L Tokyo Ghoul
Explore this model alongside other AT pachislot machines in our Japanese pachislot machine library. For the basics, read how pachislot works in Japan.
L Tokyo Ghoul (L 東京喰種) is a smart pachislot AT machine from the Tokyo Ghoul pachislot project, introduced in Japan on February 3, 2025. Normal play uses game-count and rare-role routes to reach CZ or direct AT opportunities, while the main AT Tokyo Ghoul BITES (東京喰種咬/バイツ) is a difference-coin-managed AT with an approximate 4.0 coins/G net increase.
New to pachislot? If you found this machine through the Tokyo Ghoul anime, start with our Tokyo Ghoul pachislot guide for anime fans before using the detailed settings, ceiling and AT data below.
Last verified: October 14, 2026.
Key terms: Smart Pachislot · AT · CZ · Setting · Payout rate · Initial hit · Net increase · Small role · Rare role · Replay · Bell · Weak cherry · Strong cherry · Watermelon · Chance pattern · Bonus · Ceiling · Reset · Zone · High-probability state · Quit timing · Mode · Color aiming
Quick specs
Quick specs
| Item | Value |
|---|---|
| Official Japanese name | L 東京喰種 |
| Manufacturer | SPIKY |
| Series | Tokyo Ghoul |
| Machine type | AT |
| Japan rollout date | February 3, 2025 |
| Model | L東京喰種CT / FT |
| Brand / project | SPIKY / Tokyo Ghoul pachislot project |
| Main AT | Tokyo Ghoul BITES (東京喰種咬/バイツ) |
| AT net increase | Approx. 4.0 coins/G |
| Base | Approx. 31G per 50 coins (Administrator research) |
| Normal CZ ceiling | Up to 600G + α |
| AT-interval ceiling | 1200G + α |
| Color aiming for payouts (色目押し) | Not confirmed — Payout color-navigation requirement not established. Color aiming definition. |
Setting probabilities and payout rate
Setting probabilities and payout rate
Administrator research.
| Setting | CZ | AT initial hit | Payout rate |
|---|---|---|---|
| 1 | 1/262.6 | 1/394.4 | 97.5% |
| 2 | 1/255.6 | 1/380.5 | 99.0% |
| 3 | 1/246.5 | 1/357.0 | 101.6% |
| 4 | 1/233.1 | 1/325.9 | 105.6% |
| 5 | 1/216.4 | 1/291.2 | 110.3% |
| 6 | 1/203.7 | 1/261.3 | 114.9% |
CZ routes and success conditions
CZ routes and success conditions
Administrator research. Both normal CZs are eight-game self-driven AT challenges. A freeze is the success notification; the role drawn during the CZ determines the chance of reaching it.
| CZ | Length / expectation | Success condition | Success reward |
|---|---|---|---|
| REMINISCENCE (レミニセンス) | 8G / about 50–52% | Each game draws for success from the role. Push-order bells and rare roles are the main opportunities; strong roles can succeed immediately. Freeze = success. | Tokyo Ghoul BITES (東京喰種咬/バイツ) AT |
| Big Eater Rize (大喰いの利世) | 8G / about 75% | Role-based success draw with replay/push-order bell chances and very strong rare-role treatment. Freeze = success. | Tokyo Ghoul BITES (東京喰種咬/バイツ) AT |
On the final game, some roles receive stronger or guaranteed-success treatment. The CZ occurrence probability and the CZ success probability are different quantities and should not be read as one combined rate.
Official Q&A: CZ and special challenge outcomes
REMINISCENCE (レミニセンス) can signal a successful AT through changes to its music or its final-game display. In a separate AT battle challenge, “All the world’s disadvantages…” (この世のすべての不利益は当人の能力不足) also checks the otherwise empty role: its published success chance on that role is 2.34%. That number applies to that challenge and role, not to every eight-game CZ.
The stronger Big Eater Rize (大喰いの利世) result and EPISODE BONUS (エピソードボーナス) are separate routes. A red result button after an ordinary normal-play consecutive scene indicates Big Eater Rize (大喰いの利世) or better; it does not by itself identify a machine setting.
Recognizable CZ cues: In REMINISCENCE (レミニセンス), a final-game display reading “最後” confirms AT; a red LAST display, a rainbow cabinet LED or music changing to “ON MY OWN” are other special notification patterns. These visual cues do not change the base CZ probabilities. A confirmed CZ premonition can still be upgraded by a rare role before the CZ starts.
Small-role probabilities
Small-role probabilities
Administrator research. Setting-1/common values unless noted.
| Role | Probability |
|---|---|
| Replay | 1/7.3 |
| Diagonal bell | 1/131.1 |
| Weak cherry | 1/70.3 |
| Watermelon | 1/100.5 |
| Chance pattern A | 1/585.1 |
| Chance pattern B | 1/585.1 |
| Strong cherry | 1/356.2 |
| “Ghoul” alignment (喰図柄揃い) | 1/2048 |
| Fixed cherry | 1/16384 |
Lower-row replay (赫眼リプレイ) is setting-dependent and always moves the machine into Kakugan (赫眼状態). Its published appearance rates are:
| Setting | Lower-row replay (赫眼リプレイ) |
|---|---|
| 1 | 1/1260.3 |
| 2 | 1/1213.6 |
| 3 | 1/1170.3 |
| 4 | 1/1129.9 |
| 5 | 1/1092.3 |
| 6 | 1/1024.0 |
Official clarification: The fixed cherry is one flag at 1/16,384 even though its visible stop pattern can vary with the left-reel stop position. Apart from lower-row replay (赫眼リプレイ), the published small-role appearance rates do not change by setting. Chance-pattern A and B share the same draw treatment; they are not stronger and weaker chance roles.
Game flow
Game flow
Normal play uses both game-count progression and rare-role lotteries. CZ is the main route to AT, while some lotteries can produce direct AT entry. A lower-row replay (赫眼リプレイ) moves the machine into the Kakugan (赫眼状態) high-probability state, where rare-role appearance becomes much denser.
- Normal stages shift between daytime, evening and night, with higher stages indicating stronger internal expectations.
- Game-count modes create different CZ target zones and ceilings.
- CZ success or direct-entry routes lead to Tokyo Ghoul BITES (東京喰種咬/バイツ).
- AT battles lead into feature zones such as BITES (バイツ), Centipede Awakening (百足覚醒) and One-Eyed Owl (隻眼の梟).
Where does EPISODE BONUS (エピソードボーナス) fit? It is an alternative first-hit route that already confirms AT on entry. It awards about 100 coins, with rare-role opportunities to add coins to the following AT. Reaching an ordinary CZ only gives an opportunity to win AT; entering EPISODE BONUS (エピソードボーナス) has already cleared that hurdle.
For a first visit, track the current state before interpreting a counter: normal-play CZ progression, the AT interval and the game count inside AT describe different parts of the machine. A battle countdown inside AT is not the normal-play ceiling count.
Direct AT versus EPISODE BONUS (エピソードボーナス): An initial AT reached without EPISODE BONUS (エピソードボーナス), excluding AT comeback and the AT-interval ceiling, counts as a direct AT. AT reached via EPISODE BONUS (エピソードボーナス) is counted separately. A replay-triggered direct AT can look like a hit with no obvious rare-role cue. The in-machine AT total also counts entries following a successful Arima Judgment (有馬貴将ジャッジメント), whereas the published initial-AT probability does not.
Official Q&A on watermelon game-count additions: This draw uses the same distribution across settings, normal modes and elapsed normal-play games.
| Added target games after a watermelon | Selection share |
|---|---|
| +10G | 74.0% |
| +100G | 25.0% |
| MAX | 1.0% |
The visible add-on can still be announced during a confirmed premonition. The larger central-screen watermelon animation carries approximately a 50% chance of showing +100G or more; this is an animation-conditional cue, not a new per-watermelon distribution.
Ceiling, reset and quit timing
Ceiling, reset and quit timing
Administrator research. The normal CZ ceiling is up to 600G + α on the liquid-crystal game counter. Mode-dependent ceilings can be shorter: Normal C (通常C) up to 500G + α, Heaven Preparation (天国準備モード) up to 300G + α, and Heaven (天国モード) up to 100G + α.
The AT-interval ceiling is 1200G + α and awards AT. After a setting change, a special reset mode can shorten the normal ceiling to a maximum of 200G + α.
The manufacturer states that after a failed Arima Judgment (有馬貴将ジャッジメント), the next target has a maximum prescribed game count of 100G. This is a normal-play CZ opportunity, not a guarantee of AT at exactly 100 displayed games; the ensuing premonition can extend the visible stopping point. Check this state before leaving.
Tokyo Ghoul BITES (東京喰種咬/バイツ) — AT
Tokyo Ghoul BITES (東京喰種咬/バイツ) — AT
| Type | Difference-coin-managed AT |
|---|---|
| Initial amount | Approx. 150 coins |
| Net increase | Approx. 4.0 coins/G |
| Main loop | AT → battle → feature zone / coin addition |
AT battles occur frequently, with an overall battle-win expectation around 50% depending on opponent and state. Winning can lead to BITES (バイツ). “Ghoul” alignment (喰図柄揃い) can route into premium feature zones such as Centipede Awakening (百足覚醒) or One-Eyed Owl (隻眼の梟).
The main AT is Tokyo Ghoul BITES (東京喰種咬/バイツ). BITES (バイツ) is a separate add-on zone within that AT.
What happens while the AT is running?
- Main AT: follow the navigation while the remaining coin allowance is consumed. The initial 150 coins describe this allowance; they are not a fixed number of games or your total credit balance.
- Battle opportunity: rare roles and game-count progression can start a Ghoul Battle (喰種対決). The opponent affects the chance of winning. The lower-right AT counter helps track the game-count route.
- After a win: BITES (バイツ) upgrades its reward through small roles and adds coins to extend the main AT. It is a feature within the AT loop, not a second name for every AT screen.
- At the end: when the allowance is exhausted without further continuation, the AT ends. Check the next screen before treating the machine as back in ordinary play; Arima Judgment (有馬貴将ジャッジメント) is a separate challenge with its own aftermath.
The approximate 4.0 coins/G figure describes net increase during AT. It does not mean every individual game pays four coins, and it cannot be multiplied by all the games shown on the hall counter to calculate your session profit.
Official Q&A: AT and feature-zone figures
These are conditional figures. The denominator for each row is the named state or trigger, not all normal-play games.
| Condition | Published result |
|---|---|
| AT game-count battle draw | A Ghoul Battle (喰種対決) is selected within 30G in 46.88% of the published AT game-count draws. This is a battle occurrence figure, not its win rate. |
| Guaranteed battle / AT episode | An AT episode plays in approximately 23.91% of guaranteed-battle entries. |
| Special “All the world’s disadvantages…” (この世のすべての不利益は当人の能力不足) challenge | An otherwise empty role succeeds with 2.34% probability on that challenge game. |
| BITES (バイツ) retry after an empty result | A fresh 20% retry draw is made on each applicable empty result; retries are not held as advance stocks. |
| Centipede Awakening (百足覚醒) add-on loop | Approx. 67.5% overall. The published per-game add-on chances are 60.5% for other roles and 100% for diagonal bell, replay or rare roles; the 67.5% figure is the combined average across roles. |
BITES (バイツ) entry game: With no stop-order prompt, the official Q&A reports an apparent empty result in only 0.05% of entry games; otherwise a listed bell, replay, rare role or the “Ghoul” alignment (喰図柄揃い) flag is involved. “Ghoul” alignment (喰図柄揃い) is a 1/2,048 flag, but it can look like a miss when it is not announced in that entry game. These figures do not imply that every ordinary AT game has a 0.05% miss rate.
| Rare role on BITES (バイツ) entry | Reward-table upgrade chance |
|---|---|
| Watermelon | 10.0% |
| Weak cherry | 25.0% |
| Strong cherry or chance pattern | 46.8% |
A rare role on the first game can also upgrade Centipede Awakening (百足覚醒) into One-Eyed Owl (隻眼の梟); if One-Eyed Owl (隻眼の梟) was already selected when the “Ghoul” symbols (喰図柄) aligned, that first game instead draws for an add-on. Ultra BITES (超BITES) can upgrade further after a confirmed higher reward through later rare-role draws, including its entry game. During an ending, a confirmed role on the first non-advantageous-section game after the ending is treated as an Arima Judgment (有馬貴将ジャッジメント) preparation draw and strongly indicates a long freeze. A lower-row replay (赫眼リプレイ) in the non-advantageous section strongly indicates the CCG Death God (CCGの死神) reward and the enhanced AT.
Official Q&A: CCG Death God (CCGの死神) and enhanced AT
After a successful Arima Judgment (有馬貴将ジャッジメント), the CCG Death God (CCGの死神) add-on amount depends on the role drawn. These are conditional reward distributions at this feature, not ordinary AT coin payouts.
| Role in CCG Death God (CCGの死神) | Coin-addition distribution |
|---|---|
| Other | 300 coins: 50%; 500 coins: 50% |
| Replay or diagonal bell | 500 coins: 50%; 1,000 coins: 50% |
| Watermelon or weak cherry | 1,000 coins: 50%; 2,000 coins: 50% |
| Strong cherry, chance pattern or fixed role | 2,000 coins: 100% |
In an enhanced AT, Arima (有馬貴将) appears by the fifth battle; his published selection shares are 18.75% at battle 1, 18.75% at battle 3 and 62.50% at battle 5. The Arima Judgment (有馬貴将ジャッジメント) itself checks every game’s role for success, rather than accumulating points. Its preparation phase applies different confirmed-role and rare-role treatment from an ordinary AT game. Ordinary replay (other than lower-row replay (赫眼リプレイ)) during a non-advantageous section does not trigger the transition to an advantageous section.
Battle and premium-feature clarifications: The opponent, including a possible Arima (有馬貴将) encounter, is selected at an AT-entry lever pull; a battle level is then redrawn when a character is selected, so the displayed opponent changes the role-by-role win draws. A rare role can upgrade a confirmed battle through the “All the world’s disadvantages…” (この世のすべての不利益は当人の能力不足) challenge to Ultra BITES (超BITES) and then Extreme BITES (極BITES). Entering One-Eyed Owl (隻眼の梟) on the first game of Centipede Awakening (百足覚醒) can reflect a first-game upgrade rather than an original One-Eyed Owl (隻眼の梟) selection. A lower-row replay (赫眼リプレイ) during an AT “Ghoul symbol” (喰図柄) cut-in can give a distinctive stop pattern without the corresponding one-confirmation voice.
Setting differences and identification
Setting differences and identification
CZ and AT initial-hit rates improve steadily with setting and are the primary published numerical indicators. Lower-row replay (赫眼リプレイ) is also setting-dependent.
Administrator research: direct AT from an ordinary replay rises substantially with setting, from roughly 1/28,460.6 at Setting 1 to 1/7,036.8 at Setting 6. AT return probability also rises from approximately 7.81% at Setting 1 to 15.23% at Setting 6. These are useful supporting indicators only after a sufficient sample.
The statistical ranges below include lower-row replay (赫眼リプレイ) appearance only. The published replay-triggered direct-AT figures are too sparse for the current 25/50/100/200 trigger sample display to be practically informative, and AT-return is a separate post-AT process rather than a small-role/direct-hit trial in this feature.
Setting-specific official Q&A figures
The direct-hit table concerns an ordinary replay triggering a direct AT; it is not the probability of every direct AT. The enhanced-AT table concerns an AT entry without Arima Judgment (有馬貴将ジャッジメント). Other entry paths have different conditions.
| Setting | Replay-triggered direct AT | Enhanced AT on entry without Arima Judgment (有馬貴将ジャッジメント) |
|---|---|---|
| 1 | 1/28,460.6 | 1.10% |
| 2 | 1/24,453.5 | 1.32% |
| 3 | 1/18,093.0 | 1.63% |
| 4 | 1/12,019.5 | 2.19% |
| 5 | 1/8,615.4 | 2.85% |
| 6 | 1/7,036.8 | 3.32% |
The developer also published higher enhanced-AT selection when AT follows an early prescribed game-count event: if the premonition began by 100G, 3.18% at Setting 1 to 5.14% at Setting 6; if it began by 200G, 2.99% to 4.77%. These are conditional AT-entry rates, not the chance of hitting AT by those game counts. AT comeback probability is a separate post-AT indicator.
Statistical setting ranges
Japanese Slot Guide calculation. The expected counts and central 95% count ranges below are calculated from the listed setting probabilities with a binomial model. They describe ordinary statistical variation under that probability; they are not confidence percentages for the actual setting.
Lower-row replay
Administrator-research per-game appearance rate for the lower-row replay that enters Kakugan. Count the completed lower-row replay pattern, not ordinary replay or a Kakugan state that began for another reason.
1,000 games
| Setting | Listed rate | Expected count | Central 95% range | Δ vs S1 | Relative difference |
|---|---|---|---|---|---|
| 1 | 1/1260.3 | 0.8 | 0–3 | — | — |
| 2 | 1/1213.6 | 0.8 | 0–3 | +0.0 | +3.8% |
| 3 | 1/1170.3 | 0.9 | 0–3 | +0.1 | +7.7% |
| 4 | 1/1129.9 | 0.9 | 0–3 | +0.1 | +11.5% |
| 5 | 1/1092.3 | 0.9 | 0–3 | +0.1 | +15.4% |
| 6 | 1/1024.0 | 1.0 | 0–3 | +0.2 | +23.1% |
3,000 games
| Setting | Listed rate | Expected count | Central 95% range | Δ vs S1 | Relative difference |
|---|---|---|---|---|---|
| 1 | 1/1260.3 | 2.4 | 0–6 | — | — |
| 2 | 1/1213.6 | 2.5 | 0–6 | +0.1 | +3.8% |
| 3 | 1/1170.3 | 2.6 | 0–6 | +0.2 | +7.7% |
| 4 | 1/1129.9 | 2.7 | 0–6 | +0.3 | +11.5% |
| 5 | 1/1092.3 | 2.7 | 0–6 | +0.4 | +15.4% |
| 6 | 1/1024.0 | 2.9 | 0–7 | +0.5 | +23.1% |
5,000 games
| Setting | Listed rate | Expected count | Central 95% range | Δ vs S1 | Relative difference |
|---|---|---|---|---|---|
| 1 | 1/1260.3 | 4.0 | 1–8 | — | — |
| 2 | 1/1213.6 | 4.1 | 1–9 | +0.2 | +3.8% |
| 3 | 1/1170.3 | 4.3 | 1–9 | +0.3 | +7.7% |
| 4 | 1/1129.9 | 4.4 | 1–9 | +0.5 | +11.5% |
| 5 | 1/1092.3 | 4.6 | 1–9 | +0.6 | +15.4% |
| 6 | 1/1024.0 | 4.9 | 1–10 | +0.9 | +23.1% |
8,000 games
| Setting | Listed rate | Expected count | Central 95% range | Δ vs S1 | Relative difference |
|---|---|---|---|---|---|
| 1 | 1/1260.3 | 6.3 | 2–12 | — | — |
| 2 | 1/1213.6 | 6.6 | 2–12 | +0.2 | +3.8% |
| 3 | 1/1170.3 | 6.8 | 2–12 | +0.5 | +7.7% |
| 4 | 1/1129.9 | 7.1 | 2–13 | +0.7 | +11.5% |
| 5 | 1/1092.3 | 7.3 | 3–13 | +1.0 | +15.4% |
| 6 | 1/1024.0 | 7.8 | 3–14 | +1.5 | +23.1% |
How to read this: overlapping ranges are expected. A result inside one setting’s range can also be plausible under another setting. Do not use these ranges as a setting confirmation or as a promise of future results.
Normal-play internal modes
Normal-play internal modes
Administrator research. Six normal-play modes change the CZ target zones and maximum displayed game count. A separate Special mode is used after a setting change.
| Mode | Maximum / characteristic |
|---|---|
| Normal A (通常A) | 600G + α |
| Normal B (通常B) | 600G + α; different 100G-zone emphasis |
| Normal C (通常C) | 500G + α |
| Chance (チャンスモード) | 600G + α; stronger early zones |
| Heaven Preparation (天国準備モード) | 300G + α |
| Heaven (天国モード) | 100G + α |
| Special (reset only) | 200G + α after a setting change |
Higher modes generally move the next CZ opportunity earlier. These modes are separate from the Normal / High / Super-high internal state that changes rare-role-based CZ lotteries, so both layers can matter at the same time.
After a returned AT ends in its first battle
The manufacturer clarified on September 21, 2026 that if an AT obtained through a comeback also ends in its first battle, the following normal-play mode is strongly indicated to be Mode B or higher. This condition refers specifically to a first-battle exit, not every AT ending. It is a mode indication, not confirmation of a high machine setting.
Official mode clarification: A “Normal C (通常C) or higher” end screen can mean Normal C (通常C), Chance, Heaven Preparation (天国準備モード) or Heaven (天国モード). Their published maximum prescribed game counts are 500G, 600G, 300G and 100G, respectively. If that cue is reliable and the prescribed target reaches 600G, it points to Chance mode (チャンスモード); Normal C (通常C) does not have a 600G maximum. On CZ failure the next mode is selected with reference to the current mode, with no downgrade before Heaven (天国モード) is reached. After AT, the next normal mode is drawn anew regardless of the previous mode.
Tsukiyama’s invitation (月山の招待状) gives remaining games until the next prescribed target: for example, a “within 200G” indication seen at 50G points to a target by approximately 250G, not at 200G on the displayed counter. Invitation variants can also hint at settings, so the visual cue should be identified before interpreting it.
State versus stage: The machine may internally be at super-high probability outside the visible Mental World (精神世界) stage. Even a failed premonition can continue the high-state transition draw; Mental World (精神世界) need not appear each time. During normal play, a rare-role CZ hit in super-high state can be announced immediately, while ordinary-state hits proceed through premonition. Enemy dialogue color and its stop-button timing hint at AT battle high state; Kaneki (金木研) / Owl (梟) dialogue instead hints at the next normal-play prescribed-game target.
Setting hints confirmed in official Q&A
Setting hints confirmed in official Q&A
After a failed CZ, the red Juuzou Suzuya (鈴屋什造) result card indicates an even-numbered setting (2, 4 or 6). It is a setting hint; the normal-mode indication on an end card is a different signal. Do not interpret a Normal C-or-higher mode cue as confirmation of Setting 3 or above.
How to play
How to play
No navigation: stop the left reel first. If numbered stop instructions appear, press the stop buttons in the displayed order (1, then 2, then 3). A symbol-target instruction tells you which symbol to aim for; follow it together with any displayed stop order.
When no navigation is displayed, stop the left reel first. Use this routine during normal play and unprompted AT games:
- Aim at the BAR area on the left reel to cover the cherry.
- If watermelon stops on the left reel, aim for watermelon on the middle reel, using its BAR as a visual reference. The right reel can be stopped freely.
- From the usual cherry or lower-BAR stop, the middle and right reels can be stopped freely.
- Whenever numbered navigation appears, follow that order. On a symbol-aiming cut-in, aim for the symbol shown on the screen.
Record the completed role after stopping all reels. In particular, lower-row replay (赫眼リプレイ) signals Kakugan (赫眼状態); an ordinary replay is not the same observation. These patterns help explain the next CZ or AT opportunity, but observing a rare role does not by itself confirm a hit.
Reel-pattern clarification: Cherry stop patterns can differ according to the exact left-reel stop position. A cherry/Ghoul-symbol alignment across the reels is a strong cherry; the fixed cherry at 1/16,384 can show more than one stopping pattern. A lower-row replay (赫眼リプレイ) during Kakugan (赫眼状態) draws again for Kakugan (赫眼状態) duration and adds the resulting games to the remaining duration; it is not a fixed +10G extension.
Color aiming for payouts
Administrator research. The checked material does not establish a reliable Yes or No for color-cued payout aiming across this model’s payout phases, including any upper or ending phase. Follow every displayed symbol and stop-order instruction, and use the existing small-role and bonus procedures below. Free stopping during normal play alone does not establish the payout-navigation requirement.
This field covers color-cued payout navigation. Ordinary bonus-entry alignment, presentation-only aim cut-ins, normal-play small-role targeting and fixed skill-intervention adjustments are separate. Read the Color aiming definition.
Data policy and official references
Data policy and official references
First-party manufacturer information is labelled official. Administrator research identifies the administrator’s review of published information; it does not by itself mean original measurement or manufacturer confirmation. Calculated examples are labelled separately and explain their inputs. A page verification date records an editorial check, not a new experiment or a remeasurement of every value. See our editorial policy for the distinction.
- Official product / project page
- Official developer explanation of the main AT name and BITES
- Official developer explanation of AT battle opportunities
- Official developer explanation of EPISODE BONUS
- Official developer explanation of BITES rewards
- Spiky / Cross Alpha — comeback AT and next-mode clarification
