Two lists, and they disagree
A tier list question sounds like it has one answer. In this game two answers exist, and they only partly overlap.
- What players repeat. Four styles: Pendulum, Ghost Step, Peekaboo and Slick. Several players named them in the official Discord on 27 October, with no numbers attached.
- What the developer publishes. Catch, dodge and block rates per punch type, for 12 of the 19 styles recorded on this site. The other seven — Pendulum, Ghost Step, Soviet, Dizzy, Sway, Rope-A-Dope and High Guard — have no figures at all.
- Where they overlap. Peekaboo and Slick sit on both lists. The two most-repeated names in the ranked pool, Pendulum and Ghost Step, are also the two with nothing published to check them against.
What ranked players named on 26–27 October
The thread starts with a new player asking "what styles are good?" and converges inside six minutes: "pendulum ghost step peekaboo slick" — "those are the big 4 of ranked". In the same conversation Slick is picked over Soviet, a player describing themselves as a Soviet main writes "it gets countered by almost everything" and "its impossible to win close fights as soviet", and one reply to that list is simply "not sway".
| Style | What the chat says (27 Oct) | Rarity | Published numbers |
|---|---|---|---|
| Pendulum | Named in the "big 4 of ranked" | Legendary | None |
| Ghost Step | Named in the "big 4 of ranked" | Legendary | None |
| Peekaboo | Named in the "big 4 of ranked" | Legendary | Dodges straights 46%, hooks 55% (18 Oct) |
| Slick | Named in the "big 4", and picked over Soviet | Mythic | Catches jabs 40%, crosses 36%, uppercuts 30% (20 Oct) |
| Soviet | "Gets countered by almost everything"; "impossible to win close fights" | Mythic | None; the developer names body work and pressure as its weak spots |
| Sway | "Not sway", when it came up | Epic | None |
That is chat from one channel across two days. Nobody in the thread quoted a win rate, and the styles they name are not the styles with the most published numbers. Treat it as what the pool repeats, not as a measurement.
What the patches changed, style by style
Between 18 and 22 October the developer published more style numbers than in any other window of this game's life. If what you know about a style predates those dates, this is what moved.
| Style | The published change | Direction |
|---|---|---|
| Peekaboo | 18 Oct: the bob stops counting as a dodge while you punch, and it burns more gas. 20 Oct: dodges more instead — straights 46%, hooks 55% — and guard drains less gas. | Nerfed, then partly restored |
| Slick | 18 Oct: catches more hooks, overhands, jabs and crosses. 20 Oct: jabs 40% (was 50%), crosses 36% (was 46%), uppercuts 30%, and the guard fades when you are low on gas. | Raised, then trimmed |
| Philly | 20 Oct: jabs 60%, crosses 55%, hooks 48%, overhands 28%, uppercuts 36%, body shots blocked 20%. 21 Oct: the rear hand catches a little less and picks off fewer jabs down the middle, with no new figures. | Buff, then trimmed |
| Weaver | 18 Oct: hook dodges cut to 78% from 88%, overhands to 75% from 85%. 20 Oct: no longer gets extra glove catches; the dodge is its whole defence. | Down |
| Hitman | 18 Oct: catches jabs 40%, crosses 30%. 20 Oct: jabs 50%, crosses 38%, hooks 22%, rolls 20% of hooks, blocks 20% of body shots, flicker jab at full power. | Up |
| Karate | 18 Oct: catches more straights, the blitz comes faster and hits harder. 20 Oct: jabs 55%, crosses 46%, hooks 26%, body shots blocked 20%, blitz out of a 0.5s pause. | Up |
| Muay Thai | 18 Oct: catches jabs 30% and crosses 28%, blocks uppercuts 20%. No figure since. | Up |
| Long Guard | 18 Oct: catches far more and the guard barely drains gas. 20 Oct: jab and cross about 15% faster, jab hits 12% harder, catches hooks at 18%. | Up |
| Crossarm | 18 Oct: blocks hooks 20%, body shots 15%, and drains less gas. 20 Oct: walling a straight or an uppercut opens a BLOCK COUNTER. | Up |
| Parry | 20 Oct: catches hooks at 18%, up from 7%. | Up |
| The Finisher | 20 Oct: walls power shots — hooks 24%, overhands 18%, body 22% — and answers a walled straight or uppercut with a BLOCK COUNTER. | Up |
| Bulldog | 18 Oct: blocks hooks 48% (was 55%) and body shots 50% (was 65%). | Down |
| Pendulum | 20 Oct: swing hooks cut to +15% from +25%, and only off the fast middle of the swing. 21 Oct: rolls and dodges catch fewer hooks and overhands, every save costs more stamina, and uppercuts land 15% harder on it. | Down |
| Sway | 18 Oct: dodges that fired too late against hooks were fixed. 20 Oct: far less gas drain. 21 Oct: no auto-roll or auto-slip while you throw, and guard drain cut from 1.2 to 1.0. | Reworked |
| Dizzy | 20 Oct: burns less gas and recovers faster. 21 Oct: no auto-slip while you throw. | Reworked |
| Soviet | 20 Oct: released as the second Mythic — the jab runs everything, the rear glove picks off jabs, the high guard walls hooks. 21 Oct: a clean cross or hook grants a burst of faster footwork. 22 Oct: jab while moving forward to step in, jab while backing up to slide out, on a 1.2s cooldown. | New, then trimmed |
Three of the recorded styles took no change in this window. High Guard is the starter style, Rope-A-Dope has a panel description but no published rate, and Ghost Step has no entry in the update channel at all.
How to turn this into a pick
Four steps, in the order that saves you spins.
- Name the punch that keeps landing on you. Not the style you like. If you cannot name it, watch one replay: since 20 October the knockout replay shows the hand that actually landed.
- Take the published answer for that punch. The table below keeps only the strongest published rate per punch. The full rows, including who catches a hook at 18%, are on the styles page.
- Decide whether you are spending on the pool's opinion or on a number. Two of the four names players repeat have nothing published behind them, so following those names is a bet rather than a calculation.
- Price the bet before you roll. The spin panel lists Legendary at 3% with 0.75% per Legendary style, and Mythic at 0.1%; a Lucky spin lists Mythic at 0.5% and offers no Common or Rare. Those are panel readings, not developer statements.
| Punch that keeps landing | Strongest published answer | Next best, same punch |
|---|---|---|
| Jab | Philly catches 60% | Karate 55%, Hitman 50%, Slick 40%, Muay Thai 30% |
| Cross | Philly catches 55% | Karate 46%, Hitman 38%, Slick 36%, Muay Thai 28% |
| Hook | Weaver dodges 78% | Peekaboo dodges 55%, Philly 48%, Bulldog 48%, Karate 26% |
| Overhand | Weaver dodges 75% | Philly 28%, The Finisher 18% — a steep drop from Weaver |
| Uppercut | Philly catches 36% | Slick 30%, Muay Thai blocks 20% |
| Body shot | Bulldog blocks 50% | The Finisher 22%, Philly 20%, Hitman 20%, Karate 20%, Crossarm 15% |
Two cautions sit inside that table. The Weaver and Peekaboo figures are dodge chances, so they only pay off when you are free to move; since 21 October Sway and Dizzy lose their automatic dodge while they punch, and a dodge style has no catch to fall back on. And Philly's row was published on 20 October and trimmed again on 21 October without new figures, so 60% is the strongest number ever published for a jab and not necessarily today's.
Why this page prints no S–D grades
A grade needs one number per style. The developer publishes rates per punch type: Philly catches jabs at 60% and overhands at 28%, which is one style doing two jobs with no published weight to combine them into a score. A single grade would be this site's invention, and the two tier-list pages read while writing this page say the same thing in their own text: both call their order editorial, and neither cites a measured win rate.
What would change the page: a published win rate or pick rate per style, or a developer ranking. On 18 October the developer wrote "We now track how every style performs in real fights, so future balance changes are based on actual results". That data exists on their side. It has not been published, so nothing here pretends to have it.
What this page cannot tell you
- No win rate, pick rate or ranked usage count per style appears in any source read for this page.
- Seven of the 19 recorded styles have no defensive figures at all: Pendulum, Ghost Step, Soviet, Dizzy, Sway, Rope-A-Dope and High Guard.
- Most numbers are from 18–20 October. The 21 and 22 October patches changed Philly, Sway, Dizzy, Pendulum and Soviet again without publishing replacements.
- Whether height or weight class favours a particular style is unpublished; the controls page collects what the panel shows and what players report.
- The 19 styles are the ones recorded on this site, not a confirmed full pool: the Common and Rare lists were never captured.
Each of those is missing rather than left out. Nothing on this page fills a gap with a guess.