Hit The Ground Running is an article series by Jay “Viscant’ Snyder.
This week we’re going to talk about why waking up with buttons or coming out of blockstrings with buttons is a bigger part of Street Fighter 5 as opposed to past Street Fighter titles. Specifically we’re going to talk about jabs and LKs. In theory it would seem like this would be less a part of Street Fighter 5. With fewer light to medium and light to heavy links you get less damage out of a successful wakeup jab and in theory the crush counter system seems to be designed to give you a stronger punish on wakeup buttons than in previous games. And yet people still keep on mashing on wakeup. Mostly because it works.
Before we get started I’d like to link a frame chart I found on reddit a couple weeks ago.
This chart is great because it not only has all right numbers, if you scroll to the right you can see all the hitboxes. It’s a useful reference in general but for what we’re going to talk about in this week’s post, it’s particularly useful.
Why is wakeup jab good?
The simplest answer to the question is because it was a deliberate design decision to make wakeup jab good. I’ll explain. One of the hallmarks of a new game is that the community frequently tries to play the new game exactly like the previous games in the series before we fully get the hang of it. And in Street Fighter 4, the idea of wakeup buttons just wasn’t as big a thing. Yun was the king of wakeup buttons as Kazunoko showed off and Balrog would wake up with jab sometimes, and for some unknown reason Capcom decided to make Sakura wakeup HP a thing, but it just wasn’t as prevalent in that game.
But in Street Fighter 5, the game was designed to reduce the role of setplay and okizeme. A lot of setups that people are using in competitive play, wakeup jab isn’t just on the table, it’s the specific counter given by the game to beat that setup. I’ll give some examples.
Last week we talked about how bad Alex’s okizeme post power bomb was. To recall, the move has 15 frames of advantage on quick rise and Alex lands about a dash distance away. But the dash is 19 frames making wakeup jab the legitimate counter to Alex dashing in. As you start looking at the frame data of other characters more closely you start noticing other areas where Capcom is telling you that you’re supposed to wakeup jab. Take Chun-Li’s normal forward throw. 15 frames of advantage on quick rise and her dash is…exactly 15 frames. So if you have a 3 frame jab like Necalli for example, waking up with jab if you see a dash isn’t “scrubby”, it’s the exact right thing to do. It’s possible that she could do dash into EX spinning bird. But by “possible” I mean “ridiculously and incredibly unlikely to ever happen in a real match”. So taking that off the table, the absolute worst thing that can happen in that situation if she doesn’t have full meter is a trade.
I bring up the Chun-Li example because I play Chun-Li and with the game being out for about 2 months now, she’s been in my heavy rotation for almost that whole period of time. I’ve logged hundreds of games with this character. I should KNOW by now that throw, then dash is not a valid setup against other 3f jab characters. But we’re still in the period of the new game where we’re not playing the game that exists, we’re playing the game that we THINK exists. After 7 years of Street Fighter 4, my brain thinks “I got a throw, I deserve a setup”. But this is Street Fighter 5, I deserve nothing. Maybe you’re making the same kind of mistake and you’re starting to get used to using setups that just aren’t real. Check the frame data and make sure before going forward.
The most obvious solution to beating wakeup buttons would be to use a meaty setup. Brief terminology check, meaty in fighting games means that your attack is hitting on later active frames of that move but also on the first frame of the opponent’s wakeup forcing them to block or reversal out. Why do we call this “meaty”? I mean…your guess is as good as mine. We’ve been using this term for 25 years in the Street Fighter community and FGC at large. It didn’t make sense then. Doesn’t make sense now. But it’s the accepted terminology so we’re going with it. Be careful when coming up with terminology in fledgling communities, you could be stuck with it forever. Someone on alt.games.sf2 was hungry 25 years ago and we’re all suffering for it to this day.
Anyways back on topic. Meaty attacks have always been the traditional solution to people getting frisky on wakeup, but they aren’t quite as good in this game as in some of the more classic Street Fighter titles. Take another look at the frame data chart for your character, this time looking at the active frames data. On the document linked above that would be the second column from the left. The numbers here are probably smaller than you think with some characters having shockingly low numbers. Cammy has no grounded normals with over 4 active frames. The normal that SEEMS to be active the longest, the knee bullet (towards+HK) still only has 4 active frames total, 2 on the first knee, a gap, then 2 on the second knee. That’s pretty bad, but it could be worse. She could be Rashid with almost his entire grounded set being 2s and 3s, the only 4 being the crouching HP—2 hits of 2 active frames to be exact—and the down+v-skill (which is stretching the definition of “normal” but whatever, let’s be generous).
So if meaties seem intentionally weak or intentionally hard to do, there’s a reason for that. They were designed to be difficult to use. This is another deliberate design decision on Capcom’s part. There are a couple reasons for that. Moves with large amounts of active frames that aren’t hugely negative on block can be exploitable; if you time that type of move to hit on the last active frame, it might grant a strong setup afterwards. You look at moves that seem to be active for an abnormally long period of time like Nash’s bazooka knee (6 frames active) or Chun-Li’s crouching MP (9 frames active). Capcom designed these moves to be so negative in a normal situation that even timed to hit perfectly on the last frame, they’d still only get to +1 on block. Or Dhalsim crouching MK (medium slide). 11 frames active, another abnormally large number for Street Fighter 5. But it’s -8 normally so you’re only getting to +3, a good number to be sure, but not ridiculous. Deliberate design decision. Someone put a lot of thought into this.
The other reason is because they want as little overlap as possible between moves that work on quick rise and back roll. Almost all the moves that are active long enough to hit both either have some kind of a huge visual clue on them giving the player waking up a heads up to block or reverse, or the move is greatly disadvantaged on one wakeup option or the other. Take the Dhalsim medium slide we just talked about. If you hit that dead perfectly on the final frame you get to +3. So let’s say you made a setup where that move hits on the last frame on back roll. But if they quick rose, generally that same timed slide will be at best -2 (almost always it’ll be worse than that because distance changes the setup), meaning that by doing this “setup” you just frame trapped yourself. Same with a move like Chun-Li’s towards+HK, the yokusenkyaku. This is the button used in the standard setup that Chun players do after spinning bird kick or EX lightning legs (dash into towards+HK). It’s +3 or +4 on back roll (character hitbox specific) giving her a combo opportunity followup on hit and keeps her pressure going on block. But it’s -2 on block on quick rise. Plus the move has a huge visual clue meaning that on either option, it’s easy to reversal against on reaction. Once again, deliberate design decision.
And while this doesn’t apply to competitive play very much, it has to be said. Online play + meaties don’t get along very well. This game uses a rollback netcode meaning that if the connection isn’t perfect, it “stutters”. If you’re trying to visually time the meaty, sometimes the frame YOU think you’re on, isn’t the same one the game thinks it’s on. And since we’re dealing with only 2 or 3 frames active most of the time, this leads to problems. Take Vega’s crouching MK, the ideal meaty for him since it’s +2 on block and +6 on hit normally. It’s 7 frame startup with only 2 active frames. If you’re off by 1 frame on the back, it whiffs completely and loses to wakeup buttons. If you’re off by a few frames on the front end, it can also lose to crouching jab, which is a tremendously frustrating experience.
Creating better setups
So since it’s hard to create meaty setups on sight alone both online and in live play, let’s use the frame data linked above to create a better setup to beat a wakeup 3 frame jab. I was mentioning Cammy as having some lousy active frames so let’s use her as the example.
Step 1 is to figure out what move starts the setup. I’d hoped to use throws since throws can’t be back rolled but Cammy’s throws aren’t very good for this kind of an example. 9 frames of advantage on the forward throw and 15 frames of advantage on the back throw for quick rise. So that’s no fun. Instead let’s use the HK spiral arrow as our starting move. Many Cammy combos end with this, so it’s a good choice to start with. I see a lot of players try to follow this up with walk forward into knee bullet. I can see why they want to do that; knee bullet seems like the designated meaty attack but eyeballing it is difficult. Let’s make something that takes the potential for user error out of the equation. According to the frame data linked above, this move gives us 80 frames of advantage on no recovery, 21 frames on quick rise and 26 frames on back roll.
Now let’s kill off some of these frames. Luckily this move leaves you close and Cammy’s dash is pretty good. 16 frames on a dash forward. So now we have 64 frames left if they didn’t get up, 5 left on quick rise and 10 left on back roll. So let’s start with the fastest option, the quick rise. 5 frames left. Luckily she has a 5 frame move with 3 active frames with crouching MP, which is also a combo starter. 16+5, blackjack, pay me. 6, 7 or 8 would be fine too since we’re trying to beat a 3f move with the lowest priority (remember in Street Fighter 5, if arriving on the same frame heavies beat mediums beat lights), as would lower numbers with active frames but this is perfect, this number is exactly what I’m looking for.
But if you do that same action against a back roll, you’ll get hit. So let’s look again at the frame chart and see what our options are after the dash we’ve already committed to. Ideally I’d like to find a combo starter move but knee bullet is 12f startup, meaning that even though it can beat wakeup jab, it’s easily reversed and not a true meaty. And other combo starter moves like crouching MK are 7f startup with 2f active. The move’s active frames end 1 frame before the character’s back roll is finished. So that’s no good either unless you could manually count off 1 frame before hitting that button. But the whole point of creating a setup like this is so you don’t have to do that. So let’s keep looking. Standing MK is 9f with 3f active as is crouching HP (same). This means that it’ll work as a true meaty (9f+1 of the 3 active frames). Crouching HP will combo into critical art or v-trigger. Standing MK is better if I’m not planning on using either of those options. Not ideal but looking at the data, this is the best we can do without introducing the possibility for user error. It would appear that standing HK would be better. 9 frame startup, 4f active and it’s a crush counter move but hitbox wise it just doesn’t work out all the time, it misses quite a bit. So that’s why you should make sure to test things out extensively against multiple characters and not just rely on the numbers alone.
Similar process for the no recovery option except that with bigger numbers there’s more math involved and more possibilities to make an execution error since the math is assuming you do whatever action you’re taking on the first frame. Mistakes happen here (especially to me), so you want to use as few moves as possible to get to the desired spot. So in this case you can so dash forward, whiff low MP, then jump straight up with her 46 frame jump and come down with either HP or empty jump low. This is good against characters who have no reversal anti-air or bad reversal anti-air. Or you could do dash in, whiff low MP, whiff neutral standing HK, first frame knee bullet. If done properly you’ll hit on the second knee of the knee bullet and will beat 3f crouching jabs.
This is difficult at first since it presumes that you’ll both hit every action on the first frame you’re intending to do it on and that you’ll be able to tell the difference between each wakeup option quickly. Training mode is key here, you can set the dummy to wakeup with the fastest possible attack and set it to randomly select wakeup options to give yourself practice reacting to different actions on the fly.
Now that we’ve made a setup are we done? No! We’re just getting started. We made a successful setup here but now the goal is to find a counter to that setup. We’ve successfully beaten ONE wakeup option but there are others. How can you tweak the setup vs. players who really like quick rise and then wake up shoryuken? Dash into crouch block or dash into walk back (the shimmy) works here. Since low MP is a 5 frame move, people tend to react to Cammy’s body changing shape and can’t actually see the punch. And the shimmy will select to a block vs. a wakeup shoryuken, block immediate wakeup jab, walk away from the throw and a delayed jab, allowing you to punish the latter two options. How can you tweak the setup vs. people who you’ve trained to quick rise and block somewhat frequently on wakeup? Throw is 5 frames so you don’t have to delay much to get a throw off here. In fact with a kara cancel to kill a frame you can easily get a throw that beats both block and wakeup jab. So on and so forth. Now tweak these to work on back roll. Now tweak these to work on opponents who stay down. Now let’s find new setups that look different!
Beating wakeup jab is just the first step in developing a solid okizeme game. It may seem unfair that you have to go to this kind of trouble to stop someone from waking up just mashing but this is the game we play. Adapt or die.
That’s all for this week. Next week I’m going to go a little Buzzfeed on you, with the most common moves that players aren’t dealing with properly in the first 2 months of the game (and #6 will SHOCK you). See you then!