Skip Navigation LinksHelp

The $wi$$ Immaculate Manual​

Version 7

It has been claimed that Ratings are like trying to measure the position of
a bobbing cork on turbulent water using a ruler on a piece of string in a
strong wind. This is easily believed. Therefore Ratings have been
treated accordingly. The cork has been anchored, and the string on the
ruler shortened, so that, if the cork could not be reached, no Rating
changes have been made at all. This opinion is, however, refuted by
mathematical fact. This mathematics is far simpler than the current
systems using statistics, but difficult to discover. 

Odds Ratings rely wholly on elementary probability theory, and therefore allows the
calculation of odds. Its implementation for Rating calculation and
tournament management doubles as a simulator. Called $wi$$ Immaculate,
its results are absolute proof of its efficacy, for the layman with a half
hour to spend testing its power. Rating changes can be up to 50 points
per game, and there is just one formula for all changes. Neither are
limitations required, nor do anomalies result. Perfect ranking,
consistent with game results, for an entire field of provisionally rated
players, takes one tournament. The provisional rating, bequeathed to each
new player, is the system average. It is constant, at 1500 points.
The need for Ratings to organize tournaments is obvious.

Correct appraisal of a player's ability will serve chess more generally,
encouraging and better matching Members of the fraternity. The mental
estimation of odds is easy, giving players a realistic expectation of game
results. Accurate probabilities for established Ratings can be calculated
and will be based on the latest data. Any win against a higher or equal
rated player will win 50 points. There is only one requirement to rapidly
establish a Rating within tolerance (also 50 points). Equally rated
players, from one game to the next, need to be available, until Strength is
matched. 'Odds Ratings' drive '$wi$$ Immaculate' as this table indicates.

​Win Per Loss
Rating Difference​


0. Cancel <Esc>
1. Help
2. Mathematics
3. Games
4. Tournament
5. Database
6. Fischer
7. Response
8. Printfile
9. Settings <Enter>

0. Cancel

This option in every menu returns control to the calling menu. In the
MAIN MENU it exits $wi$$ Immaculate. If a password has been supplied at
startup, it will be required here, and for every critical file changing
operation. The randomizing number requested at startup, if repeated, will
produce the same set of simulation results each time identical processes
are followed. Otherwise, each set of results will be unique.

1. Help

The Help option produces the information that you are reading now.

2. Mathematics

This facility allows high precision calculations of Ratings and
probabilities, for mathematical experimentation. Elsewhere, $wi$$
Immaculate only allows Rating points to be exchanged to the nearest whole
number. This implies that a game between players at a 718 point Rating
difference can no longer profit the better player. His chances, though,
are rated at 145 wins per loss. This will have no practical effect, but
will prevent corruption of the system average by computational rounding.
The facilities available allow game results and ratings to be entered for
odds calculations. Statistical analysis of the behavior of odds ratings,
and the effects of playing style, are also supplied. These facilities can
help to understand the nature of this important branch of mathematics, by
experimentation with the variables provided.

3. Games

Non-tournament games for Rating points should be entered here, in strictly
chronological fashion. Therefore, if a series of 'loser sits out'
lightning games are played by a small selection of players, they will need
to be written down in a fashion which will allow these to be entered in
order as played. The setting (option 9) will need to be adjusted to
Lightning. Names or identification (ID) numbers can be used to enter
results. New Names will receive the provisional Rating (the average of
1500), and may also be supplied with an assumed Rating or 'Strength'.
'Strength' is a Rating used to produce random game results for simulations.
The implied odds and the style setting will generate a mathematical
spinning wheel, proportionately divided into sectors representing game
results. This is stuck in full spin with a mathematical dart.

4. Tournament

This will be dealt with under its submenus, which manage all requirements.

5. Database

Several files are maintained by $wi$$ Immaculate, which may be managed
under the DATABASE MENU. Again, its submenus are covered below.

6. Fischer

The setting of Clocks using Fischer Increments is ideal, but somewhat
mystifying. Submenus here supply a calculator for planning matches and 
tournaments. A schedule of available playing times for each ten move
segment is provided. Various calculation methods are available.

7. Response

Banners may be delayed during processing while learning the functions
provided. These prevent errors until proficiency is achieved, for
critical responses. Thereafter, immediate response will be less
frustrating. Delays are made manifest with each choice of a response
setting. The Trial Version has optimally delayed banners by default.

8. Printfile

Tourney files (progressive tournament records) for use by $wi$$ Immaculate
are not ideal for printing. This option will generate a file that is
ideal for viewing, editing and printing. The appropriate 'Tourney.txt'
file is selected according to current settings. A reformatted version for
printing called 'Printfile.txt' is created. The display changes to
indicate the location of the created file in the database folders. If
'Printfile.txt' already exists for that Grade Category, it will be archived
and updated. Renaming a newly created printfile is offered as an option.
This identifies the event recorded, and makes that record permanent (immune
to automatic archiving).

9. Settings

There are three Categories of Members and Ratings files maintained by $wi$$
Immaculate. These are in folders (directories) called Lightning (Blitz),
Allegro (Fast Chess) and Diligent (minimum 12 minutes Clock and 3 seconds
Fischer per 2 minutes, or corresponding Non Fischer Clock limits). Each
of these categories may contain up to 9 Grades (folders A Grade to I
Grade). These allow multiple tournaments simultaneously in any Category.
The faster Categories will sponsor better games as familiarity increases
with the time constraints. An imbroglio of time limit settings, when
there is hardly time to think, would discourage the benefits of such an
excellent training opportunity, and replace fun with frustration.

$wi$$ Immaculate FILES​

The $wi$$ Immaculate file system should be explained about here. They are
in plain text and can be opened with Notepad, by double clicking, or >Right
click>Open with...>Notepad. The archived files have the extension changed
to '.old', to assist in an occasional purging. From Notepad, all files
may be sent` to the printer, except the tourney file. Tourney files have
an incompatible format, and are best converted to a printfile. Editing
the original files is highly discouraged (only the printfile has no further
use to $wi$$ Immaculate). They are in plain text for transparency of
function, but better means of recovery from disaster exist. Notably there
are the archives, or backup copies of the '0dds Ratings i386-win32' folder.

(i) The Members Files​

There is one in each Category folder, called 'Members.txt'. It contains
Names, Ratings and Strengths. A Member's ID number is simply the ordinal
position in the Members file.

(ii) The Ratings Files

Again one per Category, called 'Ratings.txt'. It contains identical lists
to the 'Members.txt' file from which it was created, but it is ranked
according to Ratings. It is also tagged with the date & time of creation
in the text, and must be updated judiciously. Updating may alter the
Grading, if Graded events are yet to be selected. This tends to indicate
that Grading needs to become a regularly scheduled event.

(iii) The Tourney Files

At most one of these can exist in any Grade within any Category, and are
called 'Tourney.txt'. A tourney file is the progressive record of a
tournament in progress, or one finalized but not yet archived. Deleting a
tourney file not only archives it, but creates (or updates and archives)
the current 'Printfile.txt'.

(iv) The Printfiles

Printfiles are called 'Printfile.txt' unless renamed. These reformatted
versions of a tourney file are suitable for viewing, editing and printing.

(v) The Archive Files

Lastly, each file deleted becomes an archive, a yyyymmddhhmmss '.old' file.
These form a low cost insurance. They make manual restoration (using
Window's facilities) to any previous state possible, so long as no relevant
archives have been purged. There is a purging option available in the
DATABASE MENU. Selective & painless deletion of all accumulated archives
is easy, whether by origin types or by archiving dates.

(vi) The Log File

Except for Menu options and Yes and No responses, every input is made using
edit mode. All transactions are recorded in the file 'Log.txt' in the
main directory. These can be recalled in reverse order with the '@'
character at any time in edit mode. The Log file may also be opened
manually and examined. It is therefore a diagnostic tool as well as a
memory aid. It will expand indefinitely at low cost, and may be emptied
at any time by manual deletion.​


0. Cancel <Esc>
1. Enter or Remove Contestants
2. Initiate a Round of Games
3. Submit Game Results
4. Random to Round Robin
5. Modify an Entry
6. Purge Last Round
7. Finalize the Tournament
8. New Grade Category Tournament
9. Settings <Enter>

0. Cancel <Esc>

Return to the previous menu.

1. Enter or Remove Contestants

Contestants can be entered at any time during a tournament. Contestants
may join by Name or ID number. If the Category Ratings file exists, a
range of players may be entered, distinguished from ID numbers as in
'1..14,16'. This will choose contestants by rank. The Ratings file is
created in the DATABASE MENU.
The removal may be expedient if a contestant never played in a round, but
will not be permitted if results are required to recalculate Ratings.
Ratings as well as contestants are enlisted together, without further
liaison with the Category Members file. Each round of games causes
recalculation of all Ratings from the beginning of the tournament. This
allows details to be modified throughout the tournament record where
necessary and feasible. Finalizing the tournament recalculates and
tallies statistics for display. Rating Points Gained (RPG), plus or
minus, are added to the Category Members file. The obvious Tie Break,
without more games played, is the highest RPG. No changes are made
permanent (or filed) until 'Yes' as well as <Enter> answers 'Update'. If
there are errors or any confusion, 'No <Enter>' will null and void all
changes since the last update. Better redo than risk a frustrating mess.

Note: The number of contestants at 14 or 15 is optional, especially for
lengthy simulations. Contestants should not exceed 16 or 17 in number,
as even this may take a while. After that, the combinatorial numbers of
possible pairings becomes too time consuming and impractical. Odds
Ratings allow a mathematically best match to be chosen from those
available. It minimizes total WPL (for the higher rated opponent of each
pair), providing the most evenly matched available overall. For both game
quality and for Rating purposes, the more evenly matched are best.
Therefore, a preferable strategy is several Graded tournaments played
simultaneously, and repeated in sequence to create the desired number of
rounds. For instance, rather than a Swiss with 45 contestants playing 30
rounds, try using A, B, & C Grade Swiss Round Robins of 15 contestants
each, Grade them again, and repeat. The efficiency and accuracy of $wi$$
Immaculate will be much improved, as may be observed using a simulation.
A sample database is provided in the DATABASE MENU for just this purpose.
The resultant Ratings file will produce incontrovertible verification of
the efficacy of ODDS RATINGS, by the ranking achieved. Ratings should
nigh order the Strengths assumed to perfection. Next, try a different

2. Initiate a Round of Games

This allows the selection of players available on the day, from among the
contestants who are enlisted in the tournament. Details of the next round
will appear, but will not become permanent until updated. It can be used
tentatively, in case further contestants emerge a little late.
Note: If you feel that the program has hung, press any key but <Esc> and
the pairing process will be displayed. <Enter> turns the display on, any
other key (except <Esc>) toggles the current setting, and <Esc> turns off
the display until the pairing has completed. Be aware that this display,
while illuminating, slows down the pairing process to 2 minutes per second.
Note also that there is no way to terminate the pairing process before it
completes except <Ctrl> <C> which will terminate the program altogether. In
that desperate resort, the last update will be intact, for a rethink and
another shot.

3. Submit Game Results

This allows game results to be entered into the tournament. Optionally, a
random entry can also be supplied, for simulations.

4. Random to Round Robin

At any point in a tournament, after a selection of contestants has been
made, this option will complete a Round Robin. The task will complete in
about two minutes with 14 or 15 contestants. The <Esc> key will exit the
Round Robin simulation upon completion of the current pairing.

5. Modify an Entry

There are limitations to modifications, but those possible can easily be
affected with this option. Each future pairing will recalculate every
result from scratch. This would take into account, for example, a
correction made for players taking the wrong sides during a game.

6. Purge Last Round

This may be the most painless way to deal with changes or disaster.

7. Finalize the Tournament

The RPG, plus or minus, are added to the Category Members file. The total
points must sum to zero, as points are only exchanged, none are created
except with each new Member. Once finalized, there can be no more done
with that tournament file, except deleting (as an archive file) and
creating, or updating (with archiving), a printfile.

8. New Grade Category Tournament

No new tournament can begin in a particular Grade Category, without first
deleting the current tourney file, if one exists. This option deletes 7
(archives) the current Grade Category tourney file, and a printfile will be
created or updated (archiving the old)

9. Settings <Enter>

See MAIN MENU description above.


0. Cancel <Esc>
1. View Category\Grade\Tourney.txt (ranked by Rating)
2. Delete Category\Grade\Tourney.txt (new tournament)
3. List Category\Members.txt (& update Category\Ratings.txt)
4. Initiate Category\Members.txt (clear for implementation)
5. Expunge a Member from Category\Members.txt (left chess)
6. Sample Category\Members.txt (simulation & training)
7. Reset Category\Members.txt (simulation & traiing)
8. Purge '.old' Files
9. Settings <Enter>​

0. Cancel <Esc>

Return to previous menu.

1. View Category\Grade\Tourney.txt (ranked by Rating)

This allows viewing of the current Grade Category tournament file.
Certain modifications are possible here, as in the TOURNAMENT MENU, unless
the tournament has been finalized.

2. Delete Category\Grade\Tourney.txt (new tournament)

Deletion (or archiving) is required before a new Grade Category tournament
can begin. This has the same effect as the TOURNAMENT MENU option 8. New
Grade Category Tournament.

3. List Category\Members.txt (& update Category\Ratings.txt)

A ranked list of the Category Members file is produced, displaying Ratings
and Strengths (for comparison in simulations). A Category Ratings file
may be created from this ordered listing. The Ratings file will allow the
use of ranked ranges of Members as tournament contestants. This greatly
simplifies the Grading of players.

4. Initiate Category\Members.txt (clear for implementation)

This will delete (archive) the Members file from the current Category.
Other Categories will not be affected. Without manually deleting the
three Category folders (using Windows facilities), each Category will need
to be cleared separately.

5. Expunge a Member from Category\Members.txt (left chess)

The removal of a Member from our main database (Category\Members.txt)
requires redistribution of that Member's excess points (the Rating minus
the average), and reassigning that Member's ID number to the most recent
Member to join, if needed.

6. Sample Category\Members.txt (simulation & training)

In order to gain familiarity with $wi$$ Immaculate, and to allow
simulations, a sample database is provided as a Category Members file.
The standard deviation of this random sample from a normal distribution is
extremely unlikely within the anticipated 0 to 3000 point range, but
designed to thoroughly test the system. The selection is a club admitting
both World Champions and Orangutans (honoring the Chess Opening, no doubt).

7. Reset Category\Members.txt (simulation & training)

This will produce a new set of random strengths for any Category Members
file. The Ratings will optionally be a duplicate of the Strength values,
or set to the provisional Rating. Together with Style settings, one tests 
Rating stability, while the other Rating convergence, respectively.

8. Purge '.old' Files

Archived files will accumulate, but use little space. With all possible
disasters under control, a purging from all Categories and Grades can be
performed in one shot using this option.

9. Settings <Enter>

This has been covered in the MAIN MENU.


0. Cancel <Esc>
1. Lightning (Blitz Chess) Clock Settings
2. Allegro (Fast Chess) Clock Settings
3. Forty Move (Average Endgame) Clock Settings
4. Sixty Move (Average Decision) Clock Settings
5. Variable Move (eg Long Games) Clock Settings
6. Hybrid Clock Settings (Clock & Increment Honing)


We may consider half of the force (not pawns) changed off the board to
constitute an endgame (Queen + 2 Piece, or 4 Pieces not a Queen). The
nature of the game, at such a point, becomes much clarified to allow a
faster speed of play. This given, statistics show that the average endgame
is reached at 40 moves, and the average decision at 60 moves. These
figures have assisted organizers to choose traditional Clock settings. A
Fischer Increment of 3 seconds per 2 minutes of Clock time produces twice
the rate of play after 40 moves. This is the basis of options below.​

0. Cancel <Esc>

Return to previous menu.

1. Lightning (Blitz Chess) Clock Settings

Lightning is traditionally (since Howard Staunton introduced chess Clocks
and standard pieces, at least) a guillotine game. The Fischer Clock
affords another version with the same rate of play, but of indefinite
length. The Fischer Clock settings chosen are simple and effective to
this end, as the schedule will show. In this form, endgame technique can
develop, and will not atrophy. Proficiency is a feature of every world
champion since Philidor.

2. Allegro (Fast Chess) Clock Settings

This is a somewhat slower version of Lightning, and perhaps a first step in
acquiring that fantastic speed of play. Allegro is a very popular and
appropriate speed for light hearted games, and the perfect compromise.
The above two Categories are ideal for training purposes as well as for
fun. The techniques vary beyond time alone, and may not transfer. These
should therefore have specific Rating Categories. See also the MAIN MENU
option 9. Settings <Enter> for suggested conditions.

3. Forty Move (Average Endgame) Clock Settings

This option helps provide rational Clock settings when we are considering
the time we wish to spend on the middle game. Fischer's concept is
difficult to understand in the practical application, and thus easily
misapplied. The calculator provided will supply the Clock settings, and a
schedule of total times available for every ten moves up to a hundred.
Such schedules could, and possibly should, be supplied with every rated
game or tournament.

4. Sixty Move (Average Decision) Clock Settings

This option helps provide rational Clock settings when we are considering
the time we wish to spend to likely conclusion of the game. Adjournment,
or adjudication (referring to the Bible: Ruben Fine's 'Basic Chess Endings'
for analogous positions) can be used if time is merciless. The
computations available here are for the endgame specialist, rather than the
gambit artist and middle game tactician or opening strategist.
5. Variable Move (eg Long Games) Clock Settings
If an absolute playing time is a concern, we can choose a likely maximum
number of moves based on previous games, and enter the time and moves.
The Clock settings, and schedule, will be supplied.

6. Hybrid Clock Settings (Clock & Increment Honing)

This option analyses the effect of specific Clock and Increment settings.
The resulting schedules can be illuminating, highlighting the effect that
some combinations may have on time management & playing styles.


Statistics for style settings and endgame/decision lengths are gleaned from
the classic publications listed below. While hardly complete or unbiased,
they should provide a sufficiently representative sample.
208 'The Unknown Capablanca' David Hooper (Batsford 1975)
60 '60 Memorable Games' Bobby Fischer (Faber 1969)
90 'Bobby Fischer - Profile of a Prodigy' Frank Brady (Dover 1965)
500 '500 Master Games of Chess' Dr. Tartakower & J. Du Mont (Dover 1952)