Knowledge versus Skill

I wrote this post for reddit, and despite it getting little attention, I still really like the content so I will re-post it here. 

The inspiration for this post comes from these sources: 

Kostya Kavutskiy's chess.com blogpost

His video about the article  

and this Dan Heisman video  (mostly just the introduction)

Sidenote: Dan Heisman's videos lack the flash and production quality of most current streamers but he has a ton of very useful information and over 100 videos. They are very under-watched in my opinion.

Most aspects of chess improvement are either knowledge-based or skill-based

Chess knowledge would generally be things you learn from books and lectures. You might be able to pick up a lot of these things through passive learning. Skills are, as Kostya puts it, "simply your ability to actually play chess." This includes anything you do over the board that requires analytical thinking or concrete decision-making. Heisman says these take up a bigger percentage of how good you are and what it takes to improve.

Examples of Chess Knowledge:

• pattern recognition
• openings 
• book endgames (lucena position for example)
• principles, be it opening (control center and develop), middlegame (rooks like open files), or endgame (activate your king)
• general chess terms (definitions and practical applications of chess terms such as doubled pawns, minor exchange, bad bishops, initiative, etc)
• general chess wisdom (if you find a good move look for a better one)

Examples of Skills and Abilities:

• analytical ability
• visualization
• calculation
• evaluation abilities (includes some knowledge but it's also a skill)
• perseverance
• ability to learn from losses
• stamina
• determination
• focus
• handling your nerves
• time management

Heisman thinks skills and abilities are about two-thirds of "what you are" as a chess player and knowledge is about one-third. I think most will agree there is a partnership between skills and knowledge when making the thousands of considerations and decisions one might have to make during the course of a chess game. Your knowledge is correlated to your rating but so is your skill and Heisman says the correlation is more tilted towards your skills and abilities. They very often work together but it seems like his theory is that skill and ability is overall more important.

Game Report Cards

At risk of sounding arrogant, I think my knowledge is solid, or rather good enough for my level such that I can improve at a consistent rate towards my goal of a 2000 rating in certain time controls. More broadly--and with some arrogance filtered out--I feel confident in saying my knowledge outweighs my skill. I've done enough passive reading and lecture-watching to understand on a surface level a lot of chess ideas and knowledge (apart from opening theory which I do not know well). My skills to put that knowledge to practical use needs work. Imagine someone who can understand and explain the ins and outs of American Football strategies within each play. They can comfortably explain, using the right terms, all the movements of the eleven position-players and what the goal is. Now imagine the same person trying to be the offensive coordinator of that team. Without practice and skill to draw up and call the correct plays in the best situations, they'll probably fail miserably. I think I'd fall more in that category than I want to. 

As I said, my skills in applying my knowledge need work and for that I think that analysis of those skill attributes is required to get better. The following are skills of chess that I think I need to improve the most:

• Calculate fully. Don't stop calculating when I think I've gone to the end of the critical line.
• Manage my time better in the opening. In a 15+15 game for example, I'm often using 5 or 10 minutes in the first ten moves which is absurd. Try for an advantage in the middlegame without killing your clock and save time for the conversion in the late middlegame and endgame. Trust that I can tell when a move-choice is critical and when playing solid-looking and principled moves is good enough. Don't let the perfect move be the enemy of the good move.
• Develop a killer instinct and trust it. Use analytical ability more than lazy assumptions that my opponent's move was a killer. Sometimes they're garbage and it's okay to call them garbage. I think many times, they play a move I didn't consider and my first instinct is that it's winning, or it lets them squeak out a draw. That's the reactive and scared-puppy-dog-instinct taking over. The killer instinct says, "I didn't see that move and it's probably garbage. Why is it garbage?" Analyze moves as they are, not as if my opponent is a god and their moves always have merit.
• Keep Calculating in the late middlegame and endgame. Don't just stop when you have a won endgame.
• CCT Automation: Checks, Captures, Threats. I need to automate this thinking so every move I have I'm looking for checks, captures, and threats I can give my opponent as well as safety-checks for the move I want to make: making sure I'm not giving my opponent valuable CCT. As many tactics as I've done, I still don't have this thinking process automated in my games, especially in time pressure.
• Strategic Move Automation. Once I identify a position as non-critical (no tactics available), I need to have Aagard's three questions automatically appear in my mind: 1) Where are the weaknesses? 2) What is my opponent's plan? 3) What is the worse placed piece? 
• Simple Strategic Analysis after exchanges, such as after takes-takes-takes-takes, being able to picture what the board looks like and that there might be a simple strategic imbalance I or my opponent could take advantage of. Example: in a recent game I was up a pawn and I didn't consider an idea of sacrificing the pawn back and in return I'd get the semi-open a- and b-files for both my rooks that would come with tempo on opponent's queen. A lot of imbalances would have been in my favor.
• Applying nuggets of chess wisdom. Think "when facing a flank attack, counter-attack in the center" for example. I can recite tons of these nuggets or pass a fill-in-the-blank quiz, but they don't naturally come to me when I'm in the middle of an intense game. 

Using these categorizations in a game analysis setting would mean to analyze how effectively I used my "knowledge" as well as how effectively I used my "skills." A fabricated example of applying this in post-game analysis might be: 

• "At move 20, I was doing a good job of applying my knowledge of the minority attack, pushing my queenside pawns forward to create a weakness, but I did not appropriately use my tactical skills to see that my opponent had set up a battery and was about to skewer my queen on c2 to my rook on b1."
• Pros: Good use of knowledge of minority attack I had studied in master games.
• Cons: Poor use of skills in A) asking myself what my opponent's ideas are, and B) missing the simple skewer tactic.

Conclusion

Overall, I think a lot of my problems come with trust and confidence in my abilities and assuming far too much strength in the moves of my opponent. Lately, videos of IM Andras Toth's private lessons with his students has taught me to criticize this part of my chess game more. It's really criticizing my psyche which I had never thought about!

I really like the new framework this can give me as I analyze my games. As opposed to missing a backward knight move and then telling myself, "oh I need to see backward knight moves better," I would categorize the type of mistake I made whether it's knowledge-based or skills-based and go deeper from there. I think it can be beneficial since it would be an evaluation of myself as a chess player instead of regular game analysis which primarily focuses on whether I made the right moves or not. 

Edit:

Here is another great explanation of these ideas from GM R.B. Ramesh talking with Kostya in a Chess Dojo video.