ONITAMA: Gotcha! An elegant and simple game for 2 players.

[FHenker]

'No luck involved' is almost correct. However the cards dealt at the start have a huge impact, depending on what cards are in the game. Certain constellations lead to one player being able to withhold a dangerous card from the opponent. Since this varies with the overall available cards in a game, it is not a particular card that is the linch pin. However, in many games I have played, the elephant, tiger and dragon cards were excessively strong, and one player usually sat on them to hold the threat over the opponent.

I'd say, there is luck involved. I agree with the overall feedback though: Great addition!

I disagree with you: with very few exceptions (the most obvious one being a game with 4 forward cards and only 1 diagonal), the initial hand or the starting player matter very little to the outcome of the game, because there are plenty of opportunities to outplay the other player.

You mentioned the Dragon as being one of the strongest cards of the game, but it is the weakest card in the game. Tiger indeed warps games, but not because it is too powerful, but rather because it adds a new element to the game; however, strong players will know how to defend against it. The Elephant is indeed one of the strongest cards in the game, similar in strength to the Rooster/Goose, but again only the better players know how to take full advantage of it. None of these cards break the game and create an unbeatable advantage.

Holding on to a specific card is one of the mistakes that many beginner/intermediate players make, and which better players can easily exploit. With very few exceptions (like the example I mentioned in the beginning, or very specific endgames), holding on to a card leads to weaker game structures and to a loss against good players.

I must agreed with Bobba. While it is true, that a stronger player will usually beat a (relatively) lesser player, two relatively equal players do depend on the initially dealt cards. It is not a perfectly symmetrical game (but it is really close). As Bobba mentioned, it is not particular cards, but particular constellations. Sometimes the cards monopolize a certain move, usually a diagonal, Sometimes 1 forward, or a sidestep. Of course, not ever playing such a card is a mistake. If you just keep them, you play with one card less, which is obviously a hinderance. But the fact that in certain situation you *could* play it forces your opponent to play around that. Which card/s make the difference strongly depends on the available cards though, I think.

Every higher level game of Onitama is essentially defined by the midgame play of both players and, to a lesser extent, by how precise they are in the end game. Two relatively equal (strong) players will split their games based on the mistakes or inaccuracies they make in the midgame, not based on the cards that are dealt.

It is true that certain particular constellations, to use your naming, are extremely biased: for example, a game with 4 forward cards and 1 single diagonal can be won mechanically by whoever has access to the diagonal. If the diagonal is 2-sided it is even easier to win. Another constellation that results in very warped games is Tiger + Dragon, and in this one there is no clear winner or mechanical way to force a win but it is very easy to make mistakes, so it will simply tend to fall towards the player who knows how to play this combination better (and most players are bad at it, myself included despite being a good player overall).

Apart from these two examples (and maybe 3 forwards + 2 diagonals, but to a much lesser extent), no constellation has a serious impact on the outcome of the game. What has the main impact in the outcome is imprecisions/errors in midgame play, followed by imprecisions in endgame play.

It’s true that between weaker players who both want to hold on to the “best” card, the cards might have some impact in the outcome of the game. But between stronger players, who are very proficient in the openings and who don’t make obvious mistakes in the midgame or endgame, the cards have a negligible effect in the outcome of the game, apart from the constellations I mentioned above. At that level, the game is all about position and tempo, and not about individual cards or combinations of cards.

Note however that there is a very special type of card that is very strong in the endgame, and some of the best players plan their midgame so that they reach the endgame with those cards in their possession: those are the cards that look to both sides horizontally (like the Elephant, the Boar and the Goose/Rooster), because they are excellent at walling the opponent’s Master.

But truly, don’t think a higher level game of Onitama is defined by the cards that are dealt, it is all about gaining small tempo and space advantages in the midgame that can snowball into a winning endgame position, the individual cards or their combination is not the winning factor.

I recommend that you check out the YouTube video series Onitama Beginner to Master, made by Onitama_Oni: he was by far the best player in the now dead Onitama app and is the best player I know, and he explains very clearly the main strategic concepts of Onitama. He also explains in one of these videos why the cards/constellations are very overrated, especially by weaker players. He coached me and some of the strongest players in the app, and I bet he will be joining BGA at some point too

[TreffnonX]

I think this boils down to this:
Except for certain constellations humans are not able to calculate deep enough down the decision tree for the Initial cards to matter. If perfect qantum computers were to play, they could (possibly) calculate the winner without playing. However since we are talking about a lot of move options just by two available cards, the decision tree is way too big to actually calculate (with regular computers). Therefore many constellations won't have a noticable bias. I can live with this assessment.

That the game is more 'lost by mistakes' than 'won by ingenuity' is a common phenomenon among all 2 player strategy games with perfect information and no luck involved. For kost of those games, the initial situation is usually the same in every game. The variance onitama comes with is very unusual. It gives the game a lot of replaying value, I think. Looking forward to learning from you about the mistakes in midgame.

[FHenker]

I think this boils down to this:
Except for certain constellations humans are not able to calculate deep enough down the decision tree for the Initial cards to matter. If perfect qantum computers were to play, they could (possibly) calculate the winner without playing. However since we are talking about a lot of move options just by two available cards, the decision tree is way too big to actually calculate (with regular computers). Therefore many constellations won't have a noticable bias. I can live with this assessment.

That the game is more 'lost by mistakes' than 'won by ingenuity' is a common phenomenon among all 2 player strategy games with perfect information and no luck involved. For kost of those games, the initial situation is usually the same in every game. The variance onitama comes with is very unusual. It gives the game a lot of replaying value, I think. Looking forward to learning from you about the mistakes in midgame.

That is a perfect summary indeed, with a few exceptions, like the ones I mentioned before, the initial dealing of the cards has very little impact on the winner of the game, especially if both (human) players play at a good level. For beginners, because they are generally worse at defending and at developing their position, the “best” card might be more easily held into a winning position, but that does not happen with better players.

And I agree with you again, having a randomized initial setup is one of Onitama’s biggest strengths, and one of the main reasons I prefer it to games like Chess — it gives a lot of replayability and makes each game a slightly different “puzzle”, even for players like me who have played thousands of games. And to me, the fact that this is achieved in such a simple and elegant way, but at the same time without introducing a noticeable bias regarding the winner of the game, is a point that never stops to amaze me!

What makes Onitama significantly more unforgiving than other games is the fact that the board is so small and there are so few pieces, so even a small mistake can quickly snowball into a loss. That, coupled with how hard it is to calculate many moves ahead due to the dynamic nature of the moves, makes Onitama a very hard game to play perfectly for humans. And this is how better players can get a significant edge compared to weaker players, by playing more solid, positional and efficient moves and making less mistakes. Do not discount however some games won on brilliant moves or sacrifices (the ingenuity you mention), they do not happen very often but when they do they are beautiful!