Russell Sherwood

Covering all the (Data) bases!

Russell Sherwood  Saturday, May 20, 2017

A regular question amongst CC players is: “Which Database should I use?”

The simple answer to this is “none” as no single database will meet all of your needs!

So which combination of Databases should we use?

This question is more relevant and I believe the Aspiring player needs a number of Databases to meet different needs at different times.

There are a few things to consider:

  • CC and OTB games are different beasts, except at the Highest level of OTB play
  • CC games should be separated into Pre and Post 2012. This is the point at which engine strength increased dramatically and a number of older opening lines became almost unplayable in CC
  • Opening Novelties from SuperGM’s should be considered of being a similar source as CC Games as the resources behind these players would indicate the same (or better) preparation levels as CC.


So onto the Databases!


The first database, I think necessary is one containing every game regardless of source. Megabase is probably the best known commercial base with 8m games, however, here are other ones floating around the internet which include engine games which are much, much larger – the biggest now above 20million games.

Many of the games in these kind of bases are of poor quality but the purpose of this database is for Idea generation.

“Super GM base”

Top Level GM’s (say 2700+) spend an lot of time on preparation, utilising engine checking of ideas. For this reason novelties can be mined from these databases. A note of caution, whilst the initial idea will have been checked in depth, the subsequent ongoing game will not have been.

“CC Base”

This is a database of all games played under Correspondence Chess time control – hence days rather than hours and minutes. The main purpose of this database is to determine what variations are current and any trends that may be developing.

“CC Strong base”

Opinions differ a little on this but this is a database of mid ranked CC players (say 2200-2400) games played post 2012 or so (the point at which engines changed the nature of the game). The purpose of this database is for generating ideas. Many of the players in the zone are the ones with ratings climbing rapidly who are worthy of close examination.

“”CC Superstrong base”

The purpose of this database is to determine what is currently playable at the highest level of CC. Most players will look for CC games above 2400-2450 played post 2012. This is the CC equivalent of Super GM Base but with the added bonus the games having been engine checked.

“Recent Games Base”

The purpose of this database is to keep the player in touch with new novelties and the latest lines. Many will not be suitable for CC but it is a very valuable source!

“Engine base”

These used to be a treasure trove of ideas but these days there are many millions of engine games played. These can be utilised to determine which direction the game may take if your opponent is simply an “Engine Jockey”

So how do we utilise these databases? That’s a discussion for another day but I think enough clues are given in the description!

Peidiwch รข gadael i'r Bastardiaid malu chi i lawr!

Russell Sherwood  Tuesday, May 16, 2017

Over the year’s I have played thousands of games against hundreds of opponents and 99%+ have been reasonable, nice, interesting people.

Many have become friends and with some, I have enjoyed varied and interesting conversions, despite (or probably because of) our varied backgrounds. This leaves a tiny percentage of rather unpleasant people who you occasionally cross paths with. I can add to this small list with similar characters when I have been acting various administrative roles.

I can only stress these people are very rare but it is unpleasant when you come across one.

So what to do? We can broadly classify the incidents into  two groups:

The first group come straight at you with offensive content – there is no likelihood of a misunderstanding, their intent is to upset and provoke a response. In many cases, you could suspect that they would benefit from professional help.

With these cases, the main rule is: do not engage with them, that generally what they are seeking. Instead, complain to the Tournament Director, They will review the message history and assuming they agree with you, either warn your opponent and/or set silent mode  Should your opponent offend again they will follow procedure, escalating the disciplinary measures, ultimately resulting in possible suspension from all games.


The second group tend to develop from conversations that get out of hand or where the meaning has been lost in translation, so a message not intended to be offensive turns out to not to be intended to be.

My advice to avoid these kinds of issues is to (a) avoid subjects such as Religion and Politics and (b) Give the other person the benefit of the doubt. I recall as a young Supply Chain Manager than a number of my staff got annoyed with their opposite numbers always using the word demand rather than ask. A little bit of investigation showed that the French word for Ask is Demander..........

I have had many interesting conversations (including ones on Religion and Politics) but only with people I knew fairly well and both of us were aware of the robustness of our relationship but unless you know your opponent well avoid this path.

In these cases, the first action should be to confirm if it is a genuine problem – or has been lost in translation. If it’s an issue, complain to the TD, otherwise either continue to enjoy your conversation!

Don’t EVER suffer from an obnoxious opponent!

Standing on the Shoulders of Giants

This article is available to federation members only; please log in to view the content.

11th European Team Championships Semi-Final - A Deeper Look

This article is available to federation members only; please log in to view the content.

Time of the Day!

Russell Sherwood  Saturday, May 13, 2017

A very interesting article was published with regard to the quality of moves made at various times of the day.

Article Link

Whilst there are a few question marks over the research its outline findings are interesting:

People tend to make better moves in the morning

Now this research was for rapid play server play rather than CC, so is it valid for our purposes?

That is something we cannot confirm but perhaps the question should be "Do you know when you make your best moves?"

Data was extracted from the ICCF server last year related to when the most moves were made (it was related to a debate over the so-called "Free day") and what this showed was a peak first thing in the morning, another at lunch, then again in early evening. To me, this suggests that there is no overall bias for CC in terms of move quality but that the player should consider what factors help the make better moves or

What gets them into the Zone?

So what could some of the factors be?

  • Music or the lack of
  • (Light) Alcohol or the lack of
  • Time of the Day
  • Temperature

So next time you find yourself Analysing well - consider what the "factors" were and try to replicate them!



Jumping through Hoops!

Russell Sherwood  Wednesday, May 10, 2017


From discussion with a few players recently I thought it worth clarifying a few ratings thresholds that a CC player needs to reach to be able to progress…….


Getting a Rating

If you don’t hold a FIDE rating then the first objective is to gain an ICCF provisional rating. To obtain this 12 games need to be completed. It is worth noting that until you reach 30 completed games this rating can (and will) jump around quite a bit (hence it being classed as “unfixed”). Prior to obtaining a provisional rating only 5 types of tournament are open to the player:

  • International Opens – These typically have an average rating of between 2000 and 2100
  • Champions League (Team Event) – Typical team averages in the bottom Division are 2200+
  • National Events – can vary but typically below 2000
  • Promotional Open Class (Both ICCF and Zonal) – Generally below 1900


At this point ICCF Promotional – Higher Class events (1900-2100, average 2000) become available


No new events become available but the player can now be considered (and can lobby!) for inclusion in Title Norm Events for the CCE Title


At this point the ICCF Master Class events become available (Average 2200) and the player can be considered for CCM Title Norm Events


This is a minimum rating applied for WCCF for possible selection/nomination in a number of events


This is a major milestone! At the point the player is now able to:

  • Enter Master Norm Events directly,
  • Be considered for Nomination to the British Correspondence Chess Championship ,
  • Be competitive in terms of selection for Major International Team Events
  • Be considered for lower IM Category Title Tournament Invitations


At this point the player can be consider for some of the medium Category Invitational Events.


At the point the player can be considered for Higher Category Invitations we receive and may enter Major Tournaments either at a later stage (such as the WCCC). In addition Grandmaster Norm events are now available


Hopefully this helps on a regular question!

Simple Guide to Annotation

Russell Sherwood  Friday, May 5, 2017

Annotated games are the lifeblood of any Chess magazine or website and a very valuable source of improvement for the player, yet only a small number of players actually annotate games.

To help improve this I considered if there are any basic ideas which can be utilised to help the aspiring annotator and surprisingly there are!


What makes this game interesting?

If the game is not interesting to you then this will come across in your annotation and will probably not be interesting to other people! So what could make the game interesting:

  • The Event
  • The Opponent
  • The Opening
  1. Any instructive characteristics in the game
  2. A Novelty
  • A Story behind the game
  • The importance of the game (did it give a Norm?)
  • And much more!


Tell a Story

In an extension to “What makes this game interesting?” we need our annotations to tell a story. With a couple of mouse clicks Chessbase can create a deep engine based set of annotations but without a Human “Angle”, the annotations are likely not to be interesting. This can be seen in good annotations having a lot of text description.

Consider your Audience

Who is your (target) Audience? Knowing this will aid in a number of ways:

  • What to include and what to omit
  • The level of explanation you put in (tactical themes, positional ideas, lines not taken, the length of analysis, the balance between text and variations…….


Consider writing annotation with your Opponent

The games are very rare but are often the most instructive: Games annotated with your opponent. For CC players these games can be especially interesting as the insights gained from two sides of the board can be vital in terms of improvement.


Don’t be a Megalomaniac!

A common issue seen in annotated games is players who try to give the impression that they had a complete understanding through the game. This can be boring for the reader and does not tend to read well. Of course if you are a 2700+ rated Super GM your understanding will be rather more complete that a 1600!


Draw’s and Losses can be an interesting as Wins!

Although emotionally much tougher to annotate, draws and especially losses can make very interesting games to annotate, especially as any advice on what was learned may be of interest to other players.

Be Respectful of your opponent

As a general rule, you should not deign your opponent in annotations, regardless of any issues during a CC game. There is a type of “roast” annotation where a number of Humorous remarks are made at a player's expense (typically from a third party annotator) but these are generally best left to those who are naturally humorous and know the players!

What was the Novelty?

(Almost) All games have a novelty (New Move). This is an important move to provide some in-depth commentary on. Were you aware of current theory or was this your cunning plan?

Was a move unexpected?

Choosing which moves to annotate can be difficult. A good, simple way to select them is to ask the questions.

  • Did I expect my opponents move?
  • Would the average reader expect the opponents move?


If the answer to either of these questions is yes then it is a good move to put some commentary on!

Quote Relevant Sources

Consider the level and relevance of sources – it can be very interesting to know the best players to have gone down a line but (a) make sure the number of references is relevant to the audience as the source (For example CC games can be a more interesting and relevant reference)

Don’t worry about making mistakes!

Any Analysis in your annotations will have errors! Don’t worry about this – firstly have an engine check anything you feel unsure about and then should errors be found be happy that someone has gone to the trouble of (a) reading your annotations and (b) taking the time to disprove them, in the process you have both learned something!


This is very important. Offer your annotations up to websites and magazines. Editors are always happy for new content. Do check the rules of the event in terms of public dissemination before you do though!

More Miles to the Gallon?

Russell Sherwood  Monday, May 1, 2017



I've had a nagging doubt about how the same engine performs in different Graphic User Interfaces (GUI) for quite some time, decided to do some research.

The method utilised is a small number of positions were tested and the stable nps (nodes per second) per engine/GUI combination recorded. A general average per engine was then calculated and then the performance transformed into an average performance.

Engine/GUI Fritz 15 CB14 AQ2017 Arena Hiarcs
AsmFish 89% 79% 102% 121% 109%
K10.4 96% 81% 99% 118% 105%
H5 104% 79% 91% 123% 102%
Stockfish 103% 99% 100% 104% 94%
Fritz 15 114% 98% 82% 95% 111%

So some tentative thoughts

  • The difference between the best and worst is around 30-40% in most cases. This is a lot of insight into a position that could be missed
  • Fritz15 seems to perform better than Chessbase
  • My Hardware/Software combination could be skewing the results
  • Different Stockfish versions perform differently in different GUI

So my suggestion to even the semi-serious CC player is to run your own tests and consider if you are using the right Engine/GUI combination(s) to support your analysis!


Can you have too much of a good thing?!

Russell Sherwood  Sunday, April 23, 2017

A common question from aspiring CC players is “How many games should I play at once?”. Apart from the obvious and sarcastic answer of “One!” this is a very interesting question.


With a little thought, it is obvious that playing too many games will have a significant negative effect in terms of quality. What might not be so obvious is that playing too few games can also have a negative effect – skills take time and practice to develop and the player also takes a time to focus on a game.


So this leads us to a conclusion that there is an optimal number of games. Talk to a number of stronger players and the answer will tend to be somewhere between 20 and 30, although it is rare these players can explain why!


So let's try and quantify this. When we look at it, our available time determines the number of games we can play. The time is in two areas – our time and computer time.


There are a number of variables to consider;


(1) The rate of play in the events we are going to take part in, typically between 10 moves in 40 and 10 moves in 50 days.


(2) The minutes per day we will devote to CC. (This does not need to be daily, so 6 hours a week would be 6/7 hours per day)


(3) How long we believe we need to review and analyse each move.


(4) How many PC minutes we are willing to utilise per day.


(5) How long we believe we should be using per game to analyse each move. Note 30 minutes here would not to be to leave an engine running on IA for 30 minutes – it could be looking at 15 positions for 2 minutes each.


When we combine these number we come up with two metrics – Human Capacity and Machine Capacity.


Human Capacity = (Average minute per day(2) *Average rate of play in days(1))/Time to Analyse each move(3)


Machine Capaity = (PC Minutes per day(4) * *Average rate of play in days(1))/Time to Analyse each move(5)


Our capacity is whichever of these two numbers is lower.


As an Example:


I look at my games and they are split evenly between 10/40 and 10/50 – my average rate of play in days is 4.5 (40+50)/2/10


I say I will spend 60 minutes per day on CC and I think it will take me 15 minutes to review each game.

I'm willing to let my PC stay on for 4 hours and think it will take 30 minutes per game to analyse.


In this example


HC = (60*4.5)/15 = 18


MC = (240*4.5)/30 = 36


So, in this case, my capacity would be around 18 games. If I wanted to increase my capacity I either need to spend more time on CC (2 Hours per day) or reduce my analysis time (15 minutes to 7.5 minutes) or a combination of both.


Now in practice, you can probably add around 20% to this number because during the game there will be phases when moves are made quickly which allows a little time to build up on your clock!

Lost in the Endgame

Russell Sherwood  Thursday, April 20, 2017

Is there a Good, Better, Best Engine in the Endgame?

When it comes to Engines the 2nd most common question asked is “Which Engine is the best in the X” - The “X” being the Opening, Middlegame or Endgame.

It's generally held that Engines are still relatively weak in the Endgame and this is demonstrated in a number of games where the evaluation is significantly incorrect (typically showing Win/Loss for a draw).

This raises four interesting points for the CC Player:

  1. Which kind of positions does the Engine typically misevaluate?

  2. Which Engines tend to be better in the endgame?

  3. Do any combinations of Engines improve the overall analysis?

  4. Can I leverage any of this knowledge in my own games?



The most common errors tend to be positions, which when stripped down bare tend to be about Zugzwang or Fortresses. Miss-evaluation also tends to happen as engines tend to use generic methods which will generate “good” moves but not the absolute best move. In endgame analysis, the engine often relies on search depth finding the win rather than knowledge but often the required winning move is beyond its horizon.



My first Analysis was to use a very large epd file with 432 various endgame positions – some simple, some very difficult. The outcome of this analysis was showed two interesting outcomes:

(1) That even the best engines were only scoring just over 80% or more importantly were getting around 20% of positions incorrect. That is not to say that these moves were blunders, many were 2nd best but not optimal.

(2) Here there was very little difference with the Komodo (10.4) and Stockfish (in the form of Matefinder) being out I front, with Houdini 5 just behind – we then see a chasing pack with Deep Shredder, Fritz 15 and older versions of Komodo and Stockfish. Looking at some historical analysis

We can see that it appears Komodo, in the most recent versions, has made up ground on Stockfish and Houdini

What is interesting is that each version of the Big 3 Tends to advance the score by 1-2%, this being due to improving overall general evaluation picking up a few extra “correct” moves, rather than any additional knowledge being added to the engine.

This knowledge is what tends to make a difference in Endgame analysis – even a fairly weak human player will know about Bishops of Opposite colour being generally drawish but an engine which is a pawn ahead that does not have this built in, will trade down evaluating the game as won!

That knowledge is the key factor here is supported by running the tests again with significantly increased times – as expected the scores only increased slightly.


So I then moved onto a smaller 100 position test set with a view to looking at coverage.

So out of a score of 100 I obtained….

85 CorChess 1.2

84 Matefinder

82 Raubfisch ME262

81 Sugar Xpro 1.0

80 AsmFish

80 Houdini 1.5

79 Houdini 5

78 Komodo 10.4

74 DS 13

71 Fritz 15

71 Gull 3

68 Andscacs

65 Hiarcs 14

59 Hakkapeliitta

57 Critter 1.6

57 Chiron 3

41 Fire 4


I did run other, older engines through the test but in general the scores were at the bottom of the table I also ran other Stockfish Clones through the test suite but these simply crowded out the top of the table.

So from this, we see that even the best modern engines are missing around 15% of these positions.

Some might be surprised that Houdini 1.5 does so well. To a certain extent this is due to the way that Chess Engines are developed. What is more important these days is developing engines which beat other engines and sit atop of the rating list, this aligned with the self-test method utilised means that on occasion, to gain 50 wins, 1 loss is sacrificed and the “Baby goes out with the bathwater”.

Does using a range of engines improve accuracy?

Even from this small-scale test, the answer appears to be a clear Yes!.

The best single score is 85/100, but with a combination of Engines, we reach 99/100!


So Looking at a few combinations….

CorChess 1.2/Houdini 5/Komodo 10.4 we get 92/100


For 99/100 we need(!!)

MateFinder/Houdini 5/Fritz15/Sugar/DS13/Gull3 and Hakkapeliitta!!!!! Rather impractical


A rather more interesting choice is:

CorChess (or another Stockfish Clone) with Support from Houdini 1.5 and Hakkapeliitta with a score of 96/100


Can I leverage any of this knowledge in my own games?

The most simple takeaways from this brief analysis are:

(1) Engines do have gaps in their knowledge, which needs to be taken into account both in analysis and CC play.

(2) These knowledge gaps are not universal and using a combination of engines to analyse a position is likely to bring us closer to the “right” move.

(3) Not all older engines are obsolete and some do give new insights into a position.

(4) Even with the use of a combination of engines it still requires the Player to act as arbiter and when doing this s(he) should consider that the highest rated engine is not always correct.


This brings us to the end of this brief review of Engines in the Endgame. I include a spreadsheet with engine results should the reader be interested in different combinations.


Welsh Correspondence Chess FederationBritish Correspondence Chess AssociationSchemingMind Internet Correspondence Chess ClubSocial Correspondence Chess AssociationNational Correspondence Chess ClubWelsh Chess UnionInternational Correspondence Chess Association