Math for Ninjas: Learning to think about risk and reward

[This essay was funded by my generous Patreon supporters. If you liked this and want to see more, please consider joining the crew!]

This one’s for the NorCal Ninjas! While the advice in this essay is applicable to anyone playing fighting games in general, it’s going to focus on Chipp in Rev2.

1bin is so ninja that we couldn’t Find Him for the photo.

If you’ve stuck with Chipp past the early highs of “oh my god, I can do anything” and survived the early lows of “oh my god, I died off a single counter hit”, congratulations! You survived the Chuunin Exam and you’re ready for anything.

By now, you’re probably comfortable with most of Chipp’s main tools, you can do a couple staple combos, and you’ve probably got a special tricky setup or two to your name to pull out if you’re trying to close out a tournament set. As you get better, you will learn how to use more tools, do more combos, and hit more tricky setups. You’ll collect a lot of stuff from higher-level players along the way, and use their work as inspiration to fuel your own discovery. It’s a lot of fun.

You will also almost certainly run into some very dramatic range in your tournament runs. You’re going to beat some people you didn’t think you could beat, and you’re going to lose to some people you thought you were better than. This is because Chipp is an incredibly volatile character, and sometimes shit happens. If you’re looking for competitive success, the thing you’re always going to have to deal with is managing that volatility strategically. And that means you’re going to have to do some math. (Also: I talked about this stuff a bit in my Chipp combo theory video, if you want to check that out.)

An example of risk/reward analysis

Chipps can get some pretty easy early wins by Just Doing Chipp Stuff because he’s fast and has good tools, but getting to competitive consistency requires that you understand the risk/reward calculations at play with those tools. When you enter tournaments, you’re often going to be playing against people whose habits you don’t already know, and being able to consistently math your way to the win is one of your most powerful tools to win games without requiring that you outplay your opponent with a genius read or whatever. Ideally, you don’t want to have to make a genius read every time you want to win in bracket. I’m honestly not very good at math, but I think it’s a lot easier and more reliable than mind-reading.

We’ll use this sample screenshot from a recent Rev2sday set as an example.

Not pictured: Chipp getting banished to the shadow realm shortly after this round.

I-No is in the corner, knocked down from an Alpha Blade, and I’m about to 22H teleport on top of her to continue the pressure with a jH that has an ambiguous crossup from this screen position, followed by whatever string I want. Why did I do this? Well, let’s work it out.

From this position, with an air alpha blade knockdown, I can apply meaty pressure with several Chipp tools — run up gamma blade, leaf grab, whatever. Since I-No is low on life, pretty much anything that successfully hits her will KO, so I don’t need to worry about calculating damage and can just focus on the thing that I think is most likely to work at minimal risk to me.

  • Gamma Blade is most likely going to get blocked, since there’s no real mix potential, and while it’ll give me an advantage on block, it also gives I-No a chance to escape if she can deal with the followup pressure string. I’d consider this a medium risk, because if I-No is alive, she’s dangerous.
  • Genrouzan (leaf grab) will kill if it works, but if I-No reacts to it, I might get airthrown or even comboed for big damage and a possible mixup. To me, this is high risk, because I’m betting it all on my opponent failing the reaction check, and if I’m wrong, they can steal that round right back.
  • 22H -> jH hits meaty, gives me an ambiguous left/right crossup, and lets me continue with pressure. That means it’s more likely to hit cleanly than Gamma Blade, harder to for I-No to defend against than Genrouzan, and even if I-No defends successfully, I’m still in a good position to take the round. Compared to the other two choices, this is the lowest-risk option.

As you can see, if you want to be consistent, 22H -> jH is the play for this situation — but it’s not always going to be the case every time you get that corner uncling alpha blade knockdown. That’s because the factors that define the context of the match — player history, matchup details, resource counts — all change the risk/reward math. For example:

  • Meter: If I-No had 25 meter in this situation, the risk/reward for 22H -> jH would change dramatically because Blitz Shield is a universal solution for ambiguous left/right mixups. In that case, Chipp might opt to go for 22H -> throw or 22H -> land 2K to beat the Blitz Shield, or run up Gamma Blade, which has lower stakes overall.
  • Decision History: If I-No had 25 meter in this situation, and had successfully Blitzed the 22H -> jH the last time they were here, then Chipp might be more inclined to go for that throw or 2K, thinking that the I-No player is going to Blitz that option until the Chipp demonstrates that they’re willing to bait it.
  • Health: If I-No had 25 meter, demonstrated that they were willing to Blitz the jH, and Chipp’s health bars were swapped so that I-No had the lead and Chipp was almost dead, Chipp might be more inclined to take a bigger risk with Genrouzan -> wall cling combo or even 5D, in the hopes that a big payoff would put him back in the game. Since 22H is usually the go-to play in this situation, doing something else might be enough to catch the I-No by surprise.

Eventually, every decision you make should factor in your risk/reward calculations. It doesn’t matter if they’re incomplete — tracking this stuff during a match is a tricky muscle to build up! But if you’re not thinking about this stuff while you’re playing, you’re going to lose a lot of rounds you didn’t need to. Fortunately, it’s not that hard to get started — you just have to start looking for it. (More on this later.)

Everything is risky when you don’t have a health bar

Oh my god

Now, all of this stuff is standard for every character in a fighting game, but I’m writing this for baby Chipps specifically, so let’s talk about a few Chipp-specific factors that affect his risk/reward math.

  1. Chipp’s health is very low, so he can’t afford to be wrong as often as other characters.
  2. Chipp’s base damage potential is fairly low, but it scales heavily as the Chipp player puts work into more combos (which require higher execution, situation-specific conversion experience, that kind of stuff).
  3. Chipp’s basic mixup tools and high speed is hard for many players to deal with at first, but once they learn some basic counterplay (blitzing crossup jH, punishing predictable teleports and alpha blades, etc.) they’ll fuck Chipp up if he goes for the same stuff too much.

All of this together leads to some pretty common patterns for new Chipp players. They’ll often start out very comfortable using tools that can be punished heavily on reaction, like 22H, Genrouzan, and Alpha Blade, which works until you find people who will defend those and punish with a full combo. Giving up clean hits isn’t a big deal at lower levels, but once you play people who are able to consistently punish with full combos, you’ll find you need to be more choosy in using the risky stuff. And if the baby Chipp is still pretty early in their combo development, they’ll have to win neutral and mix more often to compensate for their lower damage output, which gives the opponent more time to adapt to any patterns they’re showing.

As the Chipp player grows, they’ll get better at finding more and more variations to their setups and routes, which will let them cycle through more options during a set. They’ll get the counter-hit setups and corner-only combo routes down, so that their payoffs get bigger and scarier. And they’ll learn to appreciate the less flashy, lower-risk tools that let them handle business without putting themselves in harm’s way. Over time, the logic of Chipp’s risk/reward calculations get burned into their hands so deeply that they’ll feel the bad math instead of having to work it out consciously.

The trick is to realize that when you’re always doing the risky stuff that can get you killed, you’re actually more predictable than when you are using his speed to minimize risk by finding safe situations to poke, win neutral, and pressure. For example, think of all the times you use 22H to get in and get blown up; instead, you could use 22P to check if your opponent is trying to react to 22H (you’ll usually see someone jump back if this is the case). Try 22P from a safe distance and wait for them to jump; if they did, it’s because they’re looking for your 22H, and you’re generally pretty free to run up and 6P with minimal risk. Do this often enough and they’ll stop trying to react to your teleport startup — which is what lets you know that you’re clear to use 22H.

Eventually, you’ll be comfortable enough with Chipp’s tools that you’ll be able to go through entire games or even full sets without ever putting your opponent in the same situation or sequence twice, because you’ll just have variations on variations on deck and ready to go. The idea here is that you want to confound your opponent’s ability to react to or predict your stuff by giving them too many feints to react to (by using 22P and dash FD brake to bait reactions) and running through too many situations for them to be able to remember what you’ve used recently.

Sounds pretty sick, right? Man, Chipp is just cool as hell. Anyway:

OK so how the fuck do I do this


If all of that sounded overwhelming, don’t worry, it’s just another thing you’ll get better at with practice. Being able to do this kind of math on-the-fly is not easy! You’re going to have to free up the brain power to track stuff like resources and decision history effectively if you want to get a leg up on your opponent, and that’s asking a lot when you’re already playing a character that demands technical excellence and deep study to win. But there are a few easy things you can practice to get started.

  • Get used to keeping count of stuff. Count how many times your opponent does a reversal DP or Blitz, count how many times you use 22H or Genrouzan. As you get good at keeping track of this stuff, you’ll find that it’s easier to see the patterns in people’s behavior than you might think, and you’ll learn to sniff them out early on in a set.
  • Pay attention to critical resource points. You don’t need to be constantly checking in on all resources at every moment, but you should pay attention to critical thresholds. Note when the opponent has Burst and when they don’t. Note when the opponent goes from having <25 meter to 25+, because that’s a key inflection point where Blitz and YRC come online. At 50 meter, you have to worry about RCs, Dead Angle, and supers. For now, don’t worry about what happens at 50+ meter until the opponent shows you that you should worry about it.
  • Diversify your arsenal. If you go for the same oki off a midscreen throw every time, your opponent is going to have an easier time clearing their mental stack to focus on what’s next because they’ll be able to recognize the situation as soon as they take the throw. Yes, you will undoubtedly have some setups that are higher-value than others, but if you go to them every time, you’ll find that they’re less valuable when you need them the most. Rotate your options right and you’ll find that everything will just work more often.
  • Stop autopiloting. You probably already know where you turn your brain off and just do a sequence of things because they feel right, so I shouldn’t have to explain this in too much detail, but: assume that any player is going to be ready to deal with anything you show them after they see it once in a set. If you are not rotating your options appropriately — either by cycling through options in a situation to stay one step ahead of your opponent’s choices, or by taking different routes so you don’t recreate the same situation too often — then it’s just a matter of time until you get exploded and die.
  • Do good combos and don’t drop them. If you’re not paying attention to the health bar when you hit your opponent, you might be surprised by how much damage scaling can change the value of a hit. So pay attention to it. Go through all your combos and make sure you know which ones do more damage than the others. Look at the good combos that you see other Chipp players doing, and integrate them into your practice. Chipp can kill very quickly if he lands the right hits, but if he’s not doing the big damage stuff frequently, the math will inevitably catch up to him.

This all might sound hard. Probably because it is! You’ve got a foundation of character literacy to build off of, but this is when you start to actually build something up and make it yours. At first, it might seem like this kind of practice stifles the “I can do whatever I want!” whimsy, but over time you’ll find that you’ll actually end up doing a lot more stuff to make the character feel like your own than you would be if you were just out there doing shit and hoping it works without doing the diligence of figuring out if it’s actually worth it.

Now: Go be a ninja.

Thanks for reading! If you found this essay valuable and want to support my work, please do not hesitate to share it around on your social channels, follow me on Twitter, check out my Twitch stream (Mon-Thurs 830PM-1030PM PST), or join my Patreon.


-patrick miller




Game Designer. Learn to play Street Fighter:

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

Day 16 of July LeetCode Challenge

What Makes Something Fractal?

Liberal Arts Blog — The Pareto Distribution and the Pareto Chart — Life is Not Fair

Twin Paradox Puts Einstein’s Relativity at Risk?

Count Complete Tree Nodes

These are a few of my favorite (mathematical) things.

World of Combinatorics: Counting, Permutations, and Combinations

Quantum Computing — required Linear Algebra

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Patrick Miller

Patrick Miller

Game Designer. Learn to play Street Fighter:

More from Medium

Reflecting on personal health, burnout, and fighting games

Lost Words: Beyond The Page

Batman vs Abyss! | WEIRDO Reviews: Batman #119 *No Spoilers*