How data analytics have changed the NBA and elite chess
I recently came across one on X (formerly Twitter). Comment by grandmaster and world champion coach Peter Heine Nielson. In this he refers to the book [Goldsberry 2024], which will be released in May 2024. The author promotes the book on X with the image above. It shows the changes in shot selection in the NBA from the 2023/2004 season to the 2023/2024 season (Source: [Goldsberyy 2024])
For those readers who are not very familiar with basketball and the NBA, a few caveats. The National Basketball Association (kurz: NBA) gilt by far the strongest and most popular basketball league in the world. The NBA currently consists of 30 teams from the USA and Canada. The following image shows the basketball court.
Figure 2: The basketball court (Source: Wikipedia)
Successful throws beyond the three-point line score 3 points, throws inside the line score 2 points. The probability that a shooter will successfully convert a shot decreases as the distance to the basket increases. Therefore, it is not surprising that 3-point throws near the baseline have always been popular, as the distance between the three-point line is slightly shorter there than in the middle of the field (6.60m vs. 7.24m).
Figure 1 shows a major change in how NBA teams play. While previously (2003/2024, left side of the figure) the shooters also made good use of the two-point range, today (2023/2024, right side of the figure) they concentrate on three-point throws and throws close to the basket. There are hardly any shots from medium distance anymore. There is also now an increasing tendency to take three-point throws one or two steps behind the three-point line.
Anyone who has watched an NBA game recently (e.g. on YouTube) can confirm this. While smaller players like Stephen Curry (1.88m tall) are now increasingly aggressive towards the basket, taller players like Nikola Jokić (2.11m tall) can now also score three points very well. And with Luka Dončić you sometimes get the impression that he takes his three-point throws just behind the halfway line.
Figure 3: Current top shooters in the NBA Stephen Curry, Nikola Jokić and Luka Dončić (Source: Wikipedia)
These changes in the game are definitely not a coincidence. All NBA teams now have a team of data analysts who advise the coaches. These analyzes will have shown that a three-point throw is more worthwhile than a two-point throw from mid-range. This knowledge will then have an impact on the signing of new players, so that there are now, for example, more players who are good at three-point throws or who can develop good shooting positions near the basket. The whole thing will continue in training, where such shots and the ability to play such shots will be practiced more and more.
Traditional basketball fans in the USA are not happy about this development. There are currently discussions there about moving the three-point line further back and completely abolishing the three-point line near the baseline. But there will certainly be no changes in the short term.
What about chess?
I have the impression that we are seeing a similar development in chess. Now we chess players don’t throw the ball at a basket. But we choose our openings carefully. And I think that in top chess we see a lot less different openings these days than we used to.
Therefore, I analyzed whether my impression is correct or whether I am mistaken. To do this, I analyzed which openings the top players used before (here I chose the year 1990, back then Kasparov, Karpov and Kortchnoi were still active)) and today (here I chose 2019 to avoid distortions caused by the many online tournaments during Covid) play. I defined the term top player as TOP100 player.
The definition of top players brought with it the first problem. The ELO numbers of the top players have changed significantly upwards in recent years. In 1990, an ELO number of 2540 was enough to be in the TOP 100, but in 2019 this bar was already at 2654.
Figure 4: Development of the ELO numbers for top players since 1990 (author’s evaluation based on the FIDE ELO lists from July of one year)
I used the Chessbase Mega2023 database as a data source. I initially saved this database in PGN format. I then extracted the following data per game:
- Event
- Datum
- ELO White
- NOW Black
- Result
- ECO-Code
In the next step I edited the data as follows:
- All games from 1990 are taken into account if both players had an ELO rating >= 2540
- All games from 2019 are taken into account if both players had an ELO rating >= 2654
- Filtering out all games without a result or without an ECO code (this should also eliminate all Chess960 games)
- Try to filter out rapid and blitz games. To do this, I took out games that contained patterns such as “Blitz”, “Rapid”, “5’” or “Titled Tuesday” in the event field. I certainly didn’t catch all of the rapid and blitz games, but hopefully enough to avoid any distortion caused by such games.
- Removing duplicates
Using this database, I then determined the 30 most common ECO codes and their frequency for 2019 and 1990. These are listed in the following figure:
2019
ECO-Code
frequency
1990
ECO-Code
frequency
1.
C54
6,62%
1.
E15
3,42%
2.
C67
3,82%
2.
Q92
1,90%
3.
C42
3,14%
3.
E12
1,82%
4.
E05
2,97%
4.
E18
1,52%
5.
D37
2,80%
5.
E32
1,52%
6.
C65
2,62%
6.
D85
1,37%
7.
B90
2,05%
7.
C11
1,37%
8.
B51
1,71%
8.
A30
1,29%
9.
D38
1,60%
9.
B08
1,14%
10.
D02
1,54%
10.
C13
1,06%
11.
B31
1,48%
11.
B09
1,06%
12.
B12
1,37%
12.
A33
0,99%
13.
B33
1,37%
13.
E38
0,91%
14.
A13
1,37%
14.
Q10
0,91%
15.
A29
1,25%
15.
A46
0,91%
16.
D85
1,25%
16.
A37
0,91%
17.
D78
1,20%
17.
A29
0,91%
18.
E11
1,20%
18.
E11
0,91%
19.
A28
1,14%
19.
D37
0,91%
20.
A34
1,14%
20.
E05
0,91%
21.
A20
1,03%
21.
B93
0,91%
22.
C78
0,91%
22.
B12
0,84%
23.
C84
0,86%
23.
E97
0,84%
24.
C88
0,86%
24.
E20
0,84%
25.
E04
0,86%
25.
E94
0,84%
26.
D27
0,86%
26.
C18
0,84%
27.
D30
0,80%
27.
A26
0,76%
28.
C89
0,80%
28.
A13
0,76%
29.
A30
0,74%
29.
A90
0,76%
30.
E06
0,68%
30.
B07
0,68%
total
50,03%
total
33,79%
Figure 5: The 30 most common openings (identified by the ECO code) in 2019 and 1990. Entries marked in green can be found in both lists
The figure shows two things. On the one hand, the selection of openings has changed. Only 7 of the TOP 30 openings from 1990 can still be found in the TOP 30 list in 2019. No opening in the top 10 from 1990 is still in the top 10 in 2019. Openings such as French, Pirc, Dutch or King’s Indian have completely fallen out of the list of the 30 most played openings. Instead, Italian, Berlin defense and Russian now dominate.
The frequency with which the TOP openings are played has also changed significantly. In 2019, the 30 most common ECO codes account for over 50% of lots. In 1990, 57 ECO codes were needed to reach 50%.
Why is that? I also generated the following statistics.
Figure 6: Development of the draw rate in top games since 1990
The figure shows that the draw rate in games between top players has risen sharply since 1990. While this rate was around 57% in 1990, this rate has already risen to almost 65% in 2019.
All in all, I come to the conclusion that black players in particular try to reduce their risk by choosing their openings. You will be rewarded with a higher draw rate. Risky openings such as French, Pirc, Dutch or King’s Indian are now only used by the top players as a surprise weapon.
From a black perspective, a real success story. Therefore, it cannot be assumed that this will change in the future.
Closing remarks
The statistics shown show that my feeling was correct. In top chess you concentrate on a few openings. Openings that pose a higher risk, especially for Black, are rarely seen anymore.
Similar to traditional basketball fans in the USA, I am not happy with this development. I also think it’s a shame that this creates a gap between top chess and the rest. In the past, many people (myself included) tried to emulate the top players in the opening. When Garry played Najdorf or King’s Indian, these were openings that were also encountered in the lower divisions. But I can’t remember seeing the Berlin defense in a team tournament or at an amateur tournament in the last 20 years.
But I guess I’ll have to come to terms with it. Unfortunately, solutions such as moving the three-point line do not exist in chess.
Source directory
[Goldsberry 2024]: Kirk Goldsberry, Hoop Atlas: Mapping the Remarkable Transformation of the Modern NBA, Mariner Verlag
2024-04-24 13:03:31
#data #analytics #changed #NBA #elite #chess
