I seriously don't think thirty seconds, ten minutes, or fifty years on Rybka is going to give you the best moves in all openings. And while there's nothing wrong with those four moves, there are plenty of other good variations, too. Unless every other move is substantially worse, I don't see how I am supposed to solve the puzzle.
Smarter than a patzer?
Sort:
Eebster wrote:
tonydal wrote:
My opponent played (while I developed) 1 f3 and then walked his king back and forth between f2 and g3 for several moves while I developed (he may even have actually played it to f4 one time, since I played a KID setup). He won a pawn later on and I capitulated. This was a 5-minute game on ICC (my only consolation is that perhaps I was actually playing one of those legendary rumored games against Fischer). :)
Out of curiosity, why did you play KID against that? I mean, I like the KID, but it seems a little unambitious when white is giving you the whole board.
@Elubas, look, the whole point is that we cannot know that white is "100% winning" unless we can solve chess. The best we can say is "White looks better to me, and in most games played from this position, white wins." But that doesn't necessarily mean white is objectively better.
Let me give a counterexample. Before endgame tablebases were calculated, several endgames, such as KBBvKN, were believed to be draws. Everybody believed this. In fact, in some cases, there were published analyses that people thought had proved some of these positions were draws. Yet when the tablebases were calculated, we discovered that some of these endgames were actually winnable, but in a very large number of very precise moves. So our intuition and analysis was wrong--objectively wrong. There is no reason it can't be wrong again here.
In my opinion (and many other strong player's when evaluating positions) clear evidence pointing towards something while none towards anything else (like winning a knight for some reason gives the guy good drawing chances in most positions, no reason to think that) is good enough. And you need that to try to get into good positions in real games.
In post #51 I don't see what is so critical about 1. Bd3. Is 1. e5 going to get you in so much trouble?
Jan 22, 2010
0
#45
Eebster wrote:
And while there's nothing wrong with those four moves, there are plenty of other good variations, too. Unless every other move is substantially worse, I don't see how I am supposed to solve the puzzle.
Excellent point. Thank you. In future puzzles, I will let players know when there are other good variations. The problem that I have with the puzzle maker program on this site is this: I can't allow for two solutions to be correct.
Maybe I am missing something, but is there a way to post a puzzle that permits two correct solutions?
Regarding engines helping the beginner in post mortem analysis of openings: I read chess books on opening theory/opening principles. That's absolutely necessary for me - and it has helped. Engine analysis helps when I get off-beat responses from Black that I don't have a good reply for, yet. I'm sure that as my understanding of opening theory deepens, my need for engine analysis in the opening phase of the game will diminish. Finally, as I slowly climb up the ratings ladder, I look froward to seeing less and less of these patzer opening moves.
Please remember, these patzer moves are directed at helping the beginning chess player like myself. The view count on this thread is very high, so I will continue to make patzer puzzles. Any suggestions for improving upon my puzzles for beginners would be most appreciated.
Eebster - Thanks again for your kind attention in pointing out the weakness in my last puzzle. I hope the beginning player still found it instructive. I'll post the solution with the variations shortly. 
Jan 22, 2010
0
#46
Loomis wrote:
In post #51 I don't see what is so critical about 1. Bd3. Is 1. e5 going to get you in so much trouble?
"For the beginner, the first move is extremely important to get right, otherwise Black can surprisingly have a very good game." - Musikamole
Rybka placed e5 as a very close second to Bd3. When I said "get right", I didn't mean that there was only one good move. I need to be more precise in my wording on future puzzles. Sorry about that.
Thanks for taking the time to look at puzzle 3.
Jan 22, 2010
0
#47
Solution to Puzzle 3 with one variation. I don't know if there's a name for this opening. I call it the zig zag defense. 
At my level of play, I've seen it too many times! Surprisingly, Black can have a good game after this move sequence. If I'm missing a move for White that better deals with this goofy looking opening, then please post it!
Elubas wrote:
Eebster wrote:
@Elubas, look, the whole point is that we cannot know that white is "100% winning" unless we can solve chess. The best we can say is "White looks better to me, and in most games played from this position, white wins." But that doesn't necessarily mean white is objectively better.
Let me give a counterexample. Before endgame tablebases were calculated, several endgames, such as KBBvKN, were believed to be draws. Everybody believed this. In fact, in some cases, there were published analyses that people thought had proved some of these positions were draws. Yet when the tablebases were calculated, we discovered that some of these endgames were actually winnable, but in a very large number of very precise moves. So our intuition and analysis was wrong--objectively wrong. There is no reason it can't be wrong again here.
In my opinion (and many other strong player's when evaluating positions) clear evidence pointing towards something while none towards anything else (like winning a knight for some reason gives the guy good drawing chances in most positions, no reason to think that) is good enough. And you need that to try to get into good positions in real games.
Oh, certainly. We can definitely say that white looks far better. That is just not the same thing as "100% winning," which implies that we somehow know for certain that white has a won position.
Musikamole wrote:
Eebster wrote:
And while there's nothing wrong with those four moves, there are plenty of other good variations, too. Unless every other move is substantially worse, I don't see how I am supposed to solve the puzzle.
Excellent point. Thank you. In future puzzles, I will let players know when there are other good variations. The problem that I have with the puzzle maker program on this site is this: I can't allow for two solutions to be correct.
Maybe I am missing something, but is there a way to post a puzzle that permits two correct solutions?
Regarding engines helping the beginner in post mortem analysis of openings: I read chess books on opening theory/opening principles. That's absolutely necessary for me - and it has helped. Engine analysis helps when I get off-beat responses from Black that I don't have a good reply for, yet. I'm sure that as my understanding of opening theory deepens, my need for engine analysis in the opening phase of the game will diminish. Finally, as I slowly climb up the ratings ladder, I look froward to seeing less and less of these patzer opening moves.
Please remember, these patzer moves are directed at helping the beginning chess player like myself. The view count on this thread is very high, so I will continue to make patzer puzzles. Any suggestions for improving upon my puzzles for beginners would be most appreciated.
Eebster - Thanks again for your kind attention in pointing out the weakness in my last puzzle. I hope the beginning player still found it instructive. I'll post the solution with the variations shortly.
I wish this site allowed for multiple solutions to puzzles, but I don't think it does. But traditionally in puzzle composition, there is only one key move. When there is more than one key, the puzzle is "cooked." Of course, your puzzle is a little different, since it is just focusing on making sound opening moves rather than objectively "best" moves, so it would benefit from having multiple solutions. Probably just inserting it as a game like you did in your last post is best.
As for using a computer, I use it all the time for looking at tactical positions. But usually early on in the opening, the engine isn't very reliable. Even if you don't know much opening theory (I hardly know any), you can usually still find good moves online or in a book. And if you can't do that, they will often be pretty logical. Defend, develop, control the center, castle, etc. As long as you don't make big mistakes, you will usually come out of extremely nonstandard openings like these OK. If that weren't the case, those openings would probably be played more!
However, your point that black can still get a good game is valid and important. At most levels it is very difficult to punish a player for a bad opening (unless they make an opening blunder), because the advantages coming out of the opening for either side are usually very small. That's part of why computers don't analyze them very well and why beginners usually benefit more from working on their middlegame than opening.
Jan 23, 2010
0
#50
Eebster wrote:
Musikamole wrote:
Eebster wrote:
And while there's nothing wrong with those four moves, there are plenty of other good variations, too. Unless every other move is substantially worse, I don't see how I am supposed to solve the puzzle.
Excellent point. Thank you. In future puzzles, I will let players know when there are other good variations. The problem that I have with the puzzle maker program on this site is this: I can't allow for two solutions to be correct.
Maybe I am missing something, but is there a way to post a puzzle that permits two correct solutions?
Regarding engines helping the beginner in post mortem analysis of openings: I read chess books on opening theory/opening principles. That's absolutely necessary for me - and it has helped. Engine analysis helps when I get off-beat responses from Black that I don't have a good reply for, yet. I'm sure that as my understanding of opening theory deepens, my need for engine analysis in the opening phase of the game will diminish. Finally, as I slowly climb up the ratings ladder, I look froward to seeing less and less of these patzer opening moves.
Please remember, these patzer moves are directed at helping the beginning chess player like myself. The view count on this thread is very high, so I will continue to make patzer puzzles. Any suggestions for improving upon my puzzles for beginners would be most appreciated.
Eebster - Thanks again for your kind attention in pointing out the weakness in my last puzzle. I hope the beginning player still found it instructive. I'll post the solution with the variations shortly.
I wish this site allowed for multiple solutions to puzzles, but I don't think it does. But traditionally in puzzle composition, there is only one key move. When there is more than one key, the puzzle is "cooked." Of course, your puzzle is a little different, since it is just focusing on making sound opening moves rather than objectively "best" moves, so it would benefit from having multiple solutions. Probably just inserting it as a game like you did in your last post is best.
As for using a computer, I use it all the time for looking at tactical positions. But usually early on in the opening, the engine isn't very reliable. Even if you don't know much opening theory (I hardly know any), you can usually still find good moves online or in a book. And if you can't do that, they will often be pretty logical. Defend, develop, control the center, castle, etc. As long as you don't make big mistakes, you will usually come out of extremely nonstandard openings like these OK. If that weren't the case, those openings would probably be played more!
However, your point that black can still get a good game is valid and important. At most levels it is very difficult to punish a player for a bad opening (unless they make an opening blunder), because the advantages coming out of the opening for either side are usually very small. That's part of why computers don't analyze them very well and why beginners usually benefit more from working on their middlegame than opening.
I greatly appreciate your feedback. All that I placed in bold type has been of great value as I ponder my next move as a composer of puzzles.
1. When I find more patzer moves or traps in the opening, I'll post them as games, not puzzles.
2. I have an idea for a new thread which fits nicely with a book I'm reading on the middlegame by Silman. When this idea matures, I may start a new thread something like: "Find the best move in the middlegame". Not a catchy title yet, but a topic I find of great interest.
Thanks to all for the expert assistance in puzzle compositions and engines!
After the bishop's move White will either have doubled pawns on the c-file or will lose the pawn on the e-file. This makes 3. Bd3 better than bringing out the knight.
Eebster wrote:
Elubas wrote:
It's an example, which was my argument that although there are piece up position that are drawn, when one is up say two pawns for nothing out of the opening it should only go into some drawn piece up situation if he allows the opponent to trade off most pawns and sac his pice for the rest. However in most positions when one plays correctly that won't work (perhaps there will be still 5 pawns to three or something, when saccing a piece won't get rid of them all)
Yeah I've thought about this before, but that would probably only happen if the win of a pawn was maybe compromised by some bad feature in your position. The opponent doesn't have full compensation, but enough that he will be successful in a defense, and indeed get to that drawn position. These are all apporximations made by grandmasters and it seems to make enough sense. Right now I have a game where I have an extra central pawn, where that pawn can also gain space. But if I had a flank pawn instead and inferior king position at the cost of winning it maybe it's drawn. That would be a compromised pawn win. I'm sorry Kupov but I can't prove it, but it seems right. If it doesn't seem right, then I'd love to hear your take on it. I'm no expert on computers so I don't want to get into that, but all I'm saying is the way it works they make 1.5 (or 1.3 or something) means fritz thinks that it's winning, which is mostly useful in tactical positions where material gain is possible, as well as technical endgame positions.
But I don't want to debate on the details of that so much. I'm much more concerned for more pure chess discussion, like my second paragraph on why I think a pawn won in the opening without any compromise in the position has at least excellent winning chances.
I still think you completely misunderstand the meaning of "forced." +1.5 is "winning," yes, if you believe the chess engine. In fact, nearly every good player would agree that this position favors white, at least the way humans play. That doesn't mean that white will necessarily win, or that a true evaluation of the position can be determined. While you may think that if neither side makes a mistake, white will win, you can't prove it. Maybe the game could go perfectly and black could still checkmate in the end. The win is NOT forced.
Besides, black does have some compensation. She has developed her queen. Sure, that doesn't seem to be anywhere close to equal, but how can you know for sure? Obviously you can't.
It always has to be about my understanding. It couldn't just be that the word "forced" is ambiguous. Some people might talk about "forced" when it comes to a short term tactic, while others mean that, 70 moves later, you will lose to best play in all variations. In some sense that's "forced," because you are not able to get out of it, you are "forced" into that result. (Although granted that kind of "forced" is difficult to prove.) So there's ambiguity when it comes to the usage of that word.
Anyway, sometimes you can use logical relations to prove things. Like if two queens vs. king has been shown (every possible variation has been fully played out) to be winning, you might with some time be able to infer that two queens + one pawn (let's make it an "a" pawn) vs. king is also winning. You can use the logic of what you already know, that two queens are winning. One pawn theoretically might change things, but given the rules of chess, there is only so much that one pawn can do, only so many situations that a pawn could change. I think going over these in some systematic way will allow you to deduce that two queens + one pawn is winning, without actually having to show every single variation. At least, it's conceivable to me that such a proof is possible.
It's presumptuous sometimes to assume whether a position is won, lost, or drawn. But it's also presumptuous to assume that the only way to know the truth about something is to literally explore every single possibility. That's making a big assumption about how we acquire knowledge, which is a pretty difficult subject, really. A lot of things we know are derived.
The contributions to this discussion on patzer moves has reached almost 1000 views. Thanks for all of the great discussion thus far!
tonydal - You really got me thinking after you said the following:
"The key thing to note here is: nothing favors these sorts of moves more than the attitude that dismisses them outright (as being unworthy of consideration). If you take that approach, that is often your first (big) step toward losing."
What you said inspired Example 3.
I've dismissed this defense by Black without giving it any thought - mostly out of arrogance on my part.
In this puzzle, you must make four correct moves. I went deep with Rybka on this one. For the stronger players, I believe you will find the moves to be quite human and natural. For the beginner, the first move is extremely important to get right, otherwise Black can surprisingly have a very good game. I did not choose wisely in my blitz match to this very strange sequence of pawn moves by Black and went on to lose the game. I hope you find this puzzle instructive. Last, Rybka only gives White the smallest of advantages after all the moves in the puzzle are played!