Pokajan is a balancing act between "aggressively winning to rack up hands" and "playing defense to avoid dealing in." Here we break down how to play each phase of the game and how to identify safe tiles.
Passing up a win costs you: the basic approach is to declare as soon as any yaku is complete, even the cheapest 120-point hand. Reports suggest that holding out for bigger yaku tends to actually lower your average score over time. Since your hand refills back up to 7 cards after every Pokajan, "winning small, often" adds up into a big overall difference.
Over-defending backfires too: trying to avoid every dangerous tile leaves you stuck with a dead hand, which ends up costing you more opportunities than it saves. The optimal approach is a "moderate defense" — avoiding just one or two of the riskiest-looking tiles rather than everything.
Narrowing your "target groups" down to about 2–3 early in the round keeps your card-keeping decisions consistent for the rest of the match.
Smaller groups, like 3-member units, tend to be easier to complete and more practical in real play. Groups that include a bonus Holomen also move up in priority since you can stack extra points on top.
Large groups, like 5-member units, pay out a lot but come with a much higher bar to clear. Spreading yourself thin by chasing several of these early on tends to result in completing none of them. Learning to cut your losses early and say "not this round" on a group is important too.
Don't treat your narrowed-down groups as fixed — stay flexible and switch your target based on what's happening at the table (who's discarding what). Cards that work across multiple yaku (like the same Holomen who's also part of your target group) are worth prioritizing to keep.
4-member group units tend to feel the most balanced: weighing how easy a yaku is to complete against its payout, many players find that targeting 4-member group units as their main line is the most efficient approach. Also, building your whole hand around waiting on a single Bonus Holomen card can lock up nearly half your hand and cost you opportunities elsewhere. Hold onto Bonus Holomen cards when you draw them, but avoid over-committing your entire hand to a wait built solely around one.
Narrow your target yaku down to 1–2, and rack up wins from cheap, easy-to-complete hands to keep your hand cycling. Favor offense over defense at this stage.
If you're near the top, don't push your luck — shift toward prioritizing safe tiles. If you're behind, consider switching toward higher-value yaku. Read the discards on the table to guess which groups other players are chasing, and treat those as dangerous.
Once the deck drops under 20 cards, other players' hands become easier to read. If you've secured 1st or 2nd place, make avoiding a big deal-in your top priority. Conversely, if you're way behind, you may need to switch gears and go for a high-value comeback hand.
Especially when you're winning, it's often better to discard cards tied to lower-value yaku (like 3-member group units) first, accepting a small deal-in as the worst-case scenario, rather than risk a much bigger one. If you can no longer tell what's safe, fall back on a simple rule: discard cards tied to the lowest-value yaku (3-member group units) first. That alone can help you dodge the worst deal-ins.
・The same card another player just discarded (matching a recent discard)
・Cards that have already appeared multiple times in discard piles
・A card discarded by the player right before your turn
・Cards where most other members of that group have already hit the field (making the group yaku unlikely)
・Holomen whose matching-color cards are largely exhausted for the round (making a same-color bonus unlikely)
・Cards that haven't appeared on the field at all yet (another player may be holding them)
・Cards from a Holomen/group a specific player keeps discarding — they may actually be hoarding related cards behind the scenes
・Cards from the designated bonus Holomen (highly sought-after, and costly if you deal in)
Matching a recent discard is your basic defense: discarding the same card that was just discarded — sometimes called a "safety match" — is the most fundamental way to avoid dealing in. The logic: if that card could complete a hand, the player right before you would have already won off it instead of passing.
The number of Holomen implemented in the game changes how many copies of any one character or color you can expect to see in play. Here's how to think about safe tiles based on roster size.
Assumptions for this estimate: we assume 9 cards exist per Holomen (3 colors × 3 copies each), and that 100 cards are actually in play out of the total pool during a given round. Total card count is estimated as "roster size × 9."
| Roster Size | Total Cards | Expected copies per character |
Expected copies per character & color |
|---|---|---|---|
| 14 *1 | 126 | 7.14 | 2.38 |
| 15 | 135 | 6.67 | 2.22 |
| 16 | 144 | 6.25 | 2.08 |
| 17 | 153 | 5.88 | 1.96 |
| 18 *1 | 162 | 5.56 | 1.85 |
*"Expected copies" is the theoretical average number of that character's (or color's) cards you'd expect within 100 cards in play (a hypergeometric expectation: 100 × matching cards ÷ total cards). Actual results will vary round to round.
*1 The 14- and 18-character rows are unverified reference values — the author hasn't confirmed the roster was ever actually at exactly that size.
The bigger the roster, the harder "same-color" gets: as the table shows, once the roster hits 18 characters, the expected copies per character-and-color drops to just 1.85. On average, that's fewer than 2 copies in play at all — meaning both you and your opponents will have a much harder time collecting the 3 needed for a same-color yaku.
Applying this in practice: bigger rosters make "safe discards" easier to find: in a smaller roster environment (around 14 characters), each color is relatively easy to collect at 2.38 copies, so watching out for same-color yaku is worthwhile. In a larger roster environment (17–18 characters), the pool of cards needed for a same-color hand is simply too thin, so confidently discarding "same character, same color" cards as safe tiles becomes a viable play. As the roster keeps growing with future updates, keeping this "harder to complete" factor in the back of your mind should sharpen your tile-reading.
Some general trends we've observed through actual play. Use these as reference for your own strategy.
The link between a good starting hand and your final score is weaker than expected — your decisions after the deal matter more than the luck of your initial draw.
There's little to no noticeable advantage or disadvantage from where you're seated, so it doesn't seem worth worrying about.
Even against beginners, you can lose more often than not over a short stretch of games — real skill differences only become clear after playing a fair number of matches. Don't let any single result get you too high or too low.
Answers to the questions we hear most often — "is this just luck?" and "why can't I win?"
There's no denying your starting hand and draws involve luck. But as the "Data Worth Knowing" section above notes, the link between a good starting hand and your final score is weaker than you'd expect — what really decides the outcome is how you play from there. Luck is part of it, but there's plenty of room for skill to make a real difference.
See if any of these sound familiar:
・Are you passing up cheap yaku too often? (See Core Approach — winning small, often, is the basic plan.)
・Are you playing too defensively and getting stuck with a dead hand? (Over-defending backfires too.)
・Are you spreading yourself across too many groups instead of narrowing down? (See Narrowing Down Groups.)
A useful benchmark is "average rank," a concept mahjong players rely on often. Since Pokajan is also a 4-player game, if placements were assigned completely at random, the average would naturally work out to 2.5 ((1+2+3+4)÷4 = 2.5). That's why mahjong players commonly treat "beating 2.5" as one gauge of skill, with under 2.4 often considered an advanced-level benchmark.
Rather than fixating on any single match's result, it's more useful to average your placement over dozens of games. Landing in 4th even once drags your average down a lot, so prioritizing "avoid 4th place" tends to stabilize your average rank more than aggressively chasing 1st. You can actually see your own "Average Rank" in a popup by tapping a character icon during a match (explained in the Reading Player Info section of the Rules & How to Play page).
Forget the finer details for now — just remember "win small, often, even on cheap hands." See Core Approach for more.
Want to farm coins more efficiently? Check out the Coin Farming page as well.
💬 Comments
Got a strategy or tip that's worked for you? Feel free to share it.