Most players create a collection of full and partial games for their own pleasure or study.
Themes include favorite games of a famous player, beloved openings, offbeat or unusual mating patterns, miniatures, tactical sacrifices, positional considerations, pawn endings, rook endings, and even games they have played.
Players have been collecting games at least since Greco, who published his games sometime after 1500. He covered openings and spectacular wins in the opening, and claimed to have played all the games in his collection (but the modern opinion is that he simply copied at least some of the games from other collections and made an anthology of them).
Players have made their collections from index cards. This works well if you are deciding whether a new move in an opening is worth analyzing. You write down a game with the new move on an index card. Repeat. And keep going.
When you have collected enough index cards with the new move, you can place them on a table, organize them, and then play over the games on the cards. This is a flexible method as you can easily discard a game and substitute another card that you think will be a better study.
When done, organize the cards by any criteria you want, and then attach a rubber band around them and they are ready to use again.
Index cards are also good if you want to collect all the games of a players. When a new game is played, all you have to do is copy the newest game on a card and you are ready to add it to your collection. It also easy find out what he is most likely to do when he has to face your pet Najdorf variation.
Chess players have also created scrapbooks full of interesting and helpful games, problems, analyses, and the like from newspapers, magazines, and even old books.
This makes such collections much easier to travel and can hold more information than index cards.
For those players who can write also small, small note pads have been used to write down games and the occasional study or problem. Usually, you would meet them at tournaments, and they would be collecting games for publication. But this method was (and is) not meant to be long-lasting. Instead, the reporter, editor, or fellow chess player would transfer the game and notes to a more permanent medium.
But with the almighty laptop (and the Internet), it is now far easier to create a collection, make it a more-or-less a continuing collection and update it at one’s leisure.
The most common method of game collection in the Internet age is to use a PGN generator. PGN is short for Portable Game Notation, it is what the computer uses to display the moves of a game, and it looks like this:
[Event “?”]
[Site “Kiev”]
[Date “1954.??.??”]
[White “Kutsenok”]
[Black “Akimov”]
[Result “0-1”]
[ECO “C35”]
[PlyCount “30”]
1.e4 e5 2.f4 exf4 3.Nf3 Be7 4.Bc4 Nf6 $1 5.e5 Ng4 6.O-O d5 7.Bb3 $2 Nc6 8.d4 g5 9.h3 $2 h5 $1 10.hxg4 hxg4 11.Nh2 g3 12.Ng4 Rh4 13.Nf6+ Bxf6 14.exf6 Qxf6 15.Re1+ Kf8 0-1 {Black kingside pawns threaten to advance.}
A few notes here. The computer will translate a “$1” as an “!” and a “$2”as a “?”. Any notes in a { } will allow you to read the note when you play over the game on a computer, and a ply is exactly ½ of a move. Hence, this game is 15 moves, and the ply is 30. The game is from Clarke’s 100 Soviet Miniatures.
The big drawback of this method is that you need a computer to play, study, or simply enjoy the game. And in the case of electrical power going down, you can be out of luck.
With the rise of on-line publishing, players can also upload and print books of their favorite games, players, openings, and ideas on the game. This can be expensive (on-line printers need to make their profit), but it is rather easy to add new games and ideas, and even change one’s own annotations as fast as you can type.
Which is an important skill when writing anything on the Internet, social sites, and even a blog.
And now a game from Tal’s Tactical Treatment of an opponent. Maybe this is something you can add to your collection.

GM Tal-Kennedy
World Student Ch.
Varna, 1958
[Minev, “Digging into the Most Notorious Bulletin”, Inside Chess, Sept. 5, 1994]
1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.a3 Bxc3+ 6.bxc3 Ne7 7.Qg4 cxd4 8.Bd3 (Tal’s favorite continuation throughout his career and one that theory still gives as “unclear”.) 8…Nbc6?! (The immediate 8…Qa5 9.Ne2 is critical … After the text, White has no problems with his e5-pawn.) 9.Nf3 Bd7 (If now 9…Qa5, then 10.O-O.) 10.cxd4 Qc7 11.O-O O-O-O 12.a4 f5 13.Qxg7 h5 14.Qg5 Rdg8 15.Qd2 Na5 16.Ba3 Nec6 17.Bd6 Qd8 18.Qc3 Be8 19.Rab1 Rg7 20.Bb4 h4 21.Bxa5 Qxa5 22.Qxa5 Nxa5 23.Rfe1 h3 24.g3 Nc4 25.Bxc4 dxc4 26.d5! f4 27.Re4! exd5 28.Rxf4 Bxa4 29.Nd4 Re7 30.Rf5 Kc7 31.f4 Rd8 32.Rf6 Bd7 33.Rd6! b5 34.Rxd5 a6 35.f5 Rf8 36.Rf1 Bc8 37.f6 Rd7? 38.Ne6+ 1-0