about it.

i think the first thing to acknowledge here is that the GTO examples given are going to break even using GTO if the odds are 50/50. now if we look solely at neutral game, then we have to realize in smash that neutral game isnt always even between any two characters. some characters theoretically win neutral more so in a given mu (an example being marth vs peach). disregarding whether or not the other character has a better punish game or anything else like that, If we assume that the player who's using a GTO strategy has the game knowledge to know what tools gives their character an advantage in neutral and randomly picking between the best ones (i.e. if marth's best neutral tools in an mu are fair, dtilt and grab and picks randomly between those three instead of randomly picking dair in neutral) then id lean towards assuming that they'd probably win the vast majority of the time in that mu. this is because the GTO isnt going to statistically break even amongst neutral game interactions. I think this is somewhat where we see umbreon's opinion stand in smash.

On the other side of the spectrum, if we again disregard all aspects of gameplay that isnt neutral game, then a character who should theoretically lose neutral more often then not (by that i mean doesnt have as many effective neutral game tools to deal with a specific mu) is going to continue to statistically lose more often then not if they stick to a GTO strategy. In which case, it seems to me ideal that this given player attempts to play with at least a slight amount of focus on exploitative strategies if they hope to statistically win in a specific mu.

Agree on the first part.

About the second part, I don't think it is bad at all to only play for GTO in a bad matchup, as long as it is not a terrible matchup. It takes a firm understanding of the game to have a sense of what the GTO strategy is for even one situation, and both players will not be able to implement a GTO strategy to that degree. If you are the better player, you can win by knowing what the GTO strategy is better than them even in losing matchups. It comes down to personal preference whether you want to outplay your opponent by being more accurate or by reading them. In some situations, for example playing against a player that chooses very good standalone options, but in a very predictable way, I agree.

 

[Leeyam]

Hi guys

Watching Armada's latest video on Plup vs Leffen, he talks about how one mistake Plup (with sheik) consistently does is ftilt when the other character is not a high enough % to fall over from it, since it does not put the other character in any bad position and they can just hit back.

I was thinking how applicable this would be as a law for Marth. In a lot of situations, particularly when other characters are approaching me by jumping in, I feel a need to defend myself with for an example a sh fair, which does not really do anything to them at low % and puts both of us in an awkward position. Should I rather focus on responding with other options such as more CC grabs?

 

[Taytertot]

Agree on the first part.

About the second part, I don't think it is bad at all to only play for GTO in a bad matchup, as long as it is not a terrible matchup. It takes a firm understanding of the game to have a sense of what the GTO strategy is for even one situation, and both players will not be able to implement a GTO strategy to that degree. If you are the better player, you can win by knowing what the GTO strategy is better than them even in losing matchups. It comes down to personal preference whether you want to outplay your opponent by being more accurate or by reading them. In some situations, for example playing against a player that chooses very good standalone options, but in a very predictable way, I agree.

i guess i assumed that both players had the same level of understand about the GTO strategy and the skill they had in implementing it. i wasnt really considering that they might be less knowledgeable, in which case i agree with you.

 

[Dr Peepee]

Playing the percent game is always wise. However Plup may have been expecting Leffen to not hold down knowing that Plup knew not to do it, or Plup simply thought Leffen would jump and Ftilt is good vs jumps. I didn't watch the videos but it's possible there is more nuance here. That said, Marth should also be very strict about percents and tipper his Fairs/Dtilts and get grabs at lower percents then bring out more moves at mid and high percents.

 

[Doof_]

The other day a top player in my region said my down-tilts in the Marth ditto were unsafe, so I decided to watch your set with Mew2King at Smash@Xanadu in 2013 and record every down-tilt you did. As to help with giving me an idea on when I should be incorporating down-tilt instead than just doing it randomly and hoping they run into it. I plan on doing more of your Marth dittos you've played, but I was just wondering if you had any constructive criticism on the way I am going about this. Here is what I have so far https://docs.google.com/document/d/1_rnkaf5E5-MOjZT3d2j0Csh9WvM4de32C_2oLTydlCw/edit?usp=sharing

 

[HolidayMaker]

This of course can apply more broadly to other matchups but I'm asking this more specifically about Marth v Sheik.

I notice you use rising fair more/better than other Marths in this matchup. Am I correct in noticing you start going for it around 45-50 percent? The other question I have is that I know you say to start using rising fair to call out an opponent's patterns rather than just doing it randomly. What patterns in particular should I be watching for? Timing when they move OoS? When they go in? Defensive options?

 

[Doof_]

This of course can apply more broadly to other matchups but I'm asking this more specifically about Marth v Sheik.

I notice you use rising fair more/better than other Marths in this matchup. Am I correct in noticing you start going for it around 45-50 percent? The other question I have is that I know you say to start using rising fair to call out an opponent's patterns rather than just doing it randomly. What patterns in particular should I be watching for? Timing when they move OoS? When they go in? Defensive options?

He does it then because before then Sheik can crouch cancel it and get a grab which could be your stock if they're good enough. I think you're right with it being a call out because when in shield not on a platform Sheik only has the grounded options of WD OoS into f-tilt or dash which you beat with down-tilt or dashdance grab (more risk), once you've cover that once or twice that forces the Shiek to have to try jump OoS with a n-air or to get out, that's when you call it out with a f-air so it becomes a mix-up. This is similar to corner pressure, so instead of a wall of hitboxes to try to stop any approach out, you have to react and place hitboxes as they try to get out. I'm sure there's a lot more to it, but that's what I understand about it.

 

[Dr Peepee]

The other day a top player in my region said my down-tilts in the Marth ditto were unsafe, so I decided to watch your set with Mew2King at Smash@Xanadu in 2013 and record every down-tilt you did. As to help with giving me an idea on when I should be incorporating down-tilt instead than just doing it randomly and hoping they run into it. I plan on doing more of your Marth dittos you've played, but I was just wondering if you had any constructive criticism on the way I am going about this. Here is what I have so far https://docs.google.com/document/d/1_rnkaf5E5-MOjZT3d2j0Csh9WvM4de32C_2oLTydlCw/edit?usp=sharing

The way you're going about it is okay, but you need to be asking questions about trends. If the Dtilt misses what good does it do? If it hits why does it hit? And eventually, you should be asking how is this action setting up for later Dtilt threats so he doesn't have to use the move in the first place? The more you can take series of situations and bring out patterns/rules from them the better your analysis will be.

This of course can apply more broadly to other matchups but I'm asking this more specifically about Marth v Sheik.

I notice you use rising fair more/better than other Marths in this matchup. Am I correct in noticing you start going for it around 45-50 percent? The other question I have is that I know you say to start using rising fair to call out an opponent's patterns rather than just doing it randomly. What patterns in particular should I be watching for? Timing when they move OoS? When they go in? Defensive options?

Unless I am positive they will jump then yes that sounds about right.

And consider the rising Fair in context with Marth's other tools. Dtilt makes the opponent want to jump and grab makes them want to change their attack timing/spacing. Putting these threats together with Fair you can control your opponent. So if you want to find a pattern that loses to Fair, sometimes the opponent will do it on their own(jump if you get near them) and sometimes you have to make them do it(they stay grounded until you Dtilt them a few times and then they start jumping when you get close). You can figure out how likely these things are by watching players of various skill levels play and try to figure out what each player is responding to from the other.

 

[Doof_]

The way you're going about it is okay, but you need to be asking questions about trends. If the Dtilt misses what good does it do? If it hits why does it hit? And eventually, you should be asking how is this action setting up for later Dtilt threats so he doesn't have to use the move in the first place? The more you can take series of situations and bring out patterns/rules from them the better your analysis will be.

So my end goal should be to fully understand how each Dtilt in neutral is connected to each other, sets up the opponents reaction to what may or may not be a Dtilt, and everything that is linked to a scenario where a Dtilt is a threat? Do you suggest that I should go a lot more in depth with each match rather than trying to do as many as possible with my current method? Is there any sets/matches of the Marth ditto that you believe would be the best to study?

 

[Dr Peepee]

Yeah and to get there you need to understand each tool individually. And yeah you'd be better going for depth and then getting a lot of videos eventually. As for sets just anything M2K and I have done since that S@X or at it would be fine.

 

[Doof_]

I made some changes to my notes on the Dreamland match, do you think I'm headed the right direction or should I go more in detail? It gets a bit fuzzy between what actually happens and what I am able to infer why it happens due to previous interactions.
https://docs.google.com/document/d/1_rnkaf5E5-MOjZT3d2j0Csh9WvM4de32C_2oLTydlCw/edit?usp=sharing

 

[Dr Peepee]

As long as you eventually begin incorporating patterns and matchup/player "facts" in your analysis then I'd say you're definitely on the right track.

 

[Doof_]

Thank you so much for the help! I'll try putting it all together sometime this week. By patterns do you mean in similar fashion to Squid's Falco laser guide where he talks about being able to get a read on how your opponent will react to each laser. Or in this case, each of Marth's options you commit to in neutral?

 

[Dr Peepee]

Patterns would be things like "given how a character is prone to attacking in this position, then I should assume that this position+action combination forces the opponent to want to come in. This is because..."

 

[Doof_]

Okay, that is a lot more clear on what to look for and think about while watching/playing. I'll be certain to put all of this knowledge to use. Hopefully I can accomplish sometime worthwhile by the end of the year. I honestly can't thank you enough for teaching me how to analyze infinity better.

 

[CyberZixx]

What is the best DI to avoid multiple uparis from fox?

 

[MusicMan]

What do I do if I'm above a spacie in midair?

Also what are your thought's on Marth's reaction tech chase game? Is it something worth exploring?

 

[lokt]

What options does marth have vs sheik overshooting grab? Like if I see her starting to run forward during my dash back, is there a way to intercept her?

I know that vs fox you can just grab him out of his run, but i haven't had much success with this vs sheik because she just dash attacks.

 

[Dr Peepee]

What is the best DI to avoid multiple uparis from fox?

Usually SDI up and to one side(if they're on one side of you then go the other way). You can practice it in 20xx.

What do I do if I'm above a spacie in midair?

Also what are your thought's on Marth's reaction tech chase game? Is it something worth exploring?

Mix using DJ and drift forward and back and FF and aerial and side B. I tend to mix using DJ when they could start to jump and hit me vs falling with an aerial if I think they will wait and hit them, but there are plenty of ways to do this and other options.

What options does marth have vs sheik overshooting grab? Like if I see her she starting to run forward during my dash back, is there a way to intercept her?

I know that vs fox you can just grab him out of his run, but i haven't had much success with this vs sheik because she just dash attacks.

Run forward and grab her, Dtilt in place is good in this matchup, retreating/pivot Fair if needed, WD back to avoid grab. Those are my go-tos. Just hold down when you input grab so you can ASDI the dash attack down, or just run forward and shield/retreating Fair to beat the approach.

 

[MyNeighborBurrito]

Sup PP, are there any good write-ups of what percents certain characters can crouch-cancel some of Swordguy's aerials (Tipper Fair, Nair)?

In a similar vein, do you think utilizing Marth's limited crouch-cancel could be a good option, or is it better instead to stay mobile with his neutral game and punish with grab/etc?

Thanks so much dude! Looking forward to the healthy return <3

 

[Dr Peepee]

Sup PP, are there any good write-ups of what percents certain characters can crouch-cancel some of Swordguy's aerials (Tipper Fair, Nair)?

In a similar vein, do you think utilizing Marth's limited crouch-cancel could be a good option, or is it better instead to stay mobile with his neutral game and punish with grab/etc?

Thanks so much dude! Looking forward to the healthy return <3

Nah I have to do testing myself for the matchups still and write it all down. I started it after Apex but didn't finish for obvious reasons.

Marth can abuse CC much more than he currently does, and you can definitely work it into mobile neutral. Obviously if you can avoid taking damage and getting a punish then do that, but that doesn't mean you should ignore such a good tool.

 

[bts.mongoose]

When doing analysis, is it equally important to make note of the feints that Marth players do as well as the actual approaches? For example, if I'm trying to examine the interaction that happens every time the Marth player approaches the Fox by running at him, should I be paying attention to all of Marth's forward dashes as well as his runs? As it stands, I'd mostly be looking at when the Marth runs forward and then uses a move, but I think I might be missing a crucial part of the puzzle by ignoring what the Fox does when he THINKS that Marth will approach.

 

[JFB (JurgaBurgaFlintines)]

So I've been working on my punish game with chaingrabs for the past few weeks and I feel like I have gotten to the point where I don't drop grabs/follow ups and I know what to do with every type of DI. Now I'm trying to find other things to work on, but I'm not sure what to do. Do you have any suggestions?

 

[Vista_]

Hey y'all, I'm new to marth and I was curious about the applications of short hop double fair. Outside of comboing floaties, what are its uses? Setting up an aerial wall to stop an approach?

 

[Dr Peepee]

When doing analysis, is it equally important to make note of the feints that Marth players do as well as the actual approaches? For example, if I'm trying to examine the interaction that happens every time the Marth player approaches the Fox by running at him, should I be paying attention to all of Marth's forward dashes as well as his runs? As it stands, I'd mostly be looking at when the Marth runs forward and then uses a move, but I think I might be missing a crucial part of the puzzle by ignoring what the Fox does when he THINKS that Marth will approach.

I'd suggest doing it the other way around. Look at when they actually commit, what happens, then look at whether they tried to condition for it beforehand or took advantage of the new conditioning after. That has been pretty solid in my book.

So I've been working on my punish game with chaingrabs for the past few weeks and I feel like I have gotten to the point where I don't drop grabs/follow ups and I know what to do with every type of DI. Now I'm trying to find other things to work on, but I'm not sure what to do. Do you have any suggestions?

Tech chasing with Fthrow/Dthrow on spacies/falcon/sheik, edge stuff like close to the edge WD grab or edge dash or haxdash, followups on FFers at different percents on side platforms, neutral tools that take advantage of center stage or corner pressure, etc

Hey y'all, I'm new to marth and I was curious about the applications of short hop double fair. Outside of comboing floaties, what are its uses? Setting up an aerial wall to stop an approach?

Mostly for walling but it isn't done too often. People usually single Fair.