Why Chess Strategy Changes Every Year 

Today I learned that chess has an evolving meta despite no balance changes in 100s of years

The finger breaking technique seems most promising.

"If you see a good move, find a better one" is the best Chess advice I have ever heard. That guy had things figured out.

It’s basically a meta that comes around every year or so when the top engines like Stockfish, Leela, or AlphaZero find something new. A new meta tactic in todays game is a pawn break of h4, h5, a4, a5, etc. A dominant opening meta on todays game is the Catalan, but e4 e5 is seen regularly as one of the most used openings today still.

what a beautifully crafted game

That magic line was completely uncalled for. I don’t need to go outside because I see all of the outdoors on the land cards in my deck

0:49 Minor correction - Mathematically speaking, there *is* a perfect way to play chess, we just don't know what it is. As you mention, chess is theoretically solvable, but it is *practically* impossible to solve due to the incomprehensible size of the problem.

At 3:26 I'd rather capture the queen on f6, but perhaps that's just me.

By breaking your opponent’s fingers before or during the game you put them at a severe disadvantage because they are in so much pain they lose focus on the match. It’s a personal favourite of mine 😊

I could be wrong but chess's status as unsolved doesn't imply there is no right way to play, it just means we don't know if there exists a right way to play.

Yeah, Magic The Gathering cannot be solved so as much as I am a fan of it I cannot argue with just leaving being the only truly reliable strategy

Actually, the King's Gambit is perfectly fine for average players. Even computers don't rate it that badly. Ofc they definitely rate it worse for white but for most chess players, it's still a marginal difference and because it's not as popular, you're more likely to know it more in-depth than your opponent. There are much worse gambits that people regularly play. And the reason it's bad isn't necessarily because you give up the pawn. The Queen's Gambit does exactly the same thing just on the other side but is a well-respected opening. One of the issues with the King's Gambit is that it opens up your king too much. And ofc you also don't quite get good enough compensation for the pawn. But in general, it's not too rare that giving a pawn for tempo/initiative can be the best move.

Just realized I am still stuck in 1850's based on my strategy.The jokes are on fleek as usual.

Zermelo's theorem shows that there must be a perfect strategy for playing chess. It's a perfect information game without randomness. Recursively applying Von Neumann's algorithm to the starting position would result in mathematically perfect moves. It would also take trillions of years, so real chess computers use imperfect heuristics to find strong but not quite perfect moves in seconds.

Kinda sounds like this video was made by someone who doesn't know much about chess, but has read a wikipedia entry about its history.

you can though objectively say chess players have gotten better over time by looking at accuracy of games with engines.

As a chess player my whole life I love a video making the game sound cool. The text in the thumbnail is an outright lie though, 1. e4 e5 is still an incredibly popular way to start a game at all levels. @Half As Interesting please can you correct this by putting the king's gambit position on the thumbnail instead? The text would still be hyperbole but definitely true at the highest levels at least. I wouldn't want people to think e4 e5 is a bad opening.

3:17 Mistake! People back then considered king's gambit to be the best opening, and countless analysis and top level tournament when players could only play king's gambit, nothing else proved them wrong

Before: "if you see a good move, find a better one" Now: "if you see a checkmate, find a better one" The strategies really changed

Forgot to mention the first strategic breakthrough: Philidor's discovery of the importance of pawn structures, in the 1700s

To think that after centuries of mediocre chess playing we are finally returning to time tested methods like physically harming your opponent

Chess openings are just "you outsmarted me, but I outsmartsd your outsmarting!"

I definitely wouldn't say chess "strategy" changes every year. You basically described changes that occurred over decades or a century, which, if my math is right, is not equivalent to "every year." Openings fall in and out of favor regularly, but these are not changed "strategies" as they are typically openings that have existed for years or decades, including 1. e4 e5, and deviating from existing theory is so rare that it is a called a "novelty." According to a database of master games, e4 is still the most common first move. e5 is the second most common response. The more common response is c5, the Sicilian, which is from four centuries ago...

One of the coolest things I've learned about chess players is that they don't necessarily have better memories or higher intelligence. A study was conducted on chess players of different levels of skill where they would set up a chess board and arrange the pieces into different patterns. They asked the players to memorize as many of the piece locations as they could in a few seconds. As expected, the players with more skill and experience memorized more piece locations. But when they ran the experiment again, placing the pieces in ways that didn't make sense in terms of how the game is traditionally played, all the players memorized the same amount. The pattern recognition of the different possible moves allowed more experienced players to essentially chunk the information more concisely, helping them get more pieces. Take away the patterns and their memorization performance was equal.

it's called a META and basicly all games develop them over time why they change, well, counters mostly. there is a good strategy, eventually you find a way to beat it, so the strategy changes

I was really hoping this would cover the shift from Classical to Hypermodern openings, and the highlight the influence of AI like LeelaZero on the use of flank pawns. If anyone is curious you can look that stuff up.

The kings gambit wasn't all a bad opening, it was just super risky. But in return, it allowed you to play super aggressively, which was the advantage it gave you. The reason people stopped playing it wasn't because it was necessarily bad, but because playing risky in pro level tournaments just isn't a good idea. At least, playing as risky as kings gambit is

HAI needs to realize that what made them successful was being an informative channel which has some jokes. Nowadays, they are a comedy channel with only a little bit of information (and this feels annoying).

There's so much wrong with this video. 1. You covered the evolution of chess strategy in over a century, not "every year" like the title suggests. 2. The Open Game is still one of the main openings of high level play. 3. Just because chess is a unsolved game, it doesn't mean it doesn't have perfect play. Chess is, in fact, theoretically solvable, but the computation time to do so is so big that is practically impossible, but that doesn't change the fact that prefect play does exist, we just don't know what it is or what the optimal result is. 4. The Magic the Gathering bit is completely unnecessary, and I don't even play Magic.

i use this strategy known as "the swipe" which basically means you use your hand to swipe their team when youre losing, its a very promising strategy which gives me a 100% win rate

This is partially true. 1. Opening with d4 was popular for quite a while - Tarrasch was so obsessed with his version of the QGD that he marked all other lines as dubious. The Dutch was also popular. 2. The e4 e5 and KG systems were popular but so were more stable lines like French, Ruy, Italian, Philidor, Sicilian. In fact, the Evergreen Game is an Evan's Gambit. 3. While hypermodern theory was developed in the 1930s, its roots lie in Indian chess as well as the Sicilian, both centuries older. It was and still is at least equally popular to classical styles. 4. There's a huge jump between Romantic chess - which has little to do with masculinity as much as the style them - and the Fischer-Spassky game. You jumped over Morphy, Euwe, Capablanca, Botvinnik, Nimzowitch - huge theoretical leaps in strategy and endgame. Plus, was Tal Romantic? There is also a gap between Fischer-Spassky and computer-influenced chess, dominated by Karpov, Kasparov, Korchnoi, Kramnik, Anand. It was Kasparov who was first beaten by a machine; he also innovated opening prep.

3:26 casually blundering the queen and thus the win

That attack on magic players was unexpected but appreciated. I go out on my euc thank you very much!

"Very precisely arranged piles of sand" is easily my favourite description of computer processors

So there are few inaccuracies( that's a chess pun btw, an inaccuracy is when move is a slight mistake). E4 E5 is the king pawn opening. It's still played a lot in lower level play. The Ruy Lopez or Spanish game has always been the most popular opening and even today the Berlin defense is played at the highest level, with it being a particular favorite of Magnus Carlson. The Queens pawn opening isn't a slight difference it's a completely different class of openings. The kings gambit is never played in standard play, but is often used in rapid play, particularly Nakamura uses it a lot to throw off top players all the time. You mentioned the 1930s and romance period, but forgot to mention the hyper-modern period of the 1930s which led to more asymmetric openings, followed by the soviet period followed by the modern period. (There is the classical period, romance period (which are sometimes considered the same period really), hyper modern, Soviet, and modern period).

The thumbnail is unfortunately hugely misleading, the King's Pawn Opening is still the most popular opening in chess by far. Oh, and I'm pretty sure most chess players like me just play whatever opening we want at the start of the game. It's boring if you play the same move every time.

I've heard that before the engine days there were people who got titled off knowing just one opening, which they'd play at almost every tournament. Nowadays you can just look up your opponents ahead of time and have a way easier prep, or harder if you look at it the other way I guess.

It is terrifying how wrong the thumbnail of this video is. In the most recent Candidates Tournament (the tournament that determines who faces the world champion), 48% of the games began with the moves that the thumbnail claims aren't played anymore: King's pawn moved two squares forward played by both sides (or e4, e5). And in the last World Championship (which was in 2021), 8 of the 11 games in the match were King's pawn forward two squares by both sides (e4, e5). So, this video's thumbnail is incredibly wrong and misleading. As for the video itself... it generalizes in ways that are a little more imprecise than what a generalization needs to be.

0:52 Actually should be "no known right way". We'll find the right way after we make better quantum computers ぬ

A half as interesting chess video? Hell yea

The guy in the painting isn't Ruy Lopez. That's king Philip II of Spain. Ruy Lopez is the priest playing chess on the far left of the painting. He isn't even in the shot. Why anyone would think that the dude sitting on a fancy chair next to the queen is Ruy Lopez and neither of the two guys playing chess is beyond me.

so what youre saying is there's updates to chess every year?

Jet Lag Challenge: Get drunk on a local liquor and then attempt to play chess.

the "bad" era of romantic chess is by far the most fun

A thing that is not at ALL covered in this video is that openings in chess are actually just knowledge and preparation before the game starts. you play a certain opening that you know the strenght and weakness of and you play different openings to not fall under your opponent's preparation against you. if you're a good positionnal and strategic player you'll maybe play a closed game and go for a close opening, if you are better leading the attack you'll might want to play a more risky and open game. Chess strategy change because if it didn't, people would just memorize 40 moves of the same opening.

You can win a game of chess by simultaneously playing a game of Magick: The Gathering: Since MTG is turing complete, all you need to do is set up your game in a way that it emulates Stockfish (or any other chess engine of your choosing) and feed it your opponents moves. That way you are virtually guaranteed to win - because your opponent will leave the game out of sheer boredom while you are shuffling around your cards!

You're just plain wrong, the kings gambit is a very solid opening that only doesn't hold up in the supercomputer level

2:43 because a lot of openings haven't been studied as much, as they are today, players didn't know certain ideas. Modern grandmasters have learned a lot from chess engines like stockfish or Lc0, because they have more knowledge about openings, and knowledge about modern ideas (like pushing the h pawn to h4, and then to h5 in order to cause positional damage to opponent) they have an ideas about positional play, that no romantic player had

1:00 is brilliant, baiting idiots like me into commenting how both red and blue messed up therefore feeding "engagement" stats to channel in the youtube algorythm.

5:04 "Now, I know this is the part of the video where I'm supposed to sell you something, but today i'm gonna tell you about something that's a hundred percent free." *Shuts door blazingly fast* The apocalypse is coming!

3:26 The queen's hanging lol

I always win chess thanks to my RT-2PM2 Topol-M 2-staged intercontinental ballistic missile.

This is prolly an interesting video but at every level 1. E4 E5 is hella common

the middle board at 1:06 is known as the "game of the century" played in 1956

I'm referring to computers as "very precisely arranged piles of sand" from now on

Another thing that makes an opening "popular" or "unpopular", is how it's used by the grandmasters in big games. Let me explain with an example I recently read: The Berlin Defense of the Spanish (Ruy López) Game was rarely used in high level chess, because there were better answers by Blacks: It was seen as "draw at best" for Blacks. So... no love for the Berlin Defense. That was until Vladimir Kramnik rode the Berlin Defense to defeat the man himself, Garry Kasparov, in 2000, for the Classical World Championship. He basically used it because it was a powerful drawing weapon (drawing all 4 games he employed it). Suddenly, the Berlin Defense received a lot of attention everywhere, and started to be used more and more in high level games.

0:51 chess is solvable. So there is mathematically speaking exactly a right way to play chess.

2:31 it's not very good at the grandmaster level, but it is very strong at the intermediate begginer and advanced level. Levy Rozman who's a International master in a lot of his clips says that king's gambit is his favorite opening. Also statement used in the thumbnail is misleading, almost every player who's a beginner plays 1e4 e5, also a lot of gm play 1e4 e5, Italian game and Spanish game are very common at the top level

e4 e5 is still the best opening by far, because it leads to the unstoppable Ke2 (often matched by an equally brilliant Ke7)

Love how white totally hangs that queen in a winning end game at 3:26 lmao

Hey that's some major misinformation about the king's gambit. The opening is completely sound even by computer standards. It just requires grandmaster level play.

Well almost every game in the last world championship opened with e4 e5 and Spanish game so 💀

Wait until chess players hear of the steam game “4D chess with multidimensional time travel”

1.e4 is still one of the best opening choices. The Spanish opening/Ruy Lopez is very popular among the top players and unfortunately, often leads to a draw, especially if they chose the Berlin variant of the Spanish.

the ICBM variation of the Tenison gambit is the best opening

the kings pawn opening is literally still the most popular opening tho

I'm an avid chess player and 1.e4, e5, is one of the most common openings.

Kind of underselling Ruy Lopez considering the Ruy Lopez opening is still played today in top level tournaments...

Nice try, but everyone knows the castle move is the best strategy.

3:26 Epic positional queen sac

the right way to play chess is now the bongcloud attack

The queen was hanging at 3:25 Who else noticed?

Random Fun Fact: Cats can jump upto 5 times than their heights.

I mean the thumbnail is very wrong. e4-e5 is still the most popular opening in chess per Lichess' player database, and the second most popular at the high level.

Lmao as somebody who plays chess somewhat regularly, I can tell that the real chess-heads are gonna come after Sam for this one 😂 I’ve never once heard of e4 e5 and d4 d5 get contrasted as the open vs closed games, and I can already tell that the oversimplification of the history of openings is going to trigger thousands of players 😂

3:26 you blundered the queen lol

1. e4 e5 is not a bad opening by any means, and stockfish (the strongest brute force chess engine) still rates it highly. It has lost popularity over the years largely because the Sicilian defence (1. e4 c5) is considered better for black by computers and chess masters alike, leading black players to prefer the Sicilian and white players to play other openings to avoid the Sicilian. Furthermore, machine learning driven engines like alphazero and leela often avoid e4 as an opening move in favour of other moves, and especially openings centred around moves like d4, c4, and Nf3 for white.

So... Kinda wondering but there's this thing in Chess called a *_Promotion_* where you can convert a *Pawn* into any _other_ Chess piece _except_ for another *King* should it reach the opposite. The Wiki discussing this move explains that this theoretically allows a player to have up to 9 Queens on the board necessitating the existence of such additional Pieces to be brought to the table. While most definitely not a problem in a tournament where such an arrangement can easily be made - How is this usually being handled when you play at home or on the go? I don't recall ever seeing more than two Queens ( one of each color ) to be present in a Chess Set 🤨

Position chess: proceeds to show images of queen hangs

chess being a narrative of you chasing the other player's pieces and trying to assassinate his king sounds a lot more fun

Chess has been solved, the bongcloud opening is by far the strongest...

One of the ways we actually beat the best computers is to lock them in a position where they want to repeat forever. They are coded to eventually try something else, which is inevitably a less optimal move, which can be punished

Except it's mathematically proven that there is a way to perfectly play chess, only no one knows what it is or what is the result of that perfect game (though most suspect it's a draw).

Ruy Lopez, famous non-contributor to the opening meta

In case you didn't know, the breaking your opponents finger actually happened when a robot was playing a young player.

Wow. That Magic Gathering joke made me laugh so hard I woke up, and scared the hell out of, my dogs...lol

Today's fact: Donkeys kill more people annually than plane crashes.

>there is mathematically speaking no right way to play chess [because it's not a solved game] Um, no, not quite. There is actually mathematically speaking the right way to play chess, because there's actually a finite amount of states the board can be in - every tile can either be empty, a pawn, a rook, a knight, a bishop, a king or a queen, of each color, for 13 possibilities, and either of the players can be on the move. This gives us 2*13^64 possible states, giving us an upper bound which we can then whittle down from there by rejecting illegal states (it can't ALL be kings for example) or states which can't actually happen in the game (there will never be a white pawn on A1), but the point is we have a specific finite number. In chess there's a rule that if the same state happens 3 times the game ends in a draw, so that means we could potentially list all possible games of chess (move by move) since they moves are switching between those finite number of states, and that too would be a finite number, organised in a giant tree where white is on the move on odd levels, black is on the move on even levels, and the edges are the moves which were made. We could then also mark all the leaves of this tree as "draw", "white won", "black won". Then we go up. If white is on the move and can get to "white won", the state is marked as "white won". If white is on the move and white can ONLY get to "black won", the state is marked as "black won". If white is on the move and cannot get to "white won" but can get to "draw", we mark this state as "draw". For black we do the same with the colors swapped. Because the tree's height is guaranteed finite, you are guaranteed to finally get to the root, the opening state, and it is guaranteed to be either "white won", "black won", or "draw" (that would be solving the game). The mathematically right way to play the game is to follow this graph - always make a move which would put you into "[my color] won" or "draw" if that's not available on the graph. That provably has to exist. We just don't know which moves do it for which state since we haven't actually constructed it.

Moving same chess pieces back and forth by both players ends in a draw as far as I know.

Then there’s the one guy who plays the bongcloud

Still I'd wager close to 80% of modern GM games begin with either a Spanish a Queens gambit declined or a Catalan.

Now imagine this exact scenario happening every month or two for nearly forty years, and that's how Warhammer works.

I still play e4 e5 though lol

The Kings Gambit was the best strategy at the time. It was (and still is) a strong, aggressive opening that takes control of the center. It's not played today (at the highest levels with long time controls, it's still played everywhere else) because counters have been developed. Those counters didn't exist back then, and even if they did, the level of play was low enough for it to still dominate. Romantic play wasn't losing for fun, it was a strategy meta that consistently beat other metas until people found ways to beat it. It's a bit like saying people selling horses in the middle ages were doing so because they were romantic for horses, and not because cars hadn't been invented yet.

Im a 1200 chess player and well thats true but for the highest if not the professional levels of the game. The lower levels (the majority of players) always use the same old openings.

I got excited for half a second when i thought the ad would be for lichess lol

I'm sorry Sam, but this is just a bad video.

3:25 Queen hanging lol

I’ll just stick to the Bongcloud Attack

Remember: Always play opposite a mirror so you can see your opponents pieces