Puzzle Improvement but no game improvement

Since Jan 1st my puzzle rating has improved from 1475 to 1900. But, in the same timeframe my rapid score has improved from 766 to 785, a paltry 19 points. The range of that is also very constricted. I have also done an extensive amount of puzzle rush and easier custom puzzles to work on tactics and mate in 2. Its obvious that puzzles are not helping. Is this common?  

I felt the same thing, puzzle rating and rapid rating are in different area.

The advantage of studying puzzles is that it is a great way to develop tactical skill. Since most blunders and victories are determined by tactical situations, good tactical skill can help you use your opponent's mistakes when there is a chance to create a big winning position. At the same time, it can also help you notice tactics that opponent can use and prevent them.
 But a limit of studying puzzle is that it only deals with specific situations. It doesn't improve overall understanding of chess.
 So there's a limit to playing with tactics alone. There are other important skills in chess like strategies, openings and endgames that determine your level of chess.
 Tactics allow you to do the right thing at critical moments, strategies help you find what to do and create opportunities for various tactics, openings make your position better before getting into the middlegame, and endgames allow you to win when there is subtle difference after middlegame is over.
 At the beginner's level, most wins and losses are determined by tactics. You can win just by avoiding a blunder.
 But in the process of becoming an intermediate player, you can no longer win with tactics alone. How to plan becomes an important topic, and this is the area of strategy.
 Understanding strategies is a key part of chess skills. If you're not familiar with strategies, you'll have more moments when you don't know what to do, and it'll be difficult to figure out what your opponent is planning for. Understanding strategies can help you find a precise move when it seems like a lot of moves are ok.

Well, so...I guess these videos could help you, if you have not watched yet. They helped me learn concept about positional chess.

Yes, this is very common. My advice would be to play longer time controls (15/10 or slower) and focus on trying your very best to play games with 0 blunders (playing focused chess without distractions, do a final blunder check right before you make your move, keep a running tally of loose/undefended pieces, ask why your opponent made the last move, etc.). I know it may seem boring, but if you can get to where the majority of your games go without you blundering pieces or checkmate, you’ll reach 1000.

Also whenever you solve puzzles you should have a direct goal in mind (either improving calculation or pattern recognition). To improve pattern recognition, set the custom puzzle rating to below 1000 (to a point where they are easy for you) and do as many as you can as quickly as possible. This should be instant.

If you are trying to improve your calculation/visualization (this helps vastly reduce blunders), focus on calculating the puzzle until you are sure you’ve found all the responses your opponent THEN play the moves. This will show you the gaps you have in your calculation and what mistakes you might have made. Guessing when you find the first promising move might increase your rating temporarily but it won’t help you improve

Yeah

Tactics puzzles are overrated. Solving 4 tactics puzzles is a great warm-up, but that is about it.
In a real game nobody tells you there is a tactic, or for which side.
Also: how much time do you spend on a puzzle?
How much time do you spend on a move in a game?
The time per move in a game should be larger than the time per puzzle,
as you do not know if there is a tactic, or for which side.

Yeah

feel like puzzle getting better but not necessarily the game skills

That is a great question and I think it is a common problem. I made a video about solely focusing on tactics a few months ago: https://www.youtube.com/watch?v=3Es_LrdCHA0

I will publish another one on this very topic very soon! I may keep you posted here if you want!

All the best,

Can

All puzzle do is work on tactics.

If you want to get better at chess, you actually have to study, and best done at a table on a board with actual pieces.

Chess is not just tactics. There is FAR more to chess:

• Planning
• Strategy
• Calculation
• Piece Coordination
• Endgame Knowledge

Many naive players think endgame knowledge is unimportant because they think they can mate the king before then. That is one of the most unintelligent things I have ever heard. You need to know what endgame positions win and which ones draw for various reasons.

Take the following two diagrams. Both are White to move. In one case, trading Queens draws. In the other case, trading Queens loses. If you don't know your endgames, you would have no idea what to do here.

Which one draws for White after a queen trade?

The question is not "What is the best move?". The question is simply in which diagram does White trading queens lead to a draw. If you know your basic endgames, you ought to be able to answer this question in a split second.

The difference with this type of question is that you must have general knowledge to determine the answer. Tactics problems are merely the finding of forced sequences to win material.

There are different types of tactics training. You are doing custom puzzles and regular tactics on the tactics trainer. You've done roughly 1000 tactics in a 90 day period. Although this is a good amount of tactics training, in absences of woodpeckering a given set of tactics, from a book per say, it's simply not enough, especially for someone who wants to improve their rating.

There are also puzzles that come from beginner games that cannot be replicated in a tactics trainer, which are the types of tactics Dan Heisman shows in his book, Back to basics, tactics, which is why I recommended that book to you.

Someone mentioned something about planning, let me talk about that. I noticed in some of your games you lost because there simply was nothing to do and you blundered a queen, other times it seemed like you just wood pushed your way into the middlegame. So your lacking a sense of planning in your games. Chess.com has good planning videos as I don't think that subject is really covered on youtube. Just go to articles, and click on study plans, and go through the plan, you will eventually see videos on planning.

It also couldn't hurt to watch videos on the typical plans of the openings you use. might give you some more insight. I noticed you didn't know basic ideas in a particular opening you played. Just watch videos about those openings, youtube is full of videos like that.

Lastly, it takes time to improve. a 90 day period isn't that much. From Jan to now I would say your lack of the planning portion of the chess game is causing you to lag behind in your rating.

Also, I noticed you won a lot of games due to your opponent resigning when there was no reason to resign. Also, you won most of your games simply because your opponent blundered all the time, or just started giving away pieces. Though this is common at your rating, you don't want to only be winning in these types of games.

Also, learn king and pawn endgames, it'll help you grind out wins where before you just lost.

Trading Queens loses in the first one and draws in the second one.

In the first one, Black must specifically recapture with the pawn, not the king. This opens up the rook on e8 to cut off the White King. Black will maneuver to get the King to f1 and pawn to f2, and then execute the steps to win in what is called Lucena's Position. It entails pushing the White King one more file away with a check on the d-file, and then building a bridge and promoting the pawn.

The second one is a draw because after the King recapture, 2.Rh3 creates Philidor's Draw. The King is already on the file of the pawn and will only move if checked by the rook, staying on that file, toggling between 1st and second rank. Otherwise, the rook will toggle between safe squares on the third rank until the pawn is pushed to that rank, and then the rook will go all the way up the board since now the Black King has no hiding place once the pawn is advanced that far, and you will check the king from behind.

Just one of MANY factors that prove puzzles are MAYBE 2% of what you need to know.

And knowing things like Lucena's Position, Philidor's Draw, the Short Side Defense, and the Passive Defense aren't just things to think about when and if R+P vs R endings occur, it should also drive what you do before that, looking to achieve such positions. Determine whether or not to trade down.

Another pattern to know is wrong color Bishop and rook pawn. If the defending king is on the promotion square, the side with the Bishop and pawn cannot win if the Bishop is on the opposite color of the promotion square. So if Black has a light-squared Bishop and both rook pawns, White has just a knight, White can draw if he is able to sacrifice the Knight for the h-pawn and get the king to a1. The other way around would not work. If you sac for the a-pawn and move the king to the h-file, Black wins because he has the correct Bishop.

Yet another item puzzles won't teach you!

Your Puzzle Rush Survival is low. Don't solve the puzzle until you know for sure what it is. Patience. Also, for a lot of tactics you need to learn to train your mind to see why a variation won't work, because the opponent can defend or capture a piece.

Many tactical pattern puzzles rely on knowledge of basic endgames like knowing the drawing rules for Queen promotion. Honestly, know the Queen and King versus pawn and king and it will help you A LOT.

I wouldn't say switch, but do more survival training. Unless you can only do one a day, then yes, switch.

Puzzle solving mainly focuses on tactics, while actual games require a combination of tactical, strategic and positional skills. Improving one aspect may not automatically lead to improvement in others.