First team match with public thoughts TWIXT PP

111 replies. Last post: 2011-11-12

Reply to this topic Return to forum

First team match with public thoughts
  • maraca at 2011-06-09

    Overview

    The team match will start on 2011-06-10 as described in the rules with the following teams:

    white: Carroll, Thierry Pertuy, maraca
    black: upendra, David J Bush, Klaus Hußmanns

    Tools: Jtwixt

    Jump to move (the reason why the game starts in a new thread):
    1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40

  • maraca at 2011-06-09

    Move 1

  • maraca at 2011-06-09

    (Somehow the anchor was removed, I hope an alternative works on move 2)

    I think Alan’s first move chart on http://twixt.wetpaint.com/ is very good. If we examine first league games, then it seems to be somewhere in the middle of the opinions of the players what a good starting move is:

    I would say there are too categories of starting moves:
    1) those along the more horizontal diagonal and
    2) those in the 'opponent’s territory'.
    2) is not often played and more difficult to handle. While a common answer to 1) is J10 (or a mirrored position), there seems to be no standard answer for category 2). You can look at the moves of category 1) like shifting the diagonals, that’s the reason I think why D3 is a little bit too strong, it shifts 2 diagonals and not only 1, when we go closer to the middle the moves have more influence towards the center, but maybe don’t help that much in the corner.

    I propose the following moves for category 1): J4, I5, K6
    and if we want to go for category 2): C10

    What do you think?

  • maraca at 2011-06-09

    Sorry G4 not J4

  • Thierry Pertuy at 2011-06-10

    Hi,

    I don’t understand the 1-24 row in this Alan’s chart. It seems nosense to me playing there, i.e. above or below our own line.

    I think I would have prefered a very classical D3 first move for educational reason.
    But it makes sense playing the most neutral first peg. So I am ok with G4.

    As for playing a 1) or a 2) category move, I’d say that I would play a 1) category against a player with a lower rating, and a 2) one against a player with a better rating. In such a case the goal would be to drive the adversary in some unusual game configuration, admiting that this could be an advantage (what I’m not sure of...)

  • maraca at 2011-06-10

    (Carroll told me he won’t be here this week and we can decide which move to play)

    The 1-24 row is there for completeness. I think it’s pretty safe to say all moves smaller or equal to 45 are losing moves and all moves bigger or equal than 55 are winning moves, so you can ignore them.

    My thoughts about when to play a move of which category are about the same as yours, however my experience with that strategy wasn’t too good :) I would go for a category 1) move here too.

    If you prefer to play a ‘standard’ game, then like you say G4 would probably be the best choice (the modern D3 ;-). This is actually still my favourite first move. I suggested I5 in case you think G4 is too strong. K6 or L6 would already be slightly experimental moves, because moves that close to the middle are also rarely played.

    So we go with G4?

    (@ the other team: I’ll post the Move 2 header when we’re done and a Jtwixt file if I find a place to host it)

  • Thierry Pertuy at 2011-06-10

    I have no experience with G4, and as you may already know I’m not very experimented in Twixt as I really play it only for 5 months. So I’m not able to say if it is too strong or not.
    So Ok. Let us go with G4 ! And will see if the other team feels it too strong or not...

  • maraca at 2011-06-10

    Ok then we go with G4!

    I don’t have that much time right now, but I think all of us understand how the current position looks like without an image ;-)

  • maraca at 2011-06-10

    (Btw. you can also give comments on our turn, it’s not forbidden ;-)

    Move 2

  • David J Bush ★ at 2011-06-11



    First we should decide whether to swap or not. I would prefer to swap, but whatever the majority decides is fine with me. Speaking of which, are we going with majority rule here?

  • Dvd Avins ★ at 2011-06-11

    I’m sure Thierry already know this, but since his remark could be misinterpreted, it bears explanation.

    Although playing behind your own line on Move 1 is surely bad, it’s not useless in the sense that passing would be if legal. For instance, a latter move could simultaneously connect to both that starting peg and to a peg closer to the middle of the board. It could do that in a situation where there was no other way to connect the edge to that more central peg.

  • maraca at 2011-06-11

    @David J Bush: I thought you wanna swap :) The majority rule only kicks in if you couldn’t agree on a move, here it might be difficult to convince the others why you wanna swap, but on later positions you can certainly bring some arguments why to play a certain move.
    Would you also agree with the chart that D3 is stronger than G4?

    @Dvd: that’s right. And I’m 100% sure that playing behind your own line is a losing move. It is already on 6x6 and the border between losing moves and winning moves has the tendency to go downwards and to the right (solved boards).

  • David J Bush ★ at 2011-06-11

    I believe D3 is stronger than G4, but of course I don’t know.

    We could use an empty board to examine variations for this game.

  • Klaus Hußmanns at 2011-06-11

    I prefer to swap too! G4 is too near to the center for leaving it to our opponents.

  • upendra at 2011-06-11

    Hi everyone – sorry to be late to the party, didn’t realize it started. :)

    (And let me issue this disclaimer once, and it applies to the rest of my posts – I just started playing in the past couple months, so I have no idea what I’m talking about. :)

    I would say swap:

    1st mover advantage and having an extra peg on the board is of course extremely powerful if the first peg comes into play; so another way of asking the question is what is the chances that G4 will come into play later in the game. Either directly, or as a deterrent than prevents certain lines of play.

    I would think that G4 would mean that we likely would win any corner battle on the upper left – it pushes the Achilles “heel area” south by 2 pegs, I think.

    Of course if the probabilities of Alan’s first move chart applies to this game, and the winning odds of G4 is 46% less than half, we shouldn’t swap. But given the environment of this game, with the open thread, this isn’t a typical game. :)

    Anyway, I say swap!

  • upendra at 2011-06-11

    Downloading Jtwixt for the first time; David, Klaus – are you mapping the game in JTwixt? Any way to share your files and notes and annotations?

  • maraca at 2011-06-11

    @upendra, you can share them via save-files, but I would prefer if you use the empty board like David suggested. I think I’m only gonna use the commentator, my eyes are used to that layout.

    Btw. the first move often already can be used when the 3rd peg is played. For example: 1. G4, 2. J10, 3. N11. Without the peg in the top left corner N11 would only be semi-connected, but with G4 it is more or less connected to the top.

  • David J Bush ★ at 2011-06-11

    The “empty board” link above is to Alan’s servegame site. It is not possible to make a swap move on that board. I propose we swap sides, so now Klaus, upendra and myself are now white, and maraca Carroll and Thierry Pertuy are black.

    Is that okay?

  • Thierry Pertuy at 2011-06-11

    @Dvd: Of course you are right. I’m just surprised that some people did try to start there and that some statistics can be done.

    @Klaus: I was sure you will swap :-)

    @upendra: As maraca said, one peg can have an influence very far away on the board because of ladders.

  • Thierry Pertuy at 2011-06-11

    Ok. So this our turn again...

    What about the classic j10 ? (classic answer to d3 – no idea at what could be the classic answer to g4 if any). j10 cannot be attacked with a 2-0 move because we can reach the diagonals going to the top or the bottom. I haven’t experimented a lot on the second peg (and when I did, it wasn’t very successful...)
    Now, is it too right ? Well I feel (more than I think...) it isn’t. Is it too left ? Well I don’t know.

    I must admit that the game is so open that I cannot really answer and that any real prospection is too wide to me...

  • maraca at 2011-06-11

    Move 3

    Move 2 was swap, color update:
    white: upendra, David J Bush, Klaus Hußmanns
    black: Carroll, Thierry Pertuy, maraca

  • maraca at 2011-06-11

    Image Hosted by ImageShack.us

    First a rule of thumb for beginners: Always play the 2nd move inside the inner octagon and on the same half of the board as the first peg.

    We shouldn’t play too fast now because of Carroll. I will post my thoughts later.

  • maraca at 2011-06-12

    @Thierry: I would say J10 is the classic answer to all the category 1) moves.

    For the 2nd peg I think there are again 2 categories:
    A) connected (to the border) or
    B) semi-connected

    J10 falls in category A), it connects to the left and it is also the position which is most towards the center in this category.
    The same is true for J15, but J15 is only good with G4 as first move or a similar one, NOT with D3. You have to be a little bit careful with J15 that you don’t end up on top right with a ladder going to the left, but it also prevents white from making use of the first peg already.
    K17 is somewhere between those two categories, it is connected to the right but very weak (also not good with D3).
    In category B) are all moves more towards the center, they need an extra peg to connect to the border. I played against those moves, but never used them myself, maybe David can tell us something about them.

    So I would go for category A) with J10 or J15. They both also have the advantage of being 1.5 above the diagonal, so in case there will be a ladder, an Achilles have to be started very early to be succesful. In my games I more or less alternate between those two answers.

  • Thierry Pertuy at 2011-06-12

    @maraca: Let me write it my way in order to check if I get what you mean. When you say that J15 is not a good answer to D3, I suppose that this is because D3 makes it not connected in case of a ladder starting with H15. Right ?
    I don’t undertand what you are talking about with “but it also prevents white from making use of the first peg already”.
    I’m not sure that I get you remark about the Achilles.

    Regarding category A), J10 looks to me more agressive, and J15 looks to me having more potential because it lets wide open the diagonal to the top right and the horizontal to the right. White are very likely to answer in the horizontal right area, and black along the diagonal.

    Regarding category B), I never used it neither myself. So, as often in Twixt, we should answer a way white will hesitate playing left or right. I’d say we shouldn’t play more right than L.

    My personal preference would go finally to J15.

  • maraca at 2011-06-12

    @Thierry: Yes J15 is not a good answer to D3 because of the ladder. The distance is very long, it would be wrong to answer immediately with H15 or F15 then some kind of mousetrap can be played, but the threat is there and later it can get dangerous.

    I also agree on category B) with not playing more right than L, this corresponds to the rule of thumb below the image.

    Look here for the Achilles Defense. The last section shows how an Achilles can be answered with another Achilles and so on, so if the others want to be succesful with that kind of pattern they have to attack a ladder to the right very early.

    The part with the first peg: Like already seen, a common opening is 1. G4, 2. J10, 3. N11 with connecting to the top because of G4. But if we play 2. J15 and white responds with 3. N14, this N14 is only semi-connected to the bottom.

  • Carroll at 2011-06-15

    Nice opening!
    I can not write much. I agree with J15 but out of curiosity, wouldn’t K16 give us more influence on right side or is it wrong regarding some sort of parity with the Achilles?

  • maraca at 2011-06-15

    @Carroll
    :) Thanks, we haven’t done much yet.

    Ok lets first look only at the power to the right side: I would say if the distance is that big, that one step closer to the border is less important than how much you are above the ‘horizontal’ diagonal starting/ending at W23. So I would conclude that the connectivity to the right is as follows: K16 < J15 < K15.
    If we now also look at the connectivity to the left, then K16 and K15 aren’t connected to the left, they probably need some kind of Achilles pattern that makes a new chain threatening to connect to the old one. I would say that full connectivity to the left is worth more than that little bit of extra connectivity to the right side.

    Ok looks like we agreed on J15

    (you can still ask questions about this move when the next one started)

  • maraca at 2011-06-15

    Move 4

  • maraca at 2011-06-15

  • Carroll at 2011-06-15

    @maraa: thx for the answer, it seems convincing. As K15 is at distance 8 1/2 from the border is there the equivalent of Hex templates known? or how far would be the first white stone that would threaten this firm connexion?

  • Klaus Hußmanns at 2011-06-15

    Possible moves are Q14, P14 and O15, maybe M14 too: M16->Q15 and M18->Q17 look good, but O15 is difficult to play: N16- or L16- are bad, and 6.P13 is very complex...

  • David J Bush ★ at 2011-06-15

    I have difficulty following your moves, Klaus. I’m not sure what your -> symbol indicates. Please play out variations here. Registration is free. Click on the board to make moves and generate a move list. You can copy and paste the move list into a comment. Be sure to include the | symbol at the beginning when you copy the moves. When you post your comment, the moves will be click-able. You can click on the moves inside a post to see the position.

  • Klaus Hußmanns at 2011-06-15

    The -> symbol should be a “followed by”. I wrote down my comment on Twixt Commentator too now.

  • Klaus Hußmanns at 2011-06-15

    The best move in a certain position has to

    1. prevent, that the opponent can get a chain from his last set peg to one of his two borders of the game

    2. most effectiv for the rest of the game (which means most often near the center)

    In our position there is no chance to prevent that J15 is connected to the left border. Therefor we have to block a possible connection to the right border as L15, M15, N15, O15, P15, P14 or Q14 and maybe some more. Now we have to decide, which of these moves has the best influence for the future; O15 and Q14 save us a connection to the lower border, and P14 has no direct access to the lower border because of P16, but lies more central.

  • Thierry Pertuy at 2011-06-15

    @Klaus: (3.m14) 4.o15 seems a strange answer to me. Why are (3.m14 4.o15) 5.n16 and 5.l16 bad ? Because of 6.m12 ?

    I’d say that to be more effective as possible a move has to :

    a. In the first half of the game, set up some influence or potential path
    or
    b. Set up a double threat
    or
    c. Prevent the opponent to do the same thing

    ...or, preferabily two of the three.

  • maraca at 2011-06-16

    @Carroll: No there are no templates. To connect safely (can change in the game later) to the left, if there are no obstacles, you can be max 1.5 right of the diagonal because of ladders, so the most right pegs that connect to the left (on the lower half of the board) are: I13, I14, J15, J16 the moves on 17th row and below to be really sure are something like J17, I18, H19, G20, F21 but it all changes if it is not only a single peg, but a link and if there are some other pegs near it, so templates wouldn’t help that much. It’s better to try to understand the concept behind a move than to learn the exact pattern by heart. For example like I already said about the Achilles Defense: make a new chain that is connected to the border and threatens to connect the other chain.

  • maraca at 2011-06-22

    What’s going on team white? Klaus made his suggestions but David you didn’t say which moves you wanna play or that you just agree with Klaus.

  • upendra at 2011-06-25

    I’m ok w/ 4. q14...

  • David J Bush ★ at 2011-06-27

    I posted a response on servegame. I am also willing to play Q14. I just hope that is what Klaus meant.

  • Klaus Hußmanns at 2011-06-27

    Yes, I agree: Q14

  • maraca at 2011-06-28

    Ok then it’s our turn now. 4. Q14

    Move 5

  • maraca at 2011-06-28


    I suggest we use a new board for variations, it’s easier to keep track then. LINK

  • maraca at 2011-06-28

    At the moment I can only see 3 candidates for the 5th move:
    Q12, Q11, P11
    I will comment more on them later.

  • maraca at 2011-06-28

    And maybe, O8.
    The idea is that we have to make a short distance block or a block far away, if we choose medium range like P10, then white can play N11 and we are in trouble because of G4.

  • maraca at 2011-06-28

    Ok, I posted some variations and explanations to the 3 moves mentioned above here. I think my current favourite is Q12, but I’m not so sure yet.

  • Thierry Pertuy at 2011-06-30

    Hi Maraca,

    I’m copy-pasting the comments I wrote in Twixt-commentator.

    I agree with your consideration about a medium distance block : 4.p10 5.n11

    As for white answers to 4.q12, what about
    5.m12 6.n15 7.o13
    5.m12 6.n15 7.n14 8.o17 9.m16
    5.m12 6.o13 7.m16
    ...and I feel it as being too “local” at this stage of the game.

    As for white answers to 4.q11, what about
    4.q11 5.l12

    I cannot see a good answer to 4.p11.
    As an alternative, I’d propose 4.o9 more than 4.o8

    But all in one I think that I’d prefer 4.p11

  • Carroll at 2011-06-30

    Thanks for your analysis Maraca, I would prefer a more centered move but that is because I don’t play well! I think O8 is a bit to “defensive”.

  • maraca at 2011-06-30

    @Carroll: Yes I also don’t like O8 that much. Normally we would try to play more centered, but white’s move forces us to play there.

    @Thierry: L12 is indeed an interesting answer to Q11, it could work. But M12 as answer to Q12 isn’t good imho, I’ll post the lines on the commentator.
    4. O9 is dangerous, because of Q9 and R9.

    Q11 doesn’t form a nice pattern with J15, nobody seems to really like that move and Thierry proposed L12 as possible answer. So the question is now Q12 or P11. They span both a similar pattern with J15, P11 has M12/M14 and Q12 has M12/N15. Those points are farer away from each other for Q12, which is usually better. We also have to take care if we play P11, because all moves near the diagonal from W2 to C12 can be dangerous for our potential Achilles at R5, that’s why I’m looking a little bit sceptical at P11. I also agree that Q12 looks somehow weak, strange and ‘local’ at first sight, but that could also be the case because we choose the opposing answer J15 to G4 and not J10.

  • FatPhil at 2011-07-06

    Can you guys make your mind up – some of us are trying to learn from you experts! ;-p

  • maraca at 2011-07-06

    :D Okay. Thierry is away for about 2.5 weeks now. So the decision is up to Carroll and me. Thierry wanted to play P11. I think Q12 is a little bit stronger, I also made some comments on the commentator. So Carroll you can decide which move to play, do we go with P11 or Q12? Or you can also ask for specific information about the moves first.

  • Carroll at 2011-07-07

    All right, as I can’t really make my mind between the two moves, let follow Thierry’s mood and play P11. Answers 2-0 like Q12 make me feel like “I don't have anything better to play so I stop you there!”.

  • maraca at 2011-07-08

    Okay, so P11 it is.

    I follow the principle, “simplicity means efficiency” (Bertrand Meyer). So if this position requires such a “we don't have anything better to do”-move, then I say: why not? (Q12 is easier to play because it’s connected to the right)

  • maraca at 2011-07-08

    Move 6

  • Klaus Hußmanns at 2011-07-08

    I propose 5.L12. Any other moves to discuss? And what are possible answers to L12?

  • Klaus Hußmanns at 2011-07-08

    I propose 5.L12. Any other moves to discuss? And what are possible answers to L12?

  • Klaus Hußmanns at 2011-07-08

    I propose 5.L12. Any other moves to discuss? And what are possible answers to L12?

  • upendra at 2011-07-08

    To make myself useful, updated a new board here.

  • Carroll at 2011-07-20

    I can’t see anything moving here or on twixt-commentator, is it because of holidays?

    I hope it is not a flop, I don’t want a win by forfeit.
    I understood the one-month delay was intended for discussions, not just keeping everyone waiting... or we’ll just play one public game in a life (which we should remember is always too short).

  • maraca at 2011-07-21

    Yes Carroll the delay was intended for discussions, normally it’s just a weak. I didn’t intervene because Thierry was away, so it was also good for our team if they take that long.

    Thanks upendra, you are the most active member of the white team :) Klaus made a suggestion, but nothing happened since then.
    Btw. team white, I already proposed answers for L12, you can find at least 1 possible answer on last move discussion.

  • David J Bush ★ at 2011-07-26

    5.L12 is fine with me.

  • Klaus Hußmanns at 2011-07-29

    OK! 5.L12

  • maraca at 2011-07-31

  • Thierry Pertuy at 2011-07-31

    Hi all !
    Here I am, back from my vacation.
    Time to boost up this game hmmm ? ;-)

    Imho, either we threat to cut in the center and then try to open a path to the lower right, either we try to block white on the top.

    I see only two possible moves :
    6.n15
    6.l10

    If !6.n15 7.o13, then 8.q18
    If !6.n15 7.u14 then 8.q18 again, but after 9.u19 I don’t know if black can connect (I doubt it)
    If !6.n15 7.u14 8.o13 but after 9.s9 I don’t know if black can connect (I doubt it too)
    ...maybe if we attack from both sides ?

    If !6.l10 7.h10 8.l14
    ...and I cannot see any better answer (but I’m just as you, Maraca... having some difficulties to be critical enough at my moves...).

    So I think that I’d prefer 6.l10...

  • Thierry Pertuy at 2011-07-31

    ...and thank you having waited for me ! ;-)

  • maraca at 2011-08-01

    Yes, those two moves are the most reasonable in my opinion too. There is also a

    6.l10 forms a nice pattern with J15, but we are a little bit weak on the other side, the Achilles this way is not that good.
    An example of exploiting this weak pattern would be something like a ladder to the left to strengthen L12 and then follow with the o9 (black n11) p12 combination.

    6.n15 is my preferred move. It has to be answered and it makes P11 stronger to connect to the right. I would leave the top open and play a move there later.
    Interesting answers are 7.u14 and 7.s13, because 8.q18 won’t work then (9.s18). We could go for a move like 8.p7 or try something on the bottom like 8.q19 then.

  • maraca at 2011-08-01

    Also possible is L9, exactly the amount needed above the diagonal, with the same idea as L10, but maybe a little bit better patterns on the other side.

  • Carroll at 2011-08-01

    I liked both 6.n15 and 6.l10

    I was wondering if 6.m14 would work:
    7.m15 8.n16 9.k14 10.l17 or
    7.l13 8.k13 9.n14 10.m12 11.p12 12.n10 13.q10 14.q7

    otherwise I guess the Achilles spoted by Maraca force to abandon 6.l10 toward 6.n15
    (link here)

  • maraca at 2011-08-01

    Yes M14 is the same category as N15, but there are some differences. M14 doesn’t give as much power to the right side (threaten to go south around Q14). That’s the main reason why I mentioned N15 and not M14. The advantage of M14 is that the pattern is more solid and it takes the spot (M14 is a good spot for white too). Now I’m not so sure any more which one is better.

    (btw 2nd line can be answered with 15.r8)

  • maraca at 2011-08-01

    M14 can probably be answered with P12 and I think that would give white an advantage, with N15 this isn’t possible, then we play Q18.

  • Carroll at 2011-08-03

    So do we agree on 6.n15 or do you (@Thierry) want to stick with 6.l10 ?

  • Thierry Pertuy at 2011-08-03

    Hi guys. Sorry for the delay.
    If 6.l9 7.i9 and ?
    Maraca has pointed out clearly the weakness of 6.l10 (I have to be more attentive to ladders and be carefull at double bastard...)

    So I’m ok with 6.n15.

  • maraca at 2011-08-05

    s8 could also be an answer to i9.

    Okay, we agreed on N15

    Move 8

  • Klaus Hußmanns at 2011-08-05

    I propose 7.I13 with later on S13. I guess, that I13 is better than 7.S13, because it helps to ensure a way to the top.

  • David J Bush ★ at 2011-08-06

    We now have, what, four different boards set up on Alan’s site for discussing one game??!! I think we’re being litterbugs. How about we don’t create any more boards. I posted a comment here.

  • maraca at 2011-08-06

    I just don’t see your problem with the boards. I already explained it and still you mention it everytime. Maybe you like to scroll and look for the beginning of the discussion of a specific move each time... me not. And littering? Do we pollute the TwixT commentator? I don’t think so. It is valuable information and no trash, you can learn something from it.

  • Thierry Pertuy at 2011-08-06

    In my humble opinion... I have no problem in changing boards but I must admit that keeping with the same one, althought it forces to scroll, would allow the reader to keep track of the whole discussion, and all variations will be easier to follow later for others. Maybe should I use the past however because it seems to be too late, now.
    If we switch boards, we should at least copy all our comments in this thread so that it will be complete.

  • maraca at 2011-08-09

    I tried to explain things here in a more general way and only used the commentator for some complicated lines or more experimental stuff... like Carroll did too, if I remember correctly. I certainly understand your argument, everything has advantages and disadvantages, we might just weight them different.

    Anyway I’m more interested in the game than such a minor issue, so let’s discuss about the game itself instead of other things.

  • Thierry Pertuy at 2011-08-14

    Well, David & Upendra, we are waiting for you... Do you agree with Klaus on 7.I13 or do you propose another move ?

  • David J Bush ★ at 2011-08-15

    7.i13 looks good to me.

  • Klaus Hußmanns at 2011-08-15

    okay, we play 7.I13

  • maraca at 2011-08-15

    Interesting move, then it’s our turn now.

    Move 9

  • maraca at 2011-08-15

    /img>

  • maraca at 2011-08-15

    This is what I call an in-between-move: It has to be answered, but it’s not at the ‘current burning point’ of the game and it’s often also not that strong for a local fight, but it can help later in the game.

    As already said, I think we have to answer this move, this implies we’ll have to play in the region south of I13. The spot of the move should be chosen according to the purpose the move should have:
    - minimizing the potential ladder on I13, that improves the connection to the top for white. This is not such a big deal in our case I think. Solid ladders are much worse, like that they have a weak spot on J11 should we be coming from right to left on top. G14 would probably be the move that minimizes the ladder the most. 2nd best in this category is G18 I would say, it allows to bring in 1 more peg on F16, but if we answer with E19 the ladder is ‘dead’.
    - maximizing the power to the right side. In this category I would put moves like K17, K18, J19.
    - Of course you can also choose a move between those 2 extremes, but I normally prefer to go for an ‘extreme’ move, except if I have a special plan.

    To be able to choose the best move, we’ll have to analyze the right side and how it plays out. Then we know where we need a peg or where they need them. I haven’t analyzed that yet, a good technique to do this with the commentator is to play a move in the corner and then a potential white move and look how it plays out, then when you see where the pegs have to be you can play out the left side like it would have been played before, like this you can test many scenarios on the left side for the move we’re actually interested in. The problem is you have to speculate / estimate the right side first.

    So here is what white probably plays next: S13 or U14. I’m not 100% sure but I think S13 is too weak no matter how we defend now. U14 is hard to evaluate, lots of holes and loosely connected pegs. Okay that’s enough for the moment, I have no favoured move yet, I first have to make an analysis, only those single pegs make me somehow nervous, with a link all would be more solid.

  • Thierry Pertuy at 2011-08-15

    Hi Maraca,

    Thank you for your analysis. I agree that this move must be answered and that there are two extreme ways to do so. What I’m wondering about is which is the extreme move which maximizes the power on the right side.
    Wouldn’t be 8.n20 enough to stop white on the left ? (even if it seems so far away...)

    On the left side :
    9.f16 10.h16 11.g14 12.g20
    9.h15 10.h19 11.g18 12.g21 13.i19 14.j20
    9.h16 10.h19 11.g18 12.f20 13.i19 14.j20

    On the right side :
    9.s13 10.q18 11.s18 12.u18 13.q17 14.s19 15.p19 16.p20
    9.u14 10.s18

    By the way, I haven’t completely understood your technique to analyse where they/we need a peg in the right side struggle. Could you illustrate it with an example ?

  • Thierry Pertuy at 2011-08-16

    I’ve posted some additionnal variation comments on the previous board on Alan’s site.

  • maraca at 2011-08-16

    About N20: This is right, it could be good enough, but it’s hard to say. This is an even more extreme way to answer, I would say the moves I mentioned were those, that more or less strengthen the current paths with max power to the right, N20 is a move that introduces like a new path south of the existing one. Although N20 is very weak generally (behind the diagonal), it could work because of the help from the 2 pegs above. Example follows soon.

  • Thierry Pertuy at 2011-08-16

    Well, I’m waiting eagerly for your examples.

    I think that the way you can win is to push answer to its extreme limit on one side so that you take maximum profit on the other side, and it can make the difference. The point is that the limit is sometimes hard to find out. Anyway, I prefer to go for “extreme” moves too.

    By the way do you think that there are flaws to the variations I have posted ?

  • maraca at 2011-08-17

    Many questions :) I made an example on the commentator, more later.

    Analyzing technique exlained in words:
    First you pass a move (if this isn’t possible, just play one in the corner). You should be losing now (if not, then you’ve already won the game). You play out the part away from the move that has to be answered, in this case the right side. You continue until you see you’re losing and how you could have played out the left side without losing if this variation is played on the right. All moves preventing the loss are candidates, also those that give especially nice patterns have to be remembered.
    Then you play out another variation on the right side and see if your moves are still working. This you do with some variations, until you think you covered the different main lines. Then you choose among the remaining candidates (those that could prevent the loss in any variation) the one who is a special candidate in most variations.

  • Thierry Pertuy at 2011-08-17

    Ok ! Thank you very much for the example. This is now very clear. Interesting technique although very time eating as you pointed it out ;-)

    I’m under the impression that s13 is not enough for white to connect to the bottom if we play any of our 4 “extreme” moves to the right. So I think they are likely to answer u14. I don’t know if this is enough to go to the top but I think we should examine this answer.

  • maraca at 2011-08-18

    Your welcome. I agree we should now looking on U14 responses. For S13 my favoured move would probably be K17, you know I prefer solid links over loose patterns. I’m still unsure about N20, an interesting move.

  • maraca at 2011-08-18

    I posted an analyzation of a new line here here.

    Reducing the ladder isn’t as easy as it seems, there are probably some mistakes about the best ladder breakers in my first post about this move. I think we should now focus on some possible playouts on the left side and decide which variations we like. It seems that they have to come at least onto or behind the diagonal with their ladder to win and the diagonal is not that far away...

  • maraca at 2011-08-18

    And of course the mysterious N20 has to be analyzed deeper too.

  • Thierry Pertuy at 2011-08-19

    I have had a look at your analysis and I cannot find out any white move which both threats enough to connect on the bottom left and provide support enough to counter your black upper line.

    I must say that I’m usually not enough confident in my play to project a line so far away. I’m always afraid missing something. Nevertheless I cannot find any flaw in your line. So it seems that it works even if I’m not comfortable with a so long line. If white answers differently on the right side, maybe the line may not come to the upper right Achille and all the line is broken.

  • maraca at 2011-08-19

    You couldn’t find any good move for white? That surprises me. So you think K17 is still fine?

    I changed my opinion and I think now that I17 is better than K17, because like that we can keep the ladder above the diagonal but still bring in that solid link there.

    New suggestion: G12 followed by J11, this means we’re attacking on top instead of on the bottom.

    N20 works surprisingly good together with the surrounding stones, I couldn’t find any real weakness, just my experience somehow says don’t do it... but David J Bush or Alan Hensel probably wouldn’t have a problem playing this move behind the diagonal.

  • Thierry Pertuy at 2011-08-19

    Hum. Must have been very tired yesterday... sorry for my foolishness...
    Will have a look at your last post a bit later.

  • Thierry Pertuy at 2011-08-20

    I agree with the need of a solid link with i13.
    I’m trying to summarize : k17 is not enough to prevent white to play a move in a way it threats to connect to the bottom AND it will support the ladder. i17 is enough.

    What bothers me in this whole story is that it depends on the reliability of a very long series of moves. You have got much more experience than I do (I’m still a newbie who plays Twixt for only 7 months who has played less than 200 games). I’m confident in the line starting from the Achille, but would I play by myself I would be unsure of the start of the sequence wich drives to the Achille.

    What about 11.o18 for example? 12.p16 13.r17 14.s18 15.t18

    I still like n20, and I don’t have enough experience to feel it as a “don't do it”. Actually, I feel it as “do it”...

    Anyway, it wouldn’t bother me playing i17 if you think this is definitively the best move.

  • Thierry Pertuy at 2011-08-20

    @Maraca : As I said on the Twixt commentator, so far, I agree that i17 seems to be the best answer even if I’m curious to see where the game will go now.

  • maraca at 2011-08-21

    Carroll, what do you think after the whole discussion? I’m now down to i17 and N20. Still curious about N20, but slightly preferring i17.

  • Carroll at 2011-08-22

    Yes I haven’t been able to find a way out of Thierry’s last line after N20 so I would vote for i17.

  • maraca at 2011-08-22

    Ah yes, I haven’t seen Thierry’s analyzation, well done. I definitely vote for i17 now.

  • Thierry Pertuy at 2011-08-22

    We all agree on i17. Let’s go...

    Btw, it’s interesting enough to see that this deep analysis has drived us to what I feel is the most natural/standard way to answer.

  • maraca at 2011-08-22

    Move 10

  • David J Bush ★ at 2011-08-28



    I just want to assure everyone I am studying the problem, and I will post more analysis as soon as I can. I believe we have reached the point where pure tactics take over, and there is a lot to look at. I posted just one possible line in the usual place. Thanks to maraca and everyone for posting there as well. I will try to be less of a jerk.

  • maraca at 2011-09-27

    It’s been a while since the last move here. For those who can’t wait I almost completely commented my recent game with Klaus, read more here http://twixt-commentator.dyndns.org/game/1364875

    @white team: You wanna start the discussion again, just vote for a move or not continuing this game?

  • David J Bush ★ at 2011-09-27

    Please let’s continue, I’m sorry for the delay. I will contribute to the discussion before Friday the last day of Sep. 2011.

  • Klaus Hußmanns at 2011-10-02

    I propose S13-. I guess it is a better move than U14 or T13...

  • Robert Irvin at 2011-11-06

    Looking at the commentator site for this game, it seems like you were trying to analyze the entire game before it ever even started. Is that the way you guys play in an individual game?

  • maraca at 2011-11-08

    Not always, I’m often a little bit lazy. The problem with this is that you can easily make a mistake or miss just some small thing and everything changes. But if I play in the first league and want to have a chance I’m almost forced to do that, even if I miss something it’s often better than just play by feelings or something else.

  • Robert Irvin at 2011-11-09

    Maybe that’s why I can only hover around the first league but never break into it. I might analyze a corner battle, but I usually just go with my gut. The numerous times you’ve beaten me, have you just been playing it out in your head without all of the super in-depth analysis?

  • maraca at 2011-11-12

    Depends. In the early games here I didn’t use any help. Later I used a program, I have forgotten which one, but I used it exactly the same way as the commentator, so when the commentator came the program was not nessessairy any more.

    First I became better from making mistakes, remember them and try to avoid them in the future. Then I became better by learning to see the good moves faster. And with the commentator this is easier and a lot faster, that’s why I always use the commentator. Just clicking a little bit around until I think I’ve seen it, that can be from some seconds to some minutes.
    If you often only see 1-2 good moves it’s easy and fast to make just some playouts, good moves and variations that are likely to happen seem to accumulate, with that I mean they appear often in the playouts. As I said the drawback of this is that you can easily miss something if you are not very careful, because there might be other good moves.

Return to forum

Reply to this topic




Include game board: [game;id:123456] or [game;id:123456;move:20] or [game;id:123456;move:20;title:some text]