Chess computers have been revered as these perfect entities that can calculate positions like lightning, they are so precise that almost no human can beat them. But is this true? A lot of people that I have seen on forums really glorify the strength of engines more than they should.
I think we are missing that bigger picture, the strength of chess computers does not reflect to be absolute, there are many that can be beaten depending on multiple factors. That is what I am going to discuss in this article, keep on reading if you are interested.
Table of Contents
The difficulty level of the computer will depend on its strength
In order to answer this question there is an important caveat to understand, chess computers have different difficulty levels which will make their strength relative. Not all chess computers are the same, even with chess engines they will have varying strength levels which will determine how hard they are to beat.
Some computers are so easy that a decent chess player can win 90% of the time even without thinking, if these kinds of computers were human players they would be considered as beginners. These are your regular chess training bots that are not really made to win games, rather they are created to train developing players.
There are also some who will be in the mid-tier, basically those that are still beatable but will require serious consideration from the player in order to actually win. With these types of chess computers their strength is actually decent, meaning they are not too strong yet they are also not too weak.
If you have been following news about chess for some time now you would know that some chess computers are so out of this world that even the best human players will struggle against them (some are so strong that even the best super grandmasters cannot possibly take them down). The difficulty level will really differ from computer to computer.
The difficulty of beating a chess computer will depend on its given strength level, the stronger it gets the harder it is to take down. In most apps where you can see in these chess engines there are usually options to play only at the strength of your liking, this will determine how hard of a time you will get from a chess computer.
There is no one and done answer to all of this since a computer’s strength can be relative, of course I will also be talking about other things that may answer this question though.
Mid-level chess computers are beatable with decent play
Computers can have varying levels of strength, this is true, however if we’re talking about mid-level chess computers then they are definitely manageable with decent play. Even someone who has only been playing for a couple of months can perform reasonably well against a mid-level chess computer.
If you are familiar with lichess, a decent level chess computer in there should be around the strength of Stockfish 4-6, these are engines that can still be beaten even if you commit multiple mistakes. These kinds of engines are still strong for beginners but can be beaten since they are not as precise as high end chess computers.
If we are to put a rating to these “mid-level engines” they should be around 1200-1600 as chess.com standards, they are still pretty strong but higher rated human players can beat them in serious play. If you are a beginner however these engines can definitely put you in the right place.
Defeating a mid-level chess computer should be attainable for people under the rating of 2000, in order to beat a chess computer you just have to be a generally strong player. There might be a difference in the definition of decent though, a player who has been competing for years will mean differently when talking about decent play.
In this context what I mean by decent play is a performance rating of 1200-1600 by fide standards (which is slightly higher than chess.com standards). This level of strength is pretty attainable even without formal coaching.
The best chess player in the world cannot beat the best chess engine
If we are talking about decent level chess computers then strong titled players can definitely beat them, however, if we shift the perspective and compare the best computers against the best human players how do we fare? Who will prevail? Are chess engines so strong that no human can touch them?
The strongest chess computers in the world are basically untouchable, even the best super grandmasters would be lucky to get a win from the state-of-the-art chess engine. In the chess community, opinions of the strongest chess computers in the likes of Stockfish, Houdini, Komodo, etc. are basically treated as an absolute evaluator of one’s position.
This is for a good reason, many grandmasters are shocked at how precise the engine’s evaluation is in positions where the human cannot tell that much difference, the experts agree that computers are stronger than any human at the moment. A chess engine’s opinion will almost always be more respected than a grandmasters’.
Magnus Carlsen, who is arguably the best chess player of all time, even concedes that chess computers are far more powerful at the moment and competing against them is a waste of time. The best chess (human) player in the world is definitely far behind than the strongest chess computer.
Chess computers are so respected that using them in a competitive game is considered cheating, something that would take away the chances of the opposing player. If one of the competitors uses a strong chess engine throughout the tournament then he/she is guaranteed to win at least 80% of the time (if not more), that is how broken it is.
Beginners will have a hard time even against weak chess computers
Now let’s talk about weak chess computers. People generally have an easy time dealing with computers at this level since their algorithms still apply subpar technology from ages ago. Computers like these are usually not made to win games but rather to train weaker players.
Most people who have been playing chess for quite some time are able to defeat chess engines at this level, there aren’t any advanced maneuvers that one would need to do when trying to win games against these opponents. Only beginners are usually one who struggle against these opponents.
If you are playing bots on lichess this would constitute stockfish 1-3, if you have ever played against such a computer then you know it is quite easy to defeat. These engines will normally hang pieces and pawns with no compromise whatsoever, as long as you don’t give material for nothing in return it should lead to a win.
This might sound alien to some people but computers of the old age don’t really have a good reputation against human players, in fact they are constantly being stomped on by even mid tier competitors. This is not the case now, but these weaker chess computers are similar to the basic version from ages ago.
If you are a beginner the idea of defeating a chess computer might be a difficult task, even the weakest one might put up quite a challenge (since you yourself are also vulnerable to hanging pieces). However I believe that this is not a predicament and more of a challenge, defeating a weak chess computer is something that you need to achieve in order to get to the next level.
Most chess computers are beatable if you are good enough
Earlier in this article I have talked about the glaring difference between a human and a chess computer (in strength), although it may be true that any human cannot beat the strongest chess engine the majority of the players can actually come out on top. Not all chess engines are deemed the “strongest”, most can be beaten by a decent play.
It is not that far from the past when chess engines were considered the laughingstock of the community, almost no one believed that an algorithm could be created that would allow computers to take over human players. Chess after all is complicated, no one believed that a machine could make such advanced calculations.
What I am getting at is the complexity of creating a strong chess engine, it needs state-of-the-art technology that would cost a lot of money and manpower to be functional. Not a lot of organizations have access to such resources so the majority of available chess computers are beatable by modern standards.
There are many people who worship the strengths of chess engines and believe that every single one of them could beat any human competitor in a heartbeat, but this is not true. Only the best of the best engines are actually capable of doing that, the majority are beatable by humans.
Some people think that they need to cheat or something in order to truly beat a chess engine, this is not true in any way since you can beat most computers by just being a strong player. The strongest chess computers constitute a small fraction of the entire chess engine population, not all of them are going to be that way.
It is not true that you have to cheat in order to beat a decent level chess computer, by being good enough at the game it can lead you to some wins against these famed machines. What I mean by cheating is of course doing takebacks, this is something that should only be done in a practice setting and not in serious games.
Players should use different style against different chess computers
Now this is something that might or might not work depending on how you implement it, I’m talking about anti engine moves/strategies. These aren’t silver bullets or anything and would probably be considered a myth by some individuals, though I have personally witnessed that this can actually work.
There are many anti-engine strategies that you can implement to take advantage of a chess computer’s algorithm, though it will still be difficult to truly dominate a strong chess computer, it will significantly increase your chances. These have been thought about due to the existence of shadow moves that cannot seem to be identified by engines.
One most popular anti-engine strategy is to use tactical shots when playing against weaker/mid-tier chess computers and implementing strategic ones (moves that desire long-term advantages) when playing against stronger chess computers. This is to take advantage of an engine’s embedded algorithm.
Weaker chess engines tend to calculate surface-level of material gains rather than tactical ones, this means that they cannot comprehend positional compensation that is just giving material away. However if you have been playing chess for quite some time now, you know that there are occasions where sacrifices have to be made to gain an advantage.
This is why it is recommended to play tactically (short term plans, are positions that appear on puzzles) against weaker chess computers, whose algorithms usually cannot comprehend apparent material sacrifices. You can even play gambits that give material away (temporarily) in order to gain some positional advantages.
On the other hand if you’re playing against a strong chess computer it is recommended to play strategically (positional, long-term) since these are the kind of chess computers that can outplay a human tactically. Some of these run on brute force calculation that can basically calculate thousands of lines from a given move, it is almost impossible to out-calculate these engines.
One of the best examples of this is Rybka vs. Nakamura, Hikaru won by closing the position and taking a very renowned engine to a closed endgame. In the end Nakamura won since Rybka had a harder time thriving in these kinds of positions.
Final thoughts
If we are going to pit the strongest chess engine against the strongest human, the engine will definitely prevail, they have been crushing us since the near 2000’s. However not all chess computers can fall into this category, most are quite beatable by modern standards.
If you are going to play against an engine like Stockfish 9 then the chances of you winning will of course be incredibly slim, however, if you take some unknown computer that not a lot of people use then you have a higher probability of victory. That is all for this article, thank you for reading.
