I am playing a game against a university class in spain in which I have agreed to comment on why I am making my moves. I was thinking perhaps if you all like you could also make comments on the game or on my comments and I will let them know to watch here also. Its a class of Jose M Grau Ribas.

Wonderful initiative. Thanks, and good luck!

I'm afraid I'm new to this site. Where are the comments

going to appear? Here in the Hex forum? Someplace else?

++ kevin

This sounds like a great idea, but I think there's no way for us to read the comments unless we're one of the players:

http://www.littlegolem.net/jsp/game/game.jsp?gid=90921

Go to game 90921 and at the list of moves klick on the odd numbered moves (1,3 and 5…) and you will see my comments appear above the board.

Bill

Good idea, Bill. However, I think the comments must be made in the forum - preferrably under this topic. Only players can read and make comments on a game, I believe. I guess that will be one of the advantages of a wiki. If you say that others may comment the game also, don't be surprised if I do :-)

Marius

I'm also playing a game against the (same I suppose) students of Grau Ribas. It's game 91016. I agree with Marius that a wiki would make it possible for others to see mine and my oppontents' comments and also let others comment on the game. Now, the only options are for me to copy the comments at a forum thread, one at a time or all together when the game finishes.

An amazing thing I just noticed: The 4 first moves of the two games are identical!

Are these students the people that “invaded” the Dvonn rated-tournaments the other day? :)

Same moves in reverse, the students are on opposite sides of the two games.

If they are doing that, and they are clever, they may be

using a well-known “man in the middle” spoof, and actually

be playing LeBoeuf against Ypercube, guaranteeing they win

at least one of the games.

ok, my comments: move 1. The first move is the most critical move. A3 and a2 are time tested as being almost equal whichever side you play on and since your opponent has his choice of sides it is critical to keep things as equal as possible. Too far away (a1) has been shown to lose, and any closer to the center (even b3) is a huge advantage after the swap. Basically, strong players almost always play on the A column (but not a1) or on the second row (but not l2 on the short diagonal which is too strong).

Move 3: D6 is an interesting move. After I play in the center, it will support either a sweep along the top or a sweep along the bottom of the board. Now where do I play in the center? Well I want to play not too high which would give you too much room on the bottom and not too low leaving you too strong on the top. So I will try G7 right at the center.

Move 5: H4 is interesting. Now I must play on the right (i4 or j4 or k3 or k4 or j3) or to the left (f5 or e6 or g4 or g5 or e5) or play a waiting move in the middle (h5 or h6 or g6). If I play a waiting move, often a waiting move in response is good such as h3 or i2. The obvious move is often a trap, but I will try the obvious block F5.

Yes…except that it doesn't require that they are all that clever.

Kevin, you're right that the “spoof” will guarantee at least a win in usual play.

However, if one of the opponents (me or Bill) sees that he loses, he may resign and then the other may make a serious blunder in the following moves. In that case the “clever” students will win both of the games! But there aren't many chances that Bill will make any blunder at the stage where we see a lose.

On the other hand, the “clever” students may lose both games! From time! Or one game from time and the other because they would had lost one anyway. If both Bill and I reply all our moves at more than 36 hours, then the students will be constantly losing time at their clocks comaping it with our loss of time. Little by little at every move, yes, but a hex game can last more than 50 moves…

A note to the real students: I mean not to offend anyone of you by these remarks and I assume that you are not “spoofing”.

What is this “spoofing” you're talking about?

“Spoofing”:

If you're playing against two opponents who are both better than you, then winning one game and losing the other increases your rating. Now, say you're starting against one player and are second against the other. Then you wait in the game you're first until the player in the other game has made his move. Then you make that exact same move in the game where you're first.

Now, you wait for your opponent in that game to give his response. After he's done that, you use that response in the other game. This tecnique ensures that the two games become identical, and thus you'll win one of them and lose the other. If you make sure you're always logged on around 36 hours after your last move, you won't lose on time either.

I suppose this is one of the few ways you can cheat at Little Golem. However, I wouldn't accuse these students of using this tecnique - it could be a coincidence.

Marius

If the students play Bill's 5.F5 against yper, I would start to get suspicious. However, even if they do log in again within 36 hours of making each move, that does not imply their opponent in game A will have made a move for them to copy into game B. So, such a plan could be defeated if opponent A were to deliberately let his time run down past the point where it would be restored to 240 hours with his next move. Then, when he does reply, any delay by the spoofer in relaying this move to game B would result in more time lost on his clock on game B than opponent A lost on his clock. This technique could be applied during many moves, or it could be applied all at once. If player A waits until he is almost going to lose on time before making a reply, that may be too late for the spoofer who would have already lost on time in game B. I'm not suggesting either yper or Bill actually do this, especially since it would mean one of them would have to use up all his vacation days just to defeat the spoofer, but if the students continue to duplicate moves, perhaps this counter-strategy could be mentioned to them. I'm sure José would not want his students to do this, anyway, since it teaches little.

I am also playing the students ( 91365) but we haven't gone past my swap of their 1.A3.

Now that I have looked at this issue a little more, I am even more suspicious:

\* Bill's game 90921 is against “Universidad de Oviedo” which has never played any other game on LG, whereas yper's 91016 and my 91365 are both against “Matemáticas y Juegos” which has never played any other games. Why would different names be used like this, unless someone wants to hide what he is doing?

\* In his invitation to me, I was warned that I would be facing strong moves. Was a similar warning given to either of you, Bill or yper?

\* Maybe things are not what they seem. Is this José doing this, or is it someone pretending to be him? In the invitation I was asked “Do you remember me?” while at the same time I am playing a game against José.

BTW the spoofing strategy can be defeated without either yper or Bill losing any vacation days. Since the games are unrated, the simplest approach might be to resign immediately. I may do that in my game if I find that my game is paired with another game that some other strong player is playing against “Universidad de Bromas Prácticas”. But here's how both games can be won against the spoofer:

\* Always wait beyond 36 hours to make your move. If you ever move more quickly than that, you give the spoofer the opportunity to regain lost time.

\* Experiment. Find out what time of day you can move which will result in the greatest delay before the spoofer relays this move in the other game.

If the above policy is adhered to throughout the game, and both yper and Bill agree to make lots of time wasting moves to up the move count, the cumulative effect of this will result in a loss on time by the spoofer eventually. Then of course, whoever still has a game going merely has to win his game normally.

Another approach would be for yper or Bill to make some ridiculous move and see if it gets copied. This early in the game, the spoofer may not be able to turn any resulting advantage into a win. This does not necessarily defeat the spoofing stragegy, but it would settle the issue of whether it is being used or not.

Bill, thanks for your excellent comments so far. Here are my questions… for you and anyone else who wants to answer. Please understand these are questions from a beginner trying to learn:

—>move 1. The first move is the most critical move. A3 and a2 are time tested as being almost equal whichever side you play on and since your opponent has his choice of sides it is critical to keep things as equal as possible. Too far away (a1) has been shown to lose, and any closer to the center (even b3) is a huge advantage after the swap. <—-

I had always wondered why anyone would play on the A row with the first move. It seems you are saying it's because any move closer to the center will likely be swapped? If so, this is the most balanced and “fair” move.

—–>Move 3: D6 is an interesting move. After I play in the center, it will support either a sweep along the top or a sweep along the bottom of the board. <—–

What do you mean by sweeping? D6 looks good to me for two reasons: 1) It blocks your chaining path, and 2) It sets up a chaining path for him.

——>Move 5: H4 is interesting. Now I must play on the right (i4 or j4 or k3 or k4 or j3)<——

Why would J4 be a strong move, since a J3 follow-up by your opponent would block you and leave him closer to victory?

—–> or play a waiting move in the middle (h5 or h6 or g6). <——

What do you mean by “waiting”? Also, how are any of these moves strong, especially the ones that bring you into contact with your own red stone? They are obviously setup moves for something later in the game, but my feeble mind does not grasp them. ;-)

——>The obvious move is often a trap, but I will try the obvious block F5.<—–

Well, you're in trouble, because that's the move I would have made! :-)

Thanks,

Rex

Rex, some comments on 'sweep' and 'waiting move':

Lets say the player connecting top to bottom makes a move in the center. Then the other player must connect the left to the right side by either going above or below. By sweep, I just meant the process of setting up a connection between the sides. Now, to set up a connection above or below, one must make a move above or below, respectively, and this could also be seen as a 'block' of the opponents move. Now, if first player plays to center and you play (block) above him then he will play (block) you either to your left or to your right and connect (sweep) top to bottom from your left or your right.

A 'waiting move' is the idea that instead of blocking your opponent on either the top or the bottom, you play to the side which looks wasteful but which allows you some protection whichever side your opponent plays to. I know this may sound confusing, once you see it work in practice, it becomes very clear.

Cheers, Bill

And regarding swapping, its exactly like the old children's rule: One person cuts the pie, the second person choses which piece to take.

With a waiting move, I am thinking of a tennis player who keeps his racket in the center until he sees which alley the ball is going to come down. Of course, once you play your waiting move you can't switch its location but it will be of some support on both sides.

Regarding moves G6 and E7, both are pretty natural moves trying to extend oneself and block opponent.

Regarding move E8.

This move blocks my E7. Now I want to find a way to connect G7 to the top and then fight to connect to the bottom. There are three ways to do this: the obvious F7, or either H6 or I6 whense I can run to the top either on the left or the right of h4/g6. The choice here is often critical with only one move being viable (and usually not the obvious move). However, in this case, it appears that any of these moves is viable so I will just play the obvious F7.

Regarding move G8.

This is a common situation where you block to the side (G8), I play to the middle (F8) and you block below me (E10).

Your choice was to block now or to play any or all of the sequence of moves F9…G10…H11. This sequence would take the battle progressively away from the center and towards the edge of the board. Sometimes it is absolutely essential for you to bring the game to a certain proximity towards the edge. Although in this game my position looks viable with any of the above.

Bill, continued thanks for posting all of this… it's very helpful.

Rex

Regarding E10 and H9:

E10 is the natural block following which the key offensive square is H9. Why? Because from H9 I can connect upwards on the left or try to fight my way up on the right side. Of course, I can also be blocked from below and following that I will make another move to the pivot square and again connect up on the left side or try to fight my way up on the right side…as we shall soon see.

E10 is the natural block following which the key offensive square is H9. Why? Because from H9 I can connect upwards on the left or try to fight my way up on the right side. Of course, I can also be blocked from below and following that I will make another move to the pivot square and again connect up on the left side or try to fight my way up on the right side…as we shall soon see. ««

Bill or anyone,

1. Which hex(es) would allow you to “connect upwards on the left”?

2. What is the “pivot square?”

Thanks,

Rex

Rex, lets see if I can clarify this. H9 is the pivot square. Because I can fight up on the right starting with I7 and if he blocks this move I can go to the left by playing F9 (he blocks with F10) and I play G9. So whatever he does I can fight to connect H9 (the pivot) to the top. (pivot because I can play upwards to the left or to the right).

Cheers, Bill

Does that help?

Bill, it all makes sense except for the opponent blocking you with F10, because it seems G9 would be a better block once you've played at F9(at least the way I visualize it ;-)).

Thanks,

Rex

Rex, If the opponent plays G9 after I play F9, then I play F10 and F10 is through on the bottom and through on the top and game is over. No?

Yes, I think I just got turned around there somehow… thanks.

Regarding I7 and F9:

He blocks on right with I7 so I move on left with F9 to connect to H9…thus H9 was destined ( by virtue of being the pivot) to connect to the top and next will try to connect H9 to the bottom.

Regarding F10 and G9:

F10 blocks and then g9 connects the pivot point h9, now he will block again and a new pivot point I11 will arise, From there he must block on the left and I will to J8 and then K6 in a bid to connect to the top.

You may note that I could simplify this a little by playing J8 right now instead of G9. Playing g9 now just makes it clearer how to reach J8 if you dont see that J8 is already connected to the bottom.

http://www.gratisweb.com/hex1//Bill.htm

regarding H10, G10:

H10 defends, G10 attactks, F12 will defend, then I11 will be the new pivot point from which I will fight my way to the top on the right.

An Overview/Digression:

This is a very interesting exercise commenting on all the moves of a game. In the act of teaching one becomes clearer oneself on the subject being taught.

In this game you had the first move and, therefore, you had a theoretically won game. Yet I am thinking that after your first three moves D6, H4 and G6, my job seems simply to be to explain to you why all subsequent moves are forced and thus why you had already lost the game from the third move.

How can this be? Well, I think there are three possibilities here.

1. I am making some simple oversight (which I often do) and I have actually lost or there is much fighting/thinking left to be done.

2. There is a very subtle oversight on my part and (again) I have actually lost or there is much fighting/thinking left to be done. Or

3. You lost this game in the first three moves.

If the the correct answer is 3., then obviously the first three moves are very important. How should one play the first three moves? Since you had a theoretically won game you should try to play into a simple won position. But after I start with A3, we don't know of any simple winning sequence of moves. So what is the next best thing to do? Well, for sure, try to avoid playing into a simple lost position. This leaves playing into an unclear position but in which your side seems to be as strong as possible.

This leads into two other points:

It can be very helpful to know what sort initial moves strong players have made in the past and how they have fared (so-called Opening Theory).

If you have not followed forced lines through to their conclusion, you will not fare well against someone who has.

Couple of thoughts I have. First, it is possible that one (or both) sides are missing the best move.

For instance, I personally wouldn't have played 12.e10, 12.e11 seems better to me (attempting to force red to the smaller left side). After all while e10 may be the “natural” block, it doesn't work

It also seemed folly to enter the same sequence with 20.f12, but there doesn't seem to be any other option (the similar move of f13 doesn't work because red is too close to the edge now after g11).

And in the current possition (after 21.i11) I'm not sure how red can connect to the top on the right, so something like 22.h11 seems solid to me. Which I assume leads to 23.j9, 24.i9. And I'm not sure what red does to get around the piece at i7.

I'll admit I'm not going into great detail looking deep at the lines, but it seems like there is some play here.

- Charles

Hello Charles:

I11 can be protected but there is not defense against C11.

In the north neither is defense. A3 again has surprised me; the weak players as me underestimate the importance of this move. Espero haber aprendido la lección. I wait to have learned the lesson.

Thoughts on A3:

Of course, the point of A3 is that it is neither weak nor strong but almost exactly equal and thus neutral, such that the game remains almost neutral whether the second player swaps or does not swap.

As proof of this note how David Bush is able to win both with and without swapping A3.

Of course, ultimately, with perfect play, A3 (and every other move) is a dead loss one way or the other. But we don't even know which way yet, let alone how to carry out the perfect play.

Of course, currently, David for one tends to Swap A3 when given the choice.

Cheers.

Regarding the choice of moves H2 or H11 (G12):

Sometimes switching the area of combat is critical to winning. Here it makes no difference although H2 is an attempt to confuse the opponent. However, if you see that the reply G3 sets up a 'ladder' towards A3 such that with proper handling (see any of David's many games based on a third row ladder to the A3 stone) G3 is through to the top, then the game is the same with or without the (H2,G3)diversion.

Let us call H11 the mainline. Actually this move is a very common mistake(inaccuracy). When playing up to a pivot point the correct move is an automatic G12 rather than H11. Why? Because, if you can ever fight under the pivot point to H13 you will be connected with G12 but not with H11. Of course, in this game its an instuctional point only (it doesn't change the outcome).

I should add that since so many games are A3 games, the third row A3 ladder is quite critical to understand. Actually its quite simple: simply reply B4 when the play gets to E2.

Regarding H3,G3.

H3 is an interesting alternative to blocking me on i9. Now I could try continuing my development at the bottom with i8, however I will play G3, similar to my response to G2 in your other game and again have the potential for a third row ladder to a3.

regarding i9,j8:

Now for these two moves and for the following pairs of moves, you will be blocking and I will be trying to advance to the top up the right side of the board. And we will see whether your blocks or my advances are stronger.

First, I didn't mention this last time I posted, but this game (and the discussion) is great. Hopefully I'm not butting in too much.

Couple of notes, first on the game. Isn't g3 just connected (no ladder needed) through h2/f2. Unless blue makes a double threat type move in that area of course. Then the ladder may come into play.

On the right… its tight. I just spend a minute or two looking at some variations, and if blue can defend, it will take perfect play. And even then I'm not sure he can defend. I think I saw one line where he defended, but I didn't look at all of reds possible counters.

I want to ask about the level 3 ladder. Lets assume (for argument sake) that a ladder starts here, with g2 by blue, then f3 by red, so f2,e3,e2. If red plays b4, then blue plays d3, I'm not sure how (in isolation of other pieces) e3 gets gets connected to b4. I assume there is a way, but I'm not seeing it.

If you are talking about d3, d2, then b4 is a win for red. Or d3, d1, b4, b1, c1 is a win for red as well.

Charles

regarding i8,k6:

block (i8)

try to extent (k6)

Charles, I appreciate your questions.

Yes, this game is very close, if the students play perfectly as you say and block me from the top it may continue to be very instructive.

Regarding a ladder:

A ladder is when you play along the row and keep getting block from connecting, going from one side to the other. In this case we would be moving from right to left towards the a3 square (him on the second row, me on the third row).

When play get to E3(me), E2(him), then the play goes B4(me), D3(him), C5(me). Now note that I am not exactly through but that my position is so strong that I win easily anyway (if breaking through on a ladder is what I needed to do).

In this case we would be moving from right to left towards the a3 square (him on the second row, me on the third row).

When play get to E3(me), E2(him), then the play goes B4(me), D3(him), C5(me). Now note that I am not exactly through but that my position is so strong that I win easily anyway (if breaking through on a ladder is what I needed to do). «««

Bill, I see how this connects. But why not just respond to E2 with D3 (the sure win)? I assume I'm missing something about the ladder demonstration.

Rex

Regarding playing D3 after E2:

Doesn't d3 lose rather than win? how does it win?

Rex,

I was thinking the same thing but…

d3 loses after d1 (blue needs to go to d1 since it blocks the ladder from getting to the top, and it invades the area that a3 will use to get to the top), b4, b1.

Red only has one option, which is c1. Then c3 by blue. Now red probably plays b3, blue c4. And that's that.

Note the final position is worse than the one from Bill's line. So while the b4-c5 line is not fully connected, it is a really stong position. Personally I was looking for a “by these X moves the ladder connects”, but it can't be proved exactly that way. In this game, if there is a blue piece connected to d6 towards the top (d5, e4, e5), I don't think red can connect to the center. Or at least can't do it in an obvious manner.

Which is what I'm taking from this part of the discussion. In order for a3 to be a escape for the level 3 ladder, there has to be some way to connect c5 to the center when combined with the threat to hook up to the ladder. So I guess the strategy to defend it is to keep in mind that having the outside position in that area of the center is good. Since hex is so close (when equal players are playing), if you can eliminate the value of the first move you will win. So the next level of thinking is how to do that in the opening, and not handicap yourself. And d6 looks like a good start. ;)

Charles

Regarding J5,L4:

J5: blocking, L4:continuing to advance.

Now we are at the end stage and it is very close. It looks like for me to break through in the coming variations, I must first try to break through to the J4 square.

Rex,

I was thinking the same thing but…

d3 loses after d1 (blue needs to go to d1 since it blocks the ladder from getting to the top, and it invades the area that a3 will use to get to the top), b4, b1.««

Man, this is tremendous learning. Thank you both. So much of this game is, for me, learning to think in non-conventional ways. So rather than heading for the “ladder breaker” (A3) as in Go, in this situation you have to stop your progress toward it and think creatively.

So the next level of thinking is how to do that in the opening, and not handicap yourself. And d6 looks like a good start. ;)««

Which is a hex already occupied by blue… but a fat lot of good that did. ;-)

Rex

The Spanish team of students has stopped playing in the game against me. Why? I protest! They should dtart playing right now and kick my butt!

Rex,

I'm still learning as well, which is why I posted what I did. One key I'm seeing in hex is that after about 20 or so moves, most of the moves are completely forced. Unless I play this move my opponent will connect to the top/side, I must play to block it. This is really true during these sort of endgame/edge connection position.

I'm pretty sure that right now in this game, its to this point. Red is trying to hit the top, so blue has to stop it. This group on the right is (I think) the only chance has to hit the bottom. (Obviously, I could be wrong and there is a way to connect to the bottom on the left, but if there is it is already a lost game, so no need to worry about it.) So if blue can totally stop this group from connecting to the top, blue will win.

My technique for looking at these sorts of positions (and I'm curious how others are doing it), is to look defensively. Like, as blue I'd say “if red plays to K3 I'm finished, so what can I do about K3?” Then I'd do the same as red “ok, K3 not an option, how else can I make a forcing move that my opponent must block”. And so on. This position gets resolved after 10 or so moves. So I would know as blue if I should resign (though if I didn't know if my opponent would see it, I might keep playing), or start planning what I'm going to do when my opponent is finished here. In this game, blue doesn't have a connection to the left, so he needs to start planning what to do over there. And the inverse is as red, if I see I can't win here, I want to move over to the other area of the board as quickly as possible.

Also, my d6 comment was in the opening, not for this specific game. In fact, the d6 isn't good enough in this game, becuase blue has nothing else backing it up.

Charles

ps. This is really a great position to analyse, by the by.

Charles and Rex,

Your comments and questions are most helpful to me too, in trying to best explain this game to the students.

Charles and Rex,

Your comments and questions are most helpful to me too, in trying to best explain this game to the students.