Is there actually any strategy to this?

[Karuchi-GM]

Or is it all just down to dumb luck-of-the-draw?

[dpuza]

There is strategy. The game stops once all the cards in the hand have been played (8 turns). There are a limited number of "yaku" cards in the deck. (5 Light cards total in the deck, 3 Blue Ribbons total in the deck, 3 Red Poetry Ribbons in the deck.) There is risk/reward calling Koi Koi or not. If you call it right you could maximise your score. Call it wrong you lose it all and owe them instead. If you take the right card at the right time you could prevent your opponent from being able to get certain Yaku.

[Karuchi-GM]

Doesn't seem to matter all that much when your opponent flawlessly scores on the first turn just about every round.

[dpuza]

How are they scoring on the first turn every round? 2-5 matched cards are needed to make a Yaku. Do you mean they are matching a card on the first turn? That's no big deal. The combinations (Yaku) are what matters.

[Karuchi-GM]

I'm talking about Hanamikaze and Tsukumikaze often within the first draw. Doesn't really feel fair.

[Jxt09]

The strategy in this game reminds me a little of poker or rummy or yatzee. Where you have a choice between going for so many of a kind, or a sequence (1-2-3), or all of same suit. And it's also about preventing the opponent from getting cards they need. The moon card is the one to get at the first opportunity. You can make a yaku with just the moon and another card (forget off the top of my head). Or if you see your opponent has 4 ribbon cards, prevent him from getting a 5th ribbon card.

The more you play the more you will develop your strategy and understand the depth of this game. It's a little different than card games we are used to playing because intead of having 4 suits with lots of cards, this game has 12 suits (months) with only 4 cards per suit.

[dpuza]

Originally posted by Karuchi-GM:

I'm talking about Hanamikaze and Tsukumikaze often within the first draw. Doesn't really feel fair.

So Cup+Moon / Cup+Curtain... Hmm, if really as often as you imply, sounds glitchy. Obviously grab those cards before they can if possible.

[Jxt09]

Also, remember that both players start with 30 points and when someone wins a hand they steal points from the other person. It's different than games where everyone starts at zero and whoever has the most after so many hands wins the game. It can seem like you are way behind if your opponent has 45 points and you only have 15, but all you need is 16 points to win. I've played many a game where I was behind by the final hand but ended up winning the last hand and getting enough points to carry the game.

[Karuchi-GM]

I do brab those cards whenever possible, since they give mad points if you call Koikoi and get another match.
Seems it's pretty much down to luck for me, and it goes either really good or really bad.
It's the millenia-old problem with the handicap of initiative, I suppose.

[dpuza]

Also remember the winner of the round goes first in the next round, and so has the advantage. Something to keep in mind when deciding koi koi or stop.

[shoopy]

It's both luck and strategy. Remember that the opponent (or you) has to match a card on the table in order to score. This means you can prioritize some cards for defense by not playing them right away, so that if a dangerous card comes out you can instantly steal it.

However some times it's just plain dumb luck. But not always.

[shoopy]

Also remember that all cards of a suit are valuable because you need to match pairs to score. This means for example you need to watch out for ALL chrysanthemum cards and not just the sake cup. You can use this to your advantage with practice.

Or for example if your opponent just grabbed a deer last turn and you see a plain clover card on the table, and you have another one, it might be a good idea to snatch that up to lessen the chance of the opponent collecting a boar off it. Or of you have clovers in your hand but not the boar, delay putting them out in case the opponent has the boar, depriving them from possibly making a yaku.

[Karuchi-GM]

Sounds more or less like what I've been doing now that I've gotten some practice in.
Clearly underestimated what I could do with the right strategy. You have to prioritize stealing important matches for your opponent over your own score, in most cases.

[shoopy]

Originally posted by Karuchi-GM:

Sounds more or less like what I've been doing now that I've gotten some practice in.
Clearly underestimated what I could do with the right strategy. You have to prioritize stealing important matches for your opponent over your own score, in most cases.:trophy:

Yeah, just keep in mind that the cards you get from the start are all the ones you have in your hand for the entire round. You kind of have to get a feel of what to go for and when based on this. Between the cards you get dealt, the ones on the table, and the ones your opponent picks up you have a lot of info to plan a strategy. Some times you go for a big score and other times you just try to survive the round and prevent the opponent from winning.

[sfnhltb]

There are a number of key ways you can increase your odds of winning, although the details can be somewhat situational (unlike the AI who plays the same way regardless of the match situation). One part is to use try to maximize the number of cards you end up with, and thus reduce the number of cards the AI gets - this is particularly important in this version with the defensive way the AI plays (occasionally stopping the first score after 4 cards, often by 5 and basically always by 6th or later, and without any seeming consideration of either your threat of scoring or how many points the AI could gain by playing on). In my experience pretty much any round the AI gets to 10 kasu/5 tan/5 tane it will end (because it takes a few rounds to get this many usually), so a bias to quantity of cards compared to what would be a normal strategy for this game makes sense, although it clearly has to be tempered by going for scoring sets/blocking scoring sets of the AI.

Obviously each turn you should mainly be first looking for all your cards which match the board, as it is almost always a bad idea to play cards that don't match. But also you should be looking for which of those matches are positive and which are neutral - positive comes in two strains, one being cards which you match on the board which removes the only copy of a month currently on the board, reducing her chances of having a match to play, which will tend to increase the amount of options you have next turn.

Even better is when you have two of that month in your hand, as by matching the one on the board as long as you don't run out of neutral cards to play you can always be sure of winning all four for that month (if the last one does appear anyway), and avoid having to use up two turns just to get two cards that started in your hand if the AI matches the face up one before you get around to playing one of yours (this is less critical if you have both non-blank cards in your hand although it is risking the AI being 2 cards closers to scoring Kasu if you leave it too long)

The other, usually less urgent, but positive move is where there are two cards for a month you have a match of, at least one of the face up cards not being blank, and you choose which you want before the AI might take that choice away.

Neutral/holding moves are where you can see all 4 cards of a month face up, collected by either player, or in your hand. These are in most cases best left until after all positive moves have been made, unless you need to score rapidly as the AI is in danger of winning the round and you are just rushing a scoring set (or just the possibility of one if you get lucky with your turn cards).

Keeping one neutral card in hand that is a scoring play can be a strong strategy if the AI is in a position of no (or very low) risk of scoring, even if means you are playing unmatched cards - essentially hoping to get lucky with your turn card matching something, or the AI putting down something you can match next turn. If you are well behind in the match and still not having many points in this round I generally would just take the risk and play the neutral scoring card first rather than hold it however, although you also have to factor in the increased risk of losing Oya if this isn't the last round of a match.

The only other time I would play a non-matched card when I still had matching cards in hand is if the AI is pretty much guaranteed to score next round (9 Kasu or more, or something like that) and stop, then it can be worth running a Hail Mary and playing an unmatched card that could score for you if your turn card matchs it - you have to consider the odds are always against you getting it, something around 5% if you need the exact card from the pack to appear, but it can be about 1 in 6 or slightly higher if the card in hand is the one you need to score and all 3 others of that month are not face up and it is late in the round.

Even once you have no matches what you play can still be critical, as she likely has some matches for cards in your hand unless it is very late in a round, so you have to try to work out what is the lowest risk of helping your opponent as well as trying to retain the cards that may help you score, which to do without playing slowly and checking help all the time requires you to know all the months and which scoring sets they help towards.

Beyond that is the strategy of which scoring sets to go for, and thus which cards are priority to collect - this factors in the match situation (if you are behind a lot probably you need to take risks to get the big sets and not worry about the AI scoring chances or getting Kasu/Tan/Tane yourself, whereas if you aren't Oya and it is early enough you should be more defensive as playing first increases your odds of scoring well significantly), what is in your hand - both the actual cards that are part of a scoring set, and the blank cards you have which are the same month as the same scoring set, plus what is on the board, etc.

It can also be useful to factor in the "newness" of a card when deciding what to play - lets say you get a Red ribbon on your first move and a second one is also on the board. If the AI doesn't collect it that turn, and happens to turn the third Red ribbon at the end of her turn then it is better odds to collect the new third ribbon (if you have cards to match both) on your second turn rather than the one the AI has already had a chance to collect. It isn't a guarantee, as the AI may well have had something else it considers more important than blocking your scoring set, but the AI does like to block your sets so when it doesn't that is information you can use to increase your odds of scoring.

I guess you can also look at the stats - I have played over 100 matches and won around 80% of them (mostly 6 round games), that is incredibly unlikely if it was just all luck of the draw (and I have been playing fairly poorly trying to get the 10 wins in a row achievement and especially the 10 times scoring Gokou achievement), and I haven't played in years and hardly an expert in the first place. When I first started playing 12 round matches and not paying any attention to achievements it was about 90%. I would estimate if you play the long matches it is something like 80% skill and 20% luck (but about half the time the luck is in your favour), in the shorter format it is probably more like 70%/30%.

[TairikuOkami[SK]]

What strategy, you have 1-2 choices, sometimes not even that, it is pure luck, win or loose.