Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Site Review - World Chess Championship
0 views
Skip to first unread message
Mark Weeks
Mar 15, 2000, 3:00:00 AM3/15/00
to
The next review, following the Chess History bookmarks, is for the World
Chess Championship by Mark Weeks. If the name looks familiar, that's
because the site is my own. It's listed in the bookmarks at address...
http://www.mark-weeks.com/chess/wcc-indx.htm
...I'm not going to review my own site in any detail. I know it's
strengths & weaknesses too well, and anything I say would not be
objective. Instead, I'm going to look at what interests the visitors to
the site. In which world championship events are the site's visitors
most interested & least interested? The page access counts provided by
the site operator give enough clues to answer these questions.
First, how popular is the world chess championship? GoTo.Com has a
feature which returns the 100 most popular search terms containing the
word 'chess'. In February, for example, GoTo counted 12544 searches on
the single word 'chess'. The second most popular term was 'chess game'
which had 1632 searches.
GoTo is not one of the most popular search engines, but the search terms
are probably representative of those used on the larger search engines.
Comparing the popularity of chess with sports, there were 16524 searches
on 'soccer', 14958 on 'football', and 10115 on 'tennis'. Comparing chess
with other games, there were 6070 searches on 'poker', 2719 on
'backgammon', and 2252 on 'checkers'.
Search terms including 'chess' plus the word 'history' appear twice in
the top 100, with a total of 231 searches. There are no instances in the
top 100 of 'chess' plus some form of 'champ'. Searches on variations of
the phrase 'chess champion' occurred 23 times, while variations of
'chess championship' occurred 22 times. This is clearly a subject of
special interest. By comparison, 'chess' with variants of the word
'download' appear 18 times in the top 100 and were used in 3191
searches.
A short explanation about the content of my site might also be useful.
The site has 328 pages, of which there are four types:-
- 21 index pages
- 128 crosstable pages dedicated to a specific world championship event
- 114 highlights pages, also for specific events
- 53 best game pages for specific games
That doesn't add up to 328 pages, because there are a few pages which
don't fit into this classification. There are also 127 ZIP files; these
contain compressed PGN files
The index pages offer different paths to the pages for events. A
highlights page has more detailed information about an event and is
always linked from a crosstable, although a crosstable page may link to
more than one highlights page. The candidates matches are a common
example. A crosstable also provides the link to the PGN files for an
event.
Of the crosstable pages, 14 are for women's events & 13 are computer
events. The rest break down as follows:-
- 43 World championship title matches
- 26 Interzonals
- 19 Candidate events
- 9 Unofficial matches, including 1992 Fischer - Spassky & the Kasparov
- Deep Blue matches
- 3 London tournaments from the 1800s
- 1 Kasparov page covering 1998-99
Of the title matches, 17 are from the pre-FIDE period before 1948. I
recently created a page to start recording the events of the next FIDE
title tournament. This page only has a few links to news about FIDE
Commerce, but I'm counting it as a title match. The same holds for
Kasparov's efforts to play another match outside the jurisdiction of
FIDE.
The site presently draws about 100 unique visitors per day. They look at
an average of 600 pages and download 150 files. These numbers aren't
enormous, but since I introduced the site a few years ago, they have
been increasing steadily. During periods when general interest in the
world championship is high, the numbers go up. On public holidays & on
weekends, the numbers go down.
To compare the popularity of different pages, I'm going to use the
concept of weighted visits per day. This assumes that the main index
page gets 100 visits per day. The main index page is the page whose
address I gave at the beginning of this review. It's the only address
that I promote in directories & search engines, and it's the address
that most people use when they link to my site. It could also be called
the home page for the site & it is by far the most visited page.
The numbers for other pages are based on the percentage of visits
relative to the main index page. For example, the second most popular
page on my site is the index of FIDE events from 1948 to 1990. It gets
16.5% of the visits that the main index page gets, so I'll say that it
gets 16.5 weighted visits per day. Clear? If not, then let's just
consider it as a relative measure of popularity -- a page which gets 10
weighted visits per day is twice as popular as a page which gets 5.
After the two index pages, crosstable pages round out the five most
popular pages on the site. They are the main page for the 1999 FIDE
championship in Las Vegas, the new page for FIDE Commerce, & the page
dealing with Kasparov's ongoing struggles to find a sponsor for a new
title match. These pages each get about 13 weighted visits per day. Why
are these the most popular? Being near the top of the main index page
certainly doesn't hurt. I'm also convinced that many visitors are
searching for current news about the world championship & are less
interested in older events.
Looking at the other index pages, the index of pre-FIDE events & the
index of best games get 11 weighted visits per day. Rounding out the top
five is the site map, which gets 9. By comparison, the index of computer
events gets 6 weighted visits per day, while the index of women's events
gets only 3.
Looking at the crosstable pages, I've already mentioned that three are
among the five most popular pages on the site. Rounding out the top five
crosstable pages are the 1998 Karpov - Anand match and the 1972 Fischer
- Spassky match. These get about 7 weighted visits per day. The most
popular Candidates event is the series of 1994-95 FIDE matches at 5,
while the most popular Interzonal is from the same FIDE cycle and gets
4.
What's the most popular crosstable page from the first four decades of
FIDE events?
- from the 1950s, it's the first Botvinnik - Tal match (actually played
early in 1960) which gets 3.1 weighted visits per day;
- from the 1960s, the Botvinnik - Petrosian match at 2.7;
- from the 1970s (excluding the 1972 match), the 1975 forfeit by Fischer
to Karpov at 3.6;
- from the 1980s, the first (1984) Karpov - Kasparov match at 3.8.
Surprisingly, the most popular pre-FIDE event is the series of 1834
Labourdonnais - McDonnell matches at 3.4. It beats the 1927 Capablanca -
Alekhine match by a nose. I suspect that the apparent interest in the
1834 match is due to the structure of the site. Crosstables are linked
together in chronological order so that visitors can go directly from
one crosstable to the next without having to go back through an index.
The crosstable for the most recent event (currently the FIDE event for
the year 2000) has as its 'next' link the 1834 match, so the popularity
of the most recent event trickles down to the event which it links as
next.
What are the least popular events? There are five events clustered at
1.5 weighted visits per day. Four of them are pre-FIDE events. The least
popular event of the FIDE era is the 1985 Interzonal Tournament at
Tunis, won by Yusupov.
The highlights pages give an even better indication of an event's
popularity. A visitor normally clicks once to get a crosstable & then a
second time to get a highlights page. Anyone clicking two links to get
information on a topic shows real interest in that topic.
Looking at the highlights pages, the most popular is again the 1972
Fischer - Spassky match at 4.6 weighted visits per day. This is followed
by the 1995 Kasparov - Anand match (3.1), the 1998 Karpov - Anand match
(2.6), the 1997 Kasparov - Deep Blue match (2.5), and the first Karpov -
Kasparov match (2.2). The most popular pre-FIDE event is again the 1834
Labourdonnais - McDonnell match (2.1), followed again by the 1927
Capablanca - Alekhine match (1.6).
The least popular highlights pages are clustered at 0.4 weighted visits
per day & there are 15 of them. Sometimes I wonder if the only visitors
to these pages are search engine spiders. The least popular event of all
is the 1979 Riga Interzonal, won by Tal.
Taken together, the 53 best game pages get about 30 weighted visits per
day. The most popular is game 10 of the 1995 Kasparov - Anand match at
1.1, which is the last game on the index page. Also in the top four are
games 6 & 10 from the 1972 Fischer - Spassky match, and game 17 of the
1834 Labourdonnais - McDonnell match, which is the first on the page. It
seems that in lists, position counts for a lot. The first & last items
in a list often get more clicks than other items in the same list. The
least popular of the best games is the Smyslov - Euwe game from round 19
of the 1948 title tournament.
The download counts for PGN files are another indication of popularity.
The most popular files are the 1999 FIDE championship, the 1998 Karpov -
Anand match, the 1995 Kasparov - Anand match, the 1972 Fischer - Spassky
match, and the 1907 Lasker - Marshall match. When a visitor requests a
directory list of all downloadable files, the 1907 match is the first in
the list, which probably accounts for its popularity. The files in this
list have cryptic names & many visitors undoubtedly download the first
to find out what the files contain. Excluding the 1907 match, the most
popular pre-FIDE event is again the 1927 Capablanca - Alekhine match.
The least popular of all men's events is the 1982 Toluca Interzonal, won
by Portisch & Torre.
Another statistic I looked at for this review is the ratio of crosstable
views to PGN downloads for each event. I'm not sure what this number
means, but there are significant variations across the 124 events where
a crosstable links to a PGN file. The average for the entire site is
1.9, which means that crosstable pages are viewed 1.9 times for every
file downloaded. The highest ratio is 3.6 for the page covering the two
Kasparov - Deep Blue matches. This means that only one visitor in 3.6
downloaded the corresponding games for that page.
The games for the 1907 match, which we just saw was one of the most
popular PGN files, has a ratio of 0.8. This means that more visitors are
downloading the file than are viewing the crosstable. This confirms my
suspicion that many visitors are selecting this file because it is the
first in the list of all PGN files. Most of the women & computer events
also have ratios less than 1.0, which means that many downloads are not
initiated by clicking the link on the crosstable page, but by clicking
sequentially through the list of PGN files.
Excluding the 27 women & computer events raises the average ratio for
the entire site to 2.1. The ratios for the different types of men's
events are:-
- 1.8 for title events
- 2.4 for candidates events
- 2.3 for interzonal events
Why should title events have a lower download ratio? Probably because
there is more interest in collecting these games. On the other hand, I
would expect many visitors to have already found these PGN files on
other, better known sites, like UPitt.
The last thing I want to mention is advertising. I've experimented with
banner ads and external content to see if any of this is interesting to
the site's visitors. The answer, in a word, is 'no'. If a banner is not
related to chess, almost no one clicks on it. If a banner is related to
chess, 0.5-1.0% of the visitors to the page will click on it. Banners
are generally placed at the bottom of my pages. If I placed them at the
top, they might receive more attention, but I doubt that this would
improve clickthru rates dramatically.
Clickthru rates can be very high for targeted banners. On one page I had
a banner which linked to material directly related to the content of the
page. More than 5% of the visitors to the page clicked on that banner,
even though it was poorly designed and unattractive. This confirmed an
important point -- visitors come to the site looking for information on
the world chess championship -- anything else is a distraction.
All of this raises an important question. When considering new material
for a history site, should a site owner concentrate on what interests
visitors -or- provide balance across all historical events? I'd be
interested in feedback on this.
---
That brings us to the end of the Chess History bookmarks. We've looked
at 20 sites since the first review was posted to the discussion group
last June -- three have disappeared from the Web and two were removed as
bookmarks.
Continuing the series, we'll be looking next at the eight sites which
have been added since the first review. These will be:-
- About.com's Chess History - by David Dunbar
- Alphonso X Book of Games - by Christian J. Hartmann
- Chess Downloads - by Klaus Wrba
- Chess Graphics - by Alan Cowderoy
- Chess Variants - by Hans Bodlaender
- La storia, personaggi e biblio - by Marco A. Alberti
- Max Euwe Centrum - Amsterdam, NL
- Schach-Ereignisse - by Erhard Frolik
Bye for now,
Mark Weeks
Chess Championship by Mark Weeks. If the name looks familiar, that's
because the site is my own. It's listed in the bookmarks at address...
http://www.mark-weeks.com/chess/wcc-indx.htm
...I'm not going to review my own site in any detail. I know it's
strengths & weaknesses too well, and anything I say would not be
objective. Instead, I'm going to look at what interests the visitors to
the site. In which world championship events are the site's visitors
most interested & least interested? The page access counts provided by
the site operator give enough clues to answer these questions.
First, how popular is the world chess championship? GoTo.Com has a
feature which returns the 100 most popular search terms containing the
word 'chess'. In February, for example, GoTo counted 12544 searches on
the single word 'chess'. The second most popular term was 'chess game'
which had 1632 searches.
GoTo is not one of the most popular search engines, but the search terms
are probably representative of those used on the larger search engines.
Comparing the popularity of chess with sports, there were 16524 searches
on 'soccer', 14958 on 'football', and 10115 on 'tennis'. Comparing chess
with other games, there were 6070 searches on 'poker', 2719 on
'backgammon', and 2252 on 'checkers'.
Search terms including 'chess' plus the word 'history' appear twice in
the top 100, with a total of 231 searches. There are no instances in the
top 100 of 'chess' plus some form of 'champ'. Searches on variations of
the phrase 'chess champion' occurred 23 times, while variations of
'chess championship' occurred 22 times. This is clearly a subject of
special interest. By comparison, 'chess' with variants of the word
'download' appear 18 times in the top 100 and were used in 3191
searches.
A short explanation about the content of my site might also be useful.
The site has 328 pages, of which there are four types:-
- 21 index pages
- 128 crosstable pages dedicated to a specific world championship event
- 114 highlights pages, also for specific events
- 53 best game pages for specific games
That doesn't add up to 328 pages, because there are a few pages which
don't fit into this classification. There are also 127 ZIP files; these
contain compressed PGN files
The index pages offer different paths to the pages for events. A
highlights page has more detailed information about an event and is
always linked from a crosstable, although a crosstable page may link to
more than one highlights page. The candidates matches are a common
example. A crosstable also provides the link to the PGN files for an
event.
Of the crosstable pages, 14 are for women's events & 13 are computer
events. The rest break down as follows:-
- 43 World championship title matches
- 26 Interzonals
- 19 Candidate events
- 9 Unofficial matches, including 1992 Fischer - Spassky & the Kasparov
- Deep Blue matches
- 3 London tournaments from the 1800s
- 1 Kasparov page covering 1998-99
Of the title matches, 17 are from the pre-FIDE period before 1948. I
recently created a page to start recording the events of the next FIDE
title tournament. This page only has a few links to news about FIDE
Commerce, but I'm counting it as a title match. The same holds for
Kasparov's efforts to play another match outside the jurisdiction of
FIDE.
The site presently draws about 100 unique visitors per day. They look at
an average of 600 pages and download 150 files. These numbers aren't
enormous, but since I introduced the site a few years ago, they have
been increasing steadily. During periods when general interest in the
world championship is high, the numbers go up. On public holidays & on
weekends, the numbers go down.
To compare the popularity of different pages, I'm going to use the
concept of weighted visits per day. This assumes that the main index
page gets 100 visits per day. The main index page is the page whose
address I gave at the beginning of this review. It's the only address
that I promote in directories & search engines, and it's the address
that most people use when they link to my site. It could also be called
the home page for the site & it is by far the most visited page.
The numbers for other pages are based on the percentage of visits
relative to the main index page. For example, the second most popular
page on my site is the index of FIDE events from 1948 to 1990. It gets
16.5% of the visits that the main index page gets, so I'll say that it
gets 16.5 weighted visits per day. Clear? If not, then let's just
consider it as a relative measure of popularity -- a page which gets 10
weighted visits per day is twice as popular as a page which gets 5.
After the two index pages, crosstable pages round out the five most
popular pages on the site. They are the main page for the 1999 FIDE
championship in Las Vegas, the new page for FIDE Commerce, & the page
dealing with Kasparov's ongoing struggles to find a sponsor for a new
title match. These pages each get about 13 weighted visits per day. Why
are these the most popular? Being near the top of the main index page
certainly doesn't hurt. I'm also convinced that many visitors are
searching for current news about the world championship & are less
interested in older events.
Looking at the other index pages, the index of pre-FIDE events & the
index of best games get 11 weighted visits per day. Rounding out the top
five is the site map, which gets 9. By comparison, the index of computer
events gets 6 weighted visits per day, while the index of women's events
gets only 3.
Looking at the crosstable pages, I've already mentioned that three are
among the five most popular pages on the site. Rounding out the top five
crosstable pages are the 1998 Karpov - Anand match and the 1972 Fischer
- Spassky match. These get about 7 weighted visits per day. The most
popular Candidates event is the series of 1994-95 FIDE matches at 5,
while the most popular Interzonal is from the same FIDE cycle and gets
4.
What's the most popular crosstable page from the first four decades of
FIDE events?
- from the 1950s, it's the first Botvinnik - Tal match (actually played
early in 1960) which gets 3.1 weighted visits per day;
- from the 1960s, the Botvinnik - Petrosian match at 2.7;
- from the 1970s (excluding the 1972 match), the 1975 forfeit by Fischer
to Karpov at 3.6;
- from the 1980s, the first (1984) Karpov - Kasparov match at 3.8.
Surprisingly, the most popular pre-FIDE event is the series of 1834
Labourdonnais - McDonnell matches at 3.4. It beats the 1927 Capablanca -
Alekhine match by a nose. I suspect that the apparent interest in the
1834 match is due to the structure of the site. Crosstables are linked
together in chronological order so that visitors can go directly from
one crosstable to the next without having to go back through an index.
The crosstable for the most recent event (currently the FIDE event for
the year 2000) has as its 'next' link the 1834 match, so the popularity
of the most recent event trickles down to the event which it links as
next.
What are the least popular events? There are five events clustered at
1.5 weighted visits per day. Four of them are pre-FIDE events. The least
popular event of the FIDE era is the 1985 Interzonal Tournament at
Tunis, won by Yusupov.
The highlights pages give an even better indication of an event's
popularity. A visitor normally clicks once to get a crosstable & then a
second time to get a highlights page. Anyone clicking two links to get
information on a topic shows real interest in that topic.
Looking at the highlights pages, the most popular is again the 1972
Fischer - Spassky match at 4.6 weighted visits per day. This is followed
by the 1995 Kasparov - Anand match (3.1), the 1998 Karpov - Anand match
(2.6), the 1997 Kasparov - Deep Blue match (2.5), and the first Karpov -
Kasparov match (2.2). The most popular pre-FIDE event is again the 1834
Labourdonnais - McDonnell match (2.1), followed again by the 1927
Capablanca - Alekhine match (1.6).
The least popular highlights pages are clustered at 0.4 weighted visits
per day & there are 15 of them. Sometimes I wonder if the only visitors
to these pages are search engine spiders. The least popular event of all
is the 1979 Riga Interzonal, won by Tal.
Taken together, the 53 best game pages get about 30 weighted visits per
day. The most popular is game 10 of the 1995 Kasparov - Anand match at
1.1, which is the last game on the index page. Also in the top four are
games 6 & 10 from the 1972 Fischer - Spassky match, and game 17 of the
1834 Labourdonnais - McDonnell match, which is the first on the page. It
seems that in lists, position counts for a lot. The first & last items
in a list often get more clicks than other items in the same list. The
least popular of the best games is the Smyslov - Euwe game from round 19
of the 1948 title tournament.
The download counts for PGN files are another indication of popularity.
The most popular files are the 1999 FIDE championship, the 1998 Karpov -
Anand match, the 1995 Kasparov - Anand match, the 1972 Fischer - Spassky
match, and the 1907 Lasker - Marshall match. When a visitor requests a
directory list of all downloadable files, the 1907 match is the first in
the list, which probably accounts for its popularity. The files in this
list have cryptic names & many visitors undoubtedly download the first
to find out what the files contain. Excluding the 1907 match, the most
popular pre-FIDE event is again the 1927 Capablanca - Alekhine match.
The least popular of all men's events is the 1982 Toluca Interzonal, won
by Portisch & Torre.
Another statistic I looked at for this review is the ratio of crosstable
views to PGN downloads for each event. I'm not sure what this number
means, but there are significant variations across the 124 events where
a crosstable links to a PGN file. The average for the entire site is
1.9, which means that crosstable pages are viewed 1.9 times for every
file downloaded. The highest ratio is 3.6 for the page covering the two
Kasparov - Deep Blue matches. This means that only one visitor in 3.6
downloaded the corresponding games for that page.
The games for the 1907 match, which we just saw was one of the most
popular PGN files, has a ratio of 0.8. This means that more visitors are
downloading the file than are viewing the crosstable. This confirms my
suspicion that many visitors are selecting this file because it is the
first in the list of all PGN files. Most of the women & computer events
also have ratios less than 1.0, which means that many downloads are not
initiated by clicking the link on the crosstable page, but by clicking
sequentially through the list of PGN files.
Excluding the 27 women & computer events raises the average ratio for
the entire site to 2.1. The ratios for the different types of men's
events are:-
- 1.8 for title events
- 2.4 for candidates events
- 2.3 for interzonal events
Why should title events have a lower download ratio? Probably because
there is more interest in collecting these games. On the other hand, I
would expect many visitors to have already found these PGN files on
other, better known sites, like UPitt.
The last thing I want to mention is advertising. I've experimented with
banner ads and external content to see if any of this is interesting to
the site's visitors. The answer, in a word, is 'no'. If a banner is not
related to chess, almost no one clicks on it. If a banner is related to
chess, 0.5-1.0% of the visitors to the page will click on it. Banners
are generally placed at the bottom of my pages. If I placed them at the
top, they might receive more attention, but I doubt that this would
improve clickthru rates dramatically.
Clickthru rates can be very high for targeted banners. On one page I had
a banner which linked to material directly related to the content of the
page. More than 5% of the visitors to the page clicked on that banner,
even though it was poorly designed and unattractive. This confirmed an
important point -- visitors come to the site looking for information on
the world chess championship -- anything else is a distraction.
All of this raises an important question. When considering new material
for a history site, should a site owner concentrate on what interests
visitors -or- provide balance across all historical events? I'd be
interested in feedback on this.
---
That brings us to the end of the Chess History bookmarks. We've looked
at 20 sites since the first review was posted to the discussion group
last June -- three have disappeared from the Web and two were removed as
bookmarks.
Continuing the series, we'll be looking next at the eight sites which
have been added since the first review. These will be:-
- About.com's Chess History - by David Dunbar
- Alphonso X Book of Games - by Christian J. Hartmann
- Chess Downloads - by Klaus Wrba
- Chess Graphics - by Alan Cowderoy
- Chess Variants - by Hans Bodlaender
- La storia, personaggi e biblio - by Marco A. Alberti
- Max Euwe Centrum - Amsterdam, NL
- Schach-Ereignisse - by Erhard Frolik
Bye for now,
Mark Weeks
Sent via Deja.com http://www.deja.com/
Before you buy.
Mark Weeks
Mar 15, 2000, 3:00:00 AM3/15/00
to
In article <8ank88$91g$1...@nnrp1.deja.com>,
Mark Weeks <10004...@compuserve.com> wrote:
> In February, for example, GoTo counted 12544 searches on
> the single word 'chess'.
[...]
> Comparing the popularity of chess with sports, there were 16524
searches
> on 'soccer', 14958 on 'football', and 10115 on 'tennis'.
Can it really be that chess is a more popular search target than tennis?
It occurred to me that since these numbers were based on a single search
word, it might be that 'tennis' was used in more search phrases. I
decided to compute the total number of searches using these keywords...
football 45779
chess 31380
tennis 28104
...which means that the top 100 search terms containing the word
football produced a total of 45779 searches. Chess beats tennis again,
but I still don't believe that chess is more popular than tennis. - MW
Mark Weeks <10004...@compuserve.com> wrote:
> In February, for example, GoTo counted 12544 searches on
> the single word 'chess'.
> Comparing the popularity of chess with sports, there were 16524
searches
> on 'soccer', 14958 on 'football', and 10115 on 'tennis'.
It occurred to me that since these numbers were based on a single search
word, it might be that 'tennis' was used in more search phrases. I
decided to compute the total number of searches using these keywords...
football 45779
chess 31380
tennis 28104
...which means that the top 100 search terms containing the word
football produced a total of 45779 searches. Chess beats tennis again,
but I still don't believe that chess is more popular than tennis. - MW
0 new messages