New van der Heijden studies database

Dr Harold van der Heijden weas born in Veghel, The Netherlands, on 18 December 1960. So: congratulations today, Harold, on your 60th.

Harold finished a PhD in 2009, while he was working at the research and development laboratory of the veterinary company Royal GD. After his PhD he became a full researcher and went on to lead a group of 20 technicians at the company.

But that is not what Harold is famous for. It is related to his hobby: chess studies, of which he has to date published around 150 (winning 28 prizes and many horourable mentions and commendations in the process).

But it is not even this for which he is famous. It is for collecting endgame studies and creating the largest collection in the world.

It all started in October 1988, when ARVES was founded. That stands for the Alexander Rueb Vereniging voor SchaakEindspelStudie – the Dutch-Flemish Association for Endgame Study. At the inauguration the late Bas de Heer gave a talk about computerised databases of endgame studies –something that had not previously existed. But the new version of ChessBase, V. 2.0, had an exciting new feature: in addition to storing whole games, you could now enter positions and record the moves played from there. "The very next day I decided to build a database of studies, beginning with the 250 rook and bishop underpromotions which I already had on cards," Harold wrote in the endgame magazine EG, of which he is the editor. "After that I thought I might as well add - all the others”. He continues:

To begin with I made copies of my database free to anyone, or perhaps just in exchange for a book. Early in 1992 the availability of my database of 23,000 studies was announced by ChessBase, at a purchase price of DM 400, after which the most frequent question put to me was 'when is the next update coming?' In 1998 my collection totalled over 50,000. Many people have helped to build up the collection, and of  course many studies were taken straight from the pages of EG. In 1994 I estimated that the total number of published studies was in the region of 75,000, of which in my opinion it might to be possible to create a collection of 90% (say 70,000) of that total.

In November 2020 Harold released the sixth edition of his endgame study database (HHdbVI). This edition contains 93,839 studies – 8,000 more than the previous version HHdbV. In addition the solutions of thousands of studies were corrected or updated. This is by far the most comprehensive collection of endgame studies available.

When HHdbVI was released Harold offered to send it to me, free of charge, in return for a review on our news page. I refused. I had received the versions I-V from him as gifts, and used these databases intensely for decades. I owed it to him to actually purchase the latest version to show my gratitude. If you wish to follow suit you can go to the Harold van der Heijden web site and order the studies database. You get a download code (by email) and can store the 62 MB database on your computer. It costs €55, and is an investment for a lifetime.

The database is in PGN-format and can be accessed by commercial chess database programs (e.g. ChessBase) or chess playing software (Fritz, Komodo, etc.). Apart from the solution the studies also include sublines and analysis. In the headers you find name(s) of the composer(s), an index code for the material on the board, the place and date of the primary source (tourney, journal, magazine) and whether it is a win or a draw study.

Within the solutions of the studies there textual comments, like <or> for an alternative move very similar to the main line (minor dual); <eg> analytical proof after the solution has ended; <cook> a move that cooks the study (+ initials to the person who found the cook).

Some statistics

HHdbVI has 93,839 endgame studies by 5,671 composers. A vast majority, 4,627, have less than ten studies in the database (2,451 have published only a single study). The average number of studies per composer is slightly over 18. The most prolific composers are:

1. Ernest Pogosyants – 2,198 studies
2. Henri Rinck – 1,792
3. Aleksey Troitzky – 1,762
4. Ladislav Prokes – 1,261
5. David Gurgenidze – 1,048
6. Michael Bent – 958
7. Michal Hlinka – 917
8. Pavel Arestov – 902
9. Mario Guido Garcia – 834
10. Iuri Akobia – 829

The numbers of studies per decade gradually increased except for the decade with WWII. Since the 1970’s on average approximately 1,000 new studies were published each year. 

I would like to give you an impression of the studies that are contained in this remarkable collection. I am going to do so by showing you a random sample – every ten thousandth study. 

In our replayer you can switch on an engine (fan icon) to analyse the positions. Note that on the right of the engine window there are three useful buttons: + and – to increase the number of lines displayed (useful to see if moves are forced and if there are reasonable alternatives):

The exclamation mark tells you what a move is threatening. This is my favourite function – I use it to better understand all the subtleties of endgame studies.

I have spent a lot of time looking at studies in the HvdH database. I estimate that I spend an average of  fifteen minutes playing through the lines given – and finding out, with chess engine support, why some tries (most of the obvious tries are given) do not work. If I did this six hours a day, without break, it would take me just over ten years to go through all the studies Harold has collected. He tells me:

The collection is an ongoing process. It will never be completed or perfect. I am not error free, but I try to be as accurate as possible. I have several tools to check a lot of details, and correct all errors I find. But still mistakes are unavoidable. It is the same feeling as when you write a book, proofread the text, word-by-word, a hundred times, to discover a mistake immediately after opening the published book for the first time...