It's not really about idiocy (or lack thereof). Blunders like that are often a pattern recognition / board vision thing. Being careful helps, as do a variety of simple questions like "My opponent just moved a piece to an unprotected square. Can it be attacked?" But ultimately, you'd prefer moves like this to be handled by subverbal processes.

The draw offer was perfect. From a psychological standpoint, with you knowing that you messed up two chances to win and that all you needed was a draw anyway, I wouldn't suggest anything different. If you got to that position more cleanly, you could press a bit more. You'd be a tempo ahead, with practical chances to win, since people at this level blunder a lot. His king is slightly more confined, too. The experience would be helpful, too.

Thanks. I almost offered him a draw earlier b/c I had a feeling I would be outclassed and lose and beat myself up. Then I played on and thought I had an edge and drawing was weak (this guy already beat me anyway and leads our group). Then once he moved that pawn and discovered check I was like "Oh man I would KILL for a draw now." lol.

The two most important things in queen endgames are passed pawns and king safety. King safety is more important for the side with extra pawns, because perpetual check is a major part of the defender's arsenal. This knowledge, plus perfect understanding of Q vs P, plus a tiny smattering of QP vs Q, is all you need to play queen endings at the 2000 level.

With that in mind, Ke7 is the biggest thematic mistake. Of course the computer will say you are still totally winning, but that just shows why computers are not the best analysis partners. Your king would be extremely safe on g7, look at your amazing dark square control. White can't make a passed e-pawn, I think you were seeing ghosts there. I think you would win that position 90% of the time against a like-rated player. Once he starts giving a few checks all bets are off and he could probably even have won in the game.

Also surprised Rei didn't mention your choice of opening where you "gambit" a pawn with Bc5. If you always play 1...e5 you will get this position one time in 4 as black so you should be familiar with some ideas. Bc5 is probably an ok sideline, though strong players eschew it, but qf6 seems bad - white gets the pawn and the compensation. I recommend just playing main lines with 3...a6.

Aside from this your analysis is good - Qxe3+ is indeed appalling and Bxc2 is very nicely calculated.

I saw Bc5 but was hoping a particular Malkovich player (or anyone but me) would comment on it.

I thought we had a discussion on this Bishop thing already and I was told that if they haven't brought that knight out to pull my bishop only to the 5th rank? That's why I do that now. If they pull knight out I put it on 6th rank, if no, 5th. No real reason other than I interpreted advice itt that way.

I had no idea 3.a6 was standard. Never played it. Still in the base "develop your minor pieces" mode of opening knowledge. It seems like d6 is the only way to save a pawn once he exchanges on c6. I have never been able to resolve the fact that doing this precludes my DSB from developing. And I don't really ever fianchetto. I wonder if I need to learn the logistics of fianchettoing. I've never been too turned on by it b/c it seems like everyone who fianchettos against me ends up losing. I assume because it looks cool, but at my level, they don't know what they're doing with it.

Qf6 is not fundamental and I realize it. Not gonna lie I was hoping I could get a cheap mate on the next move.

I have a hard time resolving spots where I have multiple options and option A might be "well, if I play this, and he doesn't notice what it leads to, it's good for me" and option B might be the more fundamental option but with lower reward (if he errors).

This isn't a great example because a drunk monkey will notice it but I'm talking about more complex positions in general. I usually just end up saying "would I notice?" since I tend to play opponents of similar strength. But that's silly. An opponent of equal strength will fail to notice a creative tactic you employ once in a while. It's just going to happen. Just as I will fail to notice what my similarly-rated opponent is doing once in a while. I mean the last game I played I was against the wall and my opponent failed to notice I had a forced mate if he made 1 particular move and he made that 1 particular move.

3.-a6 4. Bxc6 dxc6 5. Nxe5 Qd4 and Black recovers the pawn with a good position. In general, White isn't threatening the e-pawn tactically until his own e-pawn is defended.

The mainline is 4. Ba4 Nf6 5. 0-0 Be7 6. Re1 and now the pawn is threatened, so 6.-b5. Note that 5.-Nxe4 in place of 5.-Be7 doesn't win a pawn. Gaining the option to interfere with the bishop-knight opposition by playing b7-b5 is one of the reasons Black begins with 3.-a6.

I agree with you on not wanting to get your bishop stuck behind the pawns on e7. It's a concession Black normally makes in this line, but you can place it more actively after 3...a6 4. Ba4 Nf6 5. 0-0 b5 6. Bb3 Bc5, when you may get two very strong bishops on b7 and c5 pointing at his king. This is actually the line I prefer to play. There are some disadvantages as well as advantages - d2-d4 may eventually win a tempo, and it won't be available for the defence of the king.

Thanks guys, I'll try to burn those into my memory. If I play a6 then he exchanges on c6, my Queenside pawn structure is a bit wonky (for my taste as a noob). Does this mean I should for sure castle Kingside if this happens? Or is castling Queenside still on the table?

Playing option A is called 'hope chess' and it's bad for your game. You should always assume your opponent plays perfectly against you.

How do you (a non-perfect player) decide on a move to make if you expect your opponent to respond perfectly? That's the part I don't understand. You just pick the best move and then wait for them to mess up?

Also, what about when you've blundered and lost a minor piece (or worse!) and been left in a bad position? Do you keep making fundamental moves and limit damage or do you go for a low percentage play that gets you a win if it works out? How do you decide if you are playing an equally strong opponent?

With this game being a good example of what I'm talking about. I made a rather large (2 point) mistake early as far as I can tell. But then I went for a low percentage play that ended up working, because he failed to recognize what I was doing.

I could have traded my Queen and a pawn for a bishop and rook and played on, which (I think) would be the fundamental way to proceed, but I hated that idea and it felt like losing slowly instead of losing (or winning) quickly. Also, I don't always know what is the fundamental next play so it's guesswork at best.

The opening in this game is a new one for me. My 2nd and 3rd moves were "by the book" afaict, but then I played on by myself so it wasn't a very familiar position. I didn't know how to respond to his 1st move so that's why I consulted "the book". Also, is this cheating? I don't know how to reconcile what is cheating and what is considered "general research". Like when he plays Nf6 I was like "I wonder what the response is here", but then that feels like cheating since I'm using help specifically for this 1 game.

It's fine to castle queenside here. Your pawn structure on the queenside is a little bad, but your king won't be especially weak - you actually have an extra pawn to shield your king assuming you recapture with the d-pawn.

Slightly more advanced advice: The pawn structure becomes relevant if you exchange your e-pawn for his d-pawn (likely) and exchange all of the pieces (not so likely) when the pawn ending may be winning for him because only he can create a passed pawn on the kingside. In return for the pawn structure problem, you have the bishop pair which gives you some immediate activity.

pretty much. Often the best moves are ones where you take the initiative and give your opponent problems to solve.

if you are losing it's definitely worth thinking about playing for tricks. I think Qg2 is a pretty good try. That said your position is not hopeless after giving the Q with Bxc4 or giving a couple of pawns with Nd5 (did you look at this?) so if I was playing a 1400 here I would lean towards one of those moves. afaik chess.com allows you to consult opening books and databases at any time in correspondence games. You shouldn't need a book to come up with some good candidate moves after 1. d4 Nf6 though.

Yeah, you just do the best you can. It's a bad habit you don't want to pick up to start playing a move that worsens your position unless your opponent responds poorly. You work out to the best of your ability what the best moves are for you and your opponent and assume that's how the game will go. There's really no way to do better than that.

Now, when you're losing, then it's a bit different. You're right in that playing on to lose slowly is torture and tends to give you no chances, so playing for tricks and traps is certainly what I'd recommend. The caveat, though, is to not give up too soon. Like if you lose a pawn, you probably should lose the game, so you might be tempted to say, "screw it, let me make unsound sacrifices and see what comes of it." But if that doesn't work, it certainly throws the game away. However, being down a pawn isn't a sure loss. There are many endings down a pawn that are drawn (some R+2P v R are drawn, and with opposite colored bishops being up even 3 pawns isn't a sure win). Also, if you keep playing good moves, your opponent may make a mistake and allow you to recover, something that wouldn't happen if you throw away all your pieces.

Typically, if I find myself in a somewhat losing position, I try to do a few things:
1) Get a material imbalance. If both sides are even in pieces but you're down a pawn or two, there's not much for you to do. But if there's some imbalance, like having a bishop vs a rook, or having some positional trump like a passed pawn, then you have some advantage to play off of and push to try to win with. Don't get me wrong, you are in a lost position so the advantage isn't enough to win, but at least you have something that you can build off of.

2) Trade pawns. If the pieces go and you're left with just pawns (and you have less than your opponent), you're probably lost. But if all the pawns go, then you can be down pieces and still have a draw. For instance, R v N, R v B, RB v R, and RN v R are all theoretically drawn (see here for more info: http://en.wikipedia.org/wiki/Pawnless_chess_endgame ).

3) Try to keep my queen. One other way to get a draw from a lost position is to put your opponent in perpetual check. Queens tend to be good at doing this, so I like to keep my queen around even if I lose my other pieces.

Nope, definitely allowed to use opening books and databases on chess.com and I believe all other correspondence sites. You can't use a computer engine or ask another person, but these are allowed. Where it gets tricky is when you are researching an opening using an engine and all (like if you are reviewing a prior game or preparing for an upcoming live tournament or ...) and then you reach that opening is an ongoing game -- do you have to stop researching with the computer at that point? I guess so, but you can at least use what you've already worked out with the computer.

There has to be another solution to this. Like, hypothetically, if I was playing a correspondence game against a strong player, and they had a live tourney coming up . . . expecting them to compromise their prep for a serious event that matters, for a game with me that doesn't even matter, would strike me as insane. I'd feel like a total ******* if I wanted them to do that. And I certainly wouldn't want them to resign / offer a draw / whatever, either.

OK good. I kind of assumed my opponents also used "books" for openings from time to time, and I never cared at the thought of that. Once we are a handful of moves in (at my level) there really isn't a book anymore so it doesn't matter.

I see what you're saying about unrelated engine research coinciding with an actual game. I feel like you would have to make your next move before researching further. Not entirely practical as corr games take several days so it's pretty impractical to stop researching that position for days. Maybe make your move, continue researching while opponent is on the clock, then when it's your turn again don't consult the engine before making your next move (though of course you've gained knowledge in the meantime). This seems reasonable. I mean it's pretty easy to know if you are "cheating" or not. You'd have to hold yourself to a pretty high ethical standard to take this level of precaution for an amateur online game that in a billion years won't mean anything.

Oh, this reminds me. I suppose I "cheated" during my Q+7p v. Q+3p corr game. I was not too familiar with Q+P endgames so I searched YT and watched a bit of a Seirawan video (that guy is great - Rei, isn't he the author of the book you told me to order?) where he talks about Q+P endgame concepts. Don't think it helped me at all, though. I ended up watching a couple games between him and Aronian that were really fun. They were just talking about their plans during the game and laughing at eachother.

Yeah he is the author.

What you described is so far away from cheating that I wouldn't even call it cheating in quotes. You can definitely look for strategy advice that is relevant to the positions you are currently playing. The big exception, for obvious reasons, would be by asking a question in a subforum where many people are watching your games. You can consult endgame manuals, but not endgame tablebases, on chess.com.

Actually, here are the basic rules.

I haven't posted a game in a while so here's kind of a fun one. I know I said I'd quit 10 minute but I still do play it. I've been watching so many IM Bartholomew bullet videos I swear I am seeing the chess board a lot more clearly now and quickly. Many of my wins are coming with several minutes left instead of under 1. Looks like I've won 6 in a row.

His 3...Nfd7 was weird. I guess he hates trading a knight for a bishop but this square just clogs the development of 2 of his pieces.

8 Be2 I wasn't sure of but it defends g4 and h5 and I was taking too long to decide what to do here so I just played it.

9 h4 is sort of newish for me but I liked the idea of it. In hindsight it looks kinda sucky.

13 Qd3 was an attempt at maybe a Queen lift to exploit his bishop pin eventually.

14 Rf3. Capturing the pawn seems not great so instead I backed my Rook up because g3 is a much better square for it than g4. Then I immediately realized I blundered a piece with my fancy play syndrome. I'm usually pretty good about avoiding pawn forks, too. 2 pawns for a minor piece, along with pretty good control of the center of the board seems not terrible, though.

I think his 19...h5 was a strong move. Especially for this level. He knows if he captures with bishop then he might be in trouble. Though in post-mortem it looks like he could have captured my free bishop b/c after I take his pawn he can get his Knight to h5, which defends g7, isn't hanging, and also discovers control of g5 via his Queen. So, oops.

25. Nf3 attacks his defender.

After move 27, it wasn't immediately clear to me how to proceed, until I looked down and saw my other rook, who offered a check!

Not sure if I closed correctly, and I think he made sub-optimal moves for sure in the end, but it was a fun mate. He told me "good move" after Rxe8+. Not sure what he was expecting there. I had about 4 minutes of clock left to his 6:20. It's amazing how much worse this game looks in post game analysis, oh well.

Congrats on the upswing!

Yes, this is a pretty bad move by him. It wastes a tempo running from an exchange that likely favors him and as you noted, it really clogs up being able to develop his queenside pieces.

g4 and h5 aren't really loose squares although since I see you later shove pawns down his throat it helps support that. However, the b1-h7 diagonal looks suuuuper juicy since he conveniently moved his pawns out of the way so Bd3 looks pretty great long-term.

Take a look at black's pieces before this move. He has a "box of junk" on the queenside but his kingside is pretty developed and well defended. While he's made weakening moves by pushing his pawns, he has a nice defensive bishop. Since his queenside is so crappy it does help since backup may not be arriving very quickly but pushing your pawns without a plan may be premature.

I might prefer f4 tbh followed up by h3-g4 potentially depending on what black does. These sorts of kingside pawn roller positions are not something I'm very experienced in though.

11 bxf6 — What was the purpose of this move? You really don't need to make this move without a purpose since even if he plays Nbd7 you can just take it then.

Unless you have some specific tactis I like the idea of continuing to develop pieces. E.g. planning how you want to get your Knight out and/or castling queenside to bring the other rook over into the fight. You've battered up your opponents kingside but you need more pieces in the fight to attack properly unless you have some specific pressure you can bring to bear immediately. Qd3 isn't all that much pressure, at least that I can see.

Why? Obv Rf3 is worse but what's wrong with taking the pawn?

Also, it's a good idea to note your rook is relatively trapped and you need a plan for getting it somewhere safe (f3 is not that square ).

Even without a fork e4 is a pretty strong move for black when that pawn is supported by another one. It would be very annoying to get rid of and if you don't it will be a "thorn in your side." Getting rid of it likely means weakening moves like f3 which is a pawn I'd love to ram down black's throat instead of just using it to try and deal with the black e4 pawn.

The bishop on e2 isn't loose, though. Your knight can recapture while developing and you're still threatening Qxh6 and you've prepared Rh1 which will come in handy.

Ahhh, so that's why you weren't developing your knight, so you could spring Nf3 against a defending knight at g5 at the opportune moment!



That was a good find b/c it wins a piece. Although in the future if you can see Qh7+ as preparation for a move like that it's even better. Otherwise Qh7+ may feel powerful but may not really do anything. It's likely to do something so perhaps at your level should be played anyway so moves such as Rf1+ present themselves.

Lol. I think you closed pretty precisely. You won a bishop, then a rook and then forced mate after he didn't bother making a helpful defensive move.

Not every move once you're winning can be checkmate immediately .

He's about to discover an attack on my hanging rook, so I tried to figure out where he was going to move his Knight. But I'm super slow so after putting some thought into it I just said F it and took the knight to clean things up a bit and not waste much more time.






I fruited out. I know you're knew here but I'm super slow and miss a lot of stuff. I was worried about his recapture and immediate attack of my Rook. I think during the game I saw it playing out as me moving my rook, then him taking my knight and messing up my pawn structure but my Queen is there to recapture. I just miss stuff and get nervous when my opponents are making moves in 3 seconds (albeit they lose about half the time anyway).




Great point, and something I need to get better/quicker at identifying. You are right that it will suck for me to get his pawn there, so now it makes it that much more obvious how bad not capturing was.


Wow, I didn't even notice that in the game or post analysis. This is what I'm talking about. I think my brain doesn't quite owrk ynaorem.



Yep. I mean I knew I had a mobile backup rook now, but when I mounted my attack I assumed he might go down the h-file, which is why I'd retreated to 24.Rh2 instead of Rh1 (to defend that pawn but also keep my first rank open for my other rook to swing down if need be). But envisioning the check on the f-file is something I hadn't thought of yet until the time came.

Was this a bullet game or a 10 minute game? You mention "moving in 3 seconds". In a 10 minute game you should never feel like you need to move that quickly until near the end.

I may be new here but that is no reason to not challenge you to think through the "whys" of what you're doing. Everyone can improve by being better able to think through and explain "why" they are doing certain things. Not just beginner and/or weak players have this issue. There are plenty of "strong" players who cannot explain wtf they are doing and they will never get better (imo) b/c of it.

Watching those videos is a good supplementary training method. Quickly replaying dozens of GM games a day is similar and more effective, but a lot less fun.

9. h4 is good. Opening up the h-file is a typical strategic motif in positions like this. It helps that 9.-g4 loses a pawn.

More effective in terms of being exposed to more patterns faster? More specifically, what is the main benefit from quickly playing through unannotated GM games?

My (possibly outmoded) assumption is that playing through annotated GM games carefully is the way to go in terms of studying GM games.

I would think the strongest argument for why playing rapidly through a bunch of GM games *might* be better than watching a bunch of an IM's bullet games is that the GM games are objectively higher quality (more of the moves are "good" moves), so perhaps the patterns you're being exposed to in the former case are more likely to be beneficial and useful in the long run?

I also don't know for sure if that argument is fully valid, and if the GM games actually are more effective as a training method. One could also argue that the more entertaining option (the bullet videos) will lead to better engagement, and thus more efficient retention of the patterns.

Pattern absorption and a keener sense of what looks normal. Pretty useful ime. I thought this was Silmanian pseudoscience, but I think it has improved my intuition a lot.

I think there are a lots of ways to study GM games.

Yeah, there's still a lot of bullet nonsense even in strong 1-minute games. Probably better to replace "GM" and "IM" with "IM+ slow games" and "IM+ 1-minute games" respectively.

Depends on what you're training for. Slow GM games are > for general chess strength. The bullet videos are > for fast game ability, particularly if you're trying to anticipate moves.

I said this.

I also have been reviewing some super GM games (like Carlsen, Kramnik, etc.) They are fun to look at. I often see moves that look quite counterintuitive to me, then I watch the game play out and go "oh, so that's what they were doing". Or they move to a square that looks weak but I realize if they capture it weakens the other guy much moreso. Every little bit of these types of moves I see helps.

EDIT after Rei's post: Yes, I even saw IM Bartholomew hang a queen in a 1-minute game. He has terrific control of his emotions and is able to just laugh off huge blunders and move on. I've also seen bullet games on chess.com where one guy hung his queen and the other missed it. Pre-move creates some funky stuff sometimes.