Like <rilkefan>, I wondered why black didn't play 20...Qxb2. For that matter, I didn't understand why not 23...Rxb2. Fritzie doesn't see a crushing white response to either move. It's a free prawn.

Larry will be kicking himself for not seeing 29... Qd8? 30. Rxe6. Presumably he thought that white would have to exchange queens so didn't take a serious look at the exchange sac on e6. After this, a black advantage of about a pawn (would have been more if he had grabbed b2) transfers to a black advantage of about the same margin.

Fritzie, being a merciless soul, also doesn't care for 31...Kf8, preferring 31...Kh8.

33...Qd7? leads to a mate in 4 when other moves (eg 33...c3) lead to a white advantage of a prawn and a bit.

So what gives? I think the key to this game is this position after black has played 29... Qd8

click for larger view

I reckon black thought he was winning by miles here. He thinks he has defused white's kingside attack. It looks as if white either has to exchange queens or allow black to snaffle the f6 prawn. And either way, black is going to play d4-d3 at some point and nasty things will happen.

So black sits back smugly. He perhaps glances over at the little kid who can barely see over the edge of the board. The little 'un has done well to stay alive this long, but true class will always win in the end. Maybe give a kindly smile of encouragement. Offer to autograph his scoresheet. Shake his adoring parents' hands after the game. Point out where he went wrong. The kid will treasure the experience for the rest of his life.

But what's this, the lad is picking up the rook on e1? Doesn't he see that his queen is en prise? Rxe6?

Oops.

Then it all falls apart, just as <phony> has described. The computer may find ways for black to cling on, but it's much harder in real life when you've just thrown away a winning position with one miscalculation.

Interesting point. Some records can be verified. For example, we know that Fischer was the youngest player to become a GM (at the time), because we have comprehensive records of all the GMs.

But this record is harder to prove. How do we know that a GM hasn't lost to someone younger, say in a practice match or at queen odds whilst hugely and improbably drunk?

Okay, that may be a little harsh. Let's rethink. How about ... making it illegal for anyone under sixteen to play chess competitively with anyone aged 47 or older.

Even still I find it unlikely that someone like Judit, Susan Sofia Polgar or Karjakin doesn't already hold this record.

I'm curious, because the only definition of the "word" I can find is as the name of a website/blogging platform.