how much logic was behind tal's sacrafices

serious question

About 8

"A good sacrifice is one that is not necessarily sound but leaves your opponent dazed and confused."  -  Rudolph Spielmann

I think many of Tal's sacrifices would be found wanting when analyzed by modern chess engines but as Spielmann once said the purpose of many are to daze and confuse.  Tal loved chaotic positions.

@melvindoucet - post of the week :-)

A combination is a symptom that the opponent was careless enough to allow it.  Any combination can be forestalled far enough back. Tal was a great player and did create problems for the opponent, which is just one way among many of winning.  The danger however is what if the opponent correctly solves that problems leading to a draw or worse? 

It was incredible how he was able to defeat Botvinnik for the title, but due to his defective creative faculties (according to Botvinnik) during the time (he was ill then so that might have contributed) he lost the rematch. 

I have frequently come across positions where I could have played a Tal-like sacrifice. Usually I didn't because I couldn't calculate all the most likely lines, but later when I analyzed them after the game I often found I had either winning lines or in one case a perpetual check against a higher rated player if I had sacrificed my rook as I had considered doing (but I had shied away and lost). I believe there is a continuum of certainty. If you have a forcing line that looks promising then there is a high *likelihood* of it being the best line, or a winning line, so if you're adventurous it makes sense to play it. It's just a matter of likelihood combined with judgment of the position, and how much chance you want to take. There is a spectrum from completely uncertain to completely certain. Players like Tal took chances when they weren't 100% certain, whereas some players (probably including Petrosian) wouldn't take any chances unless they could completely analyze a position. That's part of what constitutes a player's style: the amount of risk along that spectrum to which they are willing to drop.

----------

(p. 14)
SACRIFICE

The acceptance of a sacrifice usually heats up the battle.
Experienced chessplayers think long and hard before entering
into a sacrifice, deeply calculating all possible results, knowing
well that there may be unexpected resources (in-between moves)
available to the opponent.

   When you make a sacrifice, you should try to consider and
evaluate the positions which arise after each move by each
player. An error in calculation may lead to an incorrect sacrifice,
and as a result the position may be lost instead of won.

Palatnik, Sam, and Lev Alburt. 2013. Chess Tactics for the Tournament Player. New York, NY: Chess Information & Research Center.

Nf5?

Imagine if you were a genius and could always play the best moves.

Not only could you play the best moves on command. You are also so smart that you are able to discern what the person across from you is understanding about the game.

That is Tal during his peak.

His games have been anaylized quite a bit in the computer era, I have heard it said that Kasparov has made more sacrifices that dont hold up to further inspection then tal.

Many of Tals sacrifices are good enough to draw but the opponent had to walk a tightrope to get there often falling off in the process.

But that's just the regular sound move

eg exf5 exf5 Bxf5 Rde1

not much

nice

Just analyzed the entire variation,this line's pretty deep! What game is this from?

I agree it's a subpar move but that was just my first thought..considering it's Tal and thinking it must be something tricky..

Yes, it looks like e5 will be next, driving off the knight. Black has a vulnerable spot at d7 with two of White's major pieces threatening mate there. Admittedly I didn't see that Nf3 move, but then I didn't spend much time looking at it.

Actually, Nf3 isn't that hard to spot, I don't believe. It's just a case of deflection, which is one of only about ten (?) types of tactics used in combinations (decoy, deflection, obstruction, interference, pin, skewer, double attack, X-ray attack). First you analyze the position, note potential mating spots or potential alignments that cause loss of material, then reason backwards to see if you can produce the envisioned situation by use of forcing moves. Although I haven't seen how the rest of this game played out, this may not have been the best example of intuition-based play, since it's more like a puzzle based on forcing moves.

----------

(p. 24)
Deflection

Deflection is a tactical device used to remove a guarding piece
or pawn from defending a square, line, or another piece.

Palatnik, Sam, and Lev Alburt. 2013. Chess Tactics for the Tournament Player. New York, NY: Chess Information & Research Center.

Hey, I was right! Yes, so now you see that White is even closer to his envisioned mate, or even just capture at d7, since he forced that "d" pawn to move out of the way to open up the line of attack for White along the "d" file. The only piece protecting that square now is the Black knight at f6, which will be the next major target. Two candidate pieces for removing that knight are the White knight at c3 (via Ne4 or even Nd5) and the White rook at f1 (via a later Rxf6). f7 also looks fairly vulnerable, with White's bishop potentially aligned at that spot, so the sacrifice Bxe6 becomes a consideration in the combination. There is also that potentially devastating alignment of the Black king and queen on same diagonal where Bf7+ could be played later, supported by White's rook at f1. There is also the potential of Black's queen getting pinned against the king with one of White's rooks as the intervening units are removed. Nxe5 becomes a consideration, too, since that removes yet one more pawn that is shielding the king, and after ...Qxe5, Rfe1 would drive the queen off again while simultaneously aligning the rook and king. As one book I read recently said, often you just have to take those fragments of a combination and put them into the correct order to get your solution. I still haven't worked out the finale here, but that's my reasoning, and that's probably the approach that books would recommend.

P.S.--In answer to the OP's question, you can see there is a lot of logic from Tal behind these moves, at least in this example.

i REALLY like NxP QxN Rxf6 but perhaps it is reliant on accepting the knight sac. blacks N is just floundering in the bog. Tals pieces are truly engaged.

ne4 probably better played first.

Ng3 and Nxe5 as the next 2 moves for White.

Wouldn't work on top players nowadays