Russell Sherwood

Russell Sherwood  Sunday, July 30, 2017

The Similarity test was created a few years ago to be able to compare engines, mainly to be able to attempt to detect cloning (The process of taking someone else’s engine, making a few minor changes and renaming it as your own, without any acknowledgement of the original. There were many deep, nasty, arguments of the type “this engine is a copy of that engine”. These do not concern us here!)

An important area of generating winning chances is developing moves which are not the First choice or Principal Variation of our main engine. A number of analysis methods do exist which can do this, such as Subtractive or Monte Carlos Analysis but a starting point before all of this is to consider which Engines tend to give “Original” moves.

A note of caution here – “Original” moves are not necessarily “Good” moves but often the suggestion of a different (weaker) engine can help us find the best path forwards from a position. In some cases simply showing the move to an engine will help it find a better line. In other cases much more work is required.

So for the trial to determine originality I consulted the FastGM rating list  http://www.fastgm.de/ on its slowest time control and ran all the engines I had from this list (plus some which I consider to be the best older ones: Gandalf, Junior, Hiarcs…….)

The results are interesting – As you would expect most of the Stockfish variants have a very high score………………………..

So what we are looking at is a heat map – it works by looking down the left-hand column to find the engine we are interested in – we then read across – the similarity in move choice is shown, the higher the score the more similar the moves selected. In general it is better to ignore the numbers and look at the colours – The more red the more similar the choices, the more green the less similar.  The Engines are ordered in a rough ordering of strength – with the ones on the left/at the top being stronger and the ones on the right/bottom being weaker.  The difference between the top and bottom engines is between 300 and 500 elo, depending on which rating list you believe! There were also a few engines which failed to run in the trials – Fizbo and Giraffe. Of these Giraffe is the more interesting as its evaluation is based on a very different method, however the project is currently shelved!

 

We can divide the map into four basic zones:

 

Top Left (The Red Zone)

The strongest engines – many different variants of Stockfish, so it is unsurprising that these engines are fairly similar in outlook. In general engines in this area make poor bedfellows in terms of bringing a different point of view to the table.

 

The Middle (Central Red Zone)

Here we have a lot of engines with similar outlooks. Without going into much detail, a number of these were/are accused of being copies of other engines.

 

The Bottom Corner (Green Zone along the Right and Bottom sides)

These engines tend to come up with different moves from the Top Engines – the downside often being that they’re simply weak rather than original moves and care should be taken in any suggestions.

 

The Interesting Zone (Green-Yellow along the top and down the left hand side)

These are the engines which tend to give a different slant of a position compared to the Top Engines. Particularly interesting are Houdini 4, Boot and Deep Shredder 13 and Andscacs. These engines are not the strongest but are not weak either. Houdini 4 is the oddity of the group as it is “obsolete”, whereas the others are all current.

Anyway If you want a different opinion to HockmodoFish, these engines represent probably the current best mix of originality and playing strength.

 

Download

Russell Sherwood  Saturday, July 8, 2017

This should be read whilst humming the theme tune to the “Great Escape”!

You know the feeling. The game is looking lost, your opponent has already notched it up as a win and the Engine score is massively against you.

You could just give up but there are a few things you can try here…………………….

It’s important to remember that you will probably still lose but some of these techniques can give a glimmer of hope!

The first thing to not do is employ Dead Man’s Defense or any of its variations. DMD’s is basically to slow play the position, in the hope your opponent, well, defaults  (Hand’s up at the back who thought I was going to say Dies!) This method is against the spirit of Correspondence Chess and also against the rules!

So what can you do?  Is the position actually lost/losing, should be your question now! Is the position actually lost – there are numerous examples of situations where the engine evaluation is widely inaccurate – showing a drawn position as lost or a lost position as drawn. This comes down your chess knowledge. A technique which can help you here is “Line Extension”. Let the engine run a while, the copy the line into your notation, move to the end of the line and analyze again.

Does the evaluation change? Following this start to work your way back up the line , you may well find improvements that change the evaluation significantly.  Personally I have held positions where I was greater than -3 in terms of engine evaluation.

So let’s assume the position is against you. Our main try now is to attempt to randomize the position as much as possible. A few methods we can try:

 

1. Don’t play the Engine’s Principal variation (the main move/line it suggests). Run in Multiple line mode and look for alternative lines. The reason for doing this is to attempt to move away from the lines and type of position the engine is going towards. Here we are dealing with a basic problem in the way engine evaluation works. We have all seen the games where the evaluation is -0.1 when becomes -0.3 then -0.7 and so on – as the position goes down a slippery slope. In the same position there could have been an alternative line which would have been evaluation at a constant -0.6, which would be probably bypassed as its much worse than the first move.
2. Look for strategic needs of the position. What are the key motifs? For example ideas like opposite bishops ,the wrong colored bishop or control of a key diagonal or square. If you can see these in the position you may be able to nudge the game toward in that direction – which the engine won’t see until it is too late.
3. Try and build a Fortress - this is a large topic area in itself but a basic introduction can be found here https://en.wikipedia.org/wiki/Fortress_(chess)
4. Keep playing! We are looking at giving the opponent the maximum number of opportunities to go wrong!
5. Use psychological weapons – for example the use of non-standard moves (1) , combined with Conditional moves will increase both your opponent’s workload and he may think you have seen something s(he) has not.
6. Use multiple engines (especially ones outside of Stockfish, Komodo and Houdini) for ideas to support (1) and (2)
7. Smile – this approach is far more fun than either resigning or slow playing the position, especially the times you get the result.

 

It’s also worth mentioning that the use of these methods will improve your analysis skills in general, which is never a bad thing!

 

Russell Sherwood  Saturday, July 1, 2017

In a past post I suggested a method to estimate the number of games a player could concurrently manage.

As we move towards the start of a new season I offer a different, complementary approach to aid the player in planning their season.

So why is Season Planning important? In simple terms, it allows us to make sensible choices about which events to participate in and which to not!

So how do we undertake this? Much depends on if you are an established or new player. For the Established player the steps are:

1. List all the events you took part in the last (current) season. ICCF “My Events” is a good place to get this information from, but don’t forget those events which were not on the server (Postal, email…..)
2. For each of the events you need the number of games, the level (Category or Average rating)
3. Add to the list any events you did not enter this season but would like to consider for next season.
4. Add the Start date for each event.
5. Go to ICCF Ratings and calculate your average number of games completed per rating period. This approach will give you an estimate of how many games you can play per season.
6. Go back to the Events list you have created and add an “importance” rating for each one. This is a very personal measurement based on a combination of factors of why an event is important to you. 1 is the most important, 3 is the Least.
7. Order the list with “1’s” at the top and “3’s” at the bottom.
8. Go through the list, allocating your Total games to events, on a reducing balance until all are allocated.
9. At this point you can see which events you have capacity to play in and which you do not. In addition you can look at periods of the season which will be busy (Typically October and March), which may affect your choice of entry.

 

For new players the process is very similar , although step 3 takes much more relevance. If you are unsure of which events are around, details of repeating events are held at http://www.welshccf.org.uk/ under Play – Tournament Calendar

Russell Sherwood  Saturday, July 1, 2017

In a past post I suggested a method to estimate the number of games a player could concurrently manage.

As we move towards the start of a new season I offer a different, complementary approach to aid the player in planning their season.

So why is Season Planning important? In simple terms, it allows us to make sensible choices about which events to participate in and which to not!

So how do we undertake this? Much depends on if you are an established or new player. For the Established player the steps are:

1. List all the events you took part in the last (current) season. ICCF “My Events” is a good place to get this information from, but don’t forget those events which were not on the server (Postal, email…..)

2. For each of the events you need the number of games, the level (Category or Average rating)

3. Add to the list any events you did not enter this season but would like to consider for next season.

4. Add the Start date for each event.

5. Go to ICCF Ratings and calculate your average number of games completed per rating period. This approach will give you an estimate of how many games you can play per season.

6. Go back to the Events list you have created and add an “importance” rating for each one. This is a very personal measurement based on a combination of factors of why an event is important to you. 1 is the most important, 3 is the Least.

7. Order the list with “1’s” at the top and “3’s” at the bottom.

8. Go through the list, allocating your Total games to events, on a reducing balance until all are allocated.

9. At this point you can see which events you have capacity to play in and which you do not. In addition you can look at periods of the season which will be busy (Typically October and March), which may affect your choice of entry.

 

For new players the process is very similar, although step 3 takes much more relevance. If you are unsure of which events are around, details of repeating events are held at http://www.welshccf.org.uk/ under Play – Tournament Calendar

Russell Sherwood  Saturday, July 1, 2017

In a past post I suggested a method to estimate the number of games a player could concurrently manage.

As we move towards the start of a new season I offer a different, complementary approach to aid the player in planning their season.

So why is Season Planning important? In simple terms, it allows us to make sensible choices about which events to participate in and which to not!

So how do we undertake this? Much depends on if you are an established or new player. For the Established player the steps are:

1. List all the events you took part in the last (current) season. ICCF “My Events” is a good place to get this information from, but don’t forget those events which were not on the server (Postal, email…..)

2. For each of the events you need the number of games, the level (Category or Average rating)

3. Add to the list any events you did not enter this season but would like to consider for next season.

4. Add the Start date for each event.

5. Go to ICCF Ratings and calculate your average number of games completed per rating period. This approach will give you an estimate of how many games you can play per season.

6. Go back to the Events list you have created and add an “importance” rating for each one. This is a very personal measurement based on a combination of factors of why an event is important to you. 1 is the most important, 3 is the Least.

7. Order the list with “1’s” at the top and “3’s” at the bottom.

8. Go through the list, allocating your Total games to events, on a reducing balance until all are allocated.

9. At this point you can see which events you have capacity to play in and which you do not. In addition you can look at periods of the season which will be busy (Typically October and March), which may affect your choice of entry.

 

For new players the process is very similar, although step 3 takes much more relevance. If you are unsure of which events are around, details of repeating events are held at http://www.welshccf.org.uk/ under Play – Tournament Calendar

Russell Sherwood  Wednesday, June 21, 2017

In the real world, I spend an awful lot of time optimising and improving processes, something I am rather good at!

Where I can, I try and transfer some of the techniques to my CC play, often with success. The methodology I will now offer, I had very mixed views when first exposed to, mainly as it’s so “UnBritish” but it does work!

Ask a typical British person to list their (a) strengths and (b) weaknesses; the typical response will have the far more in list (b) than (a). To have more in (a) is seen as “Boastful and Bragging”, something negative. This in itself is something that is unhelpful.

So the technique I am looking at here is the of a “Bragging Board”. The technique is used to real world to record any successes, however small, by an individual or team. This is very useful for three main reasons:

• A rich source of information for compilation of reports

• A way to demonstrate progress, especially when success is hard to come by, which is a very useful positive psychological benefit.

• A method to manage progress.

So how can the player use this method for CC?

Practically I believe a simple approach can be as follows:

• Record your test best & worst wins

• Record your 10 best & worst draws

• Record your 10 best & worst losses

• Record any Tournaments where you:

  • Perform better than expected

  • Place 1^st,2^nd or 3^rd

  • Do anything that noteworthy (in your opinion)

• Anything else you find noteworthy

It is worth mentioning that “Best and Worst” is your definition.  It could be by rating or by Title held or any other method you think worthy.

The reason for the “best” results is obvious, but the “worst” needs a explanation. The draws and losses here are painful but very informative and keeping them in mind is something the player can continue to learn from.

Once you have written up this document, keep it somewhere safe (and probably private!) and maintain it. Once you have done this, anytime you need a positive boost in Chess terms have a read through it. This is very useful in the situations discussed in “Form is temporary, Class is Permanent”.

On a personal note, being British, I can confirm the first sentence was rather difficult to write!

Russell Sherwood  Tuesday, June 20, 2017

Form in Chess is a nebulous thing, doubly so in Correspondence Chess! In OTB changes in a player’s form are easy to spot – a few games lost in a row, mediocre performance in tournaments and lost rating points all are good indicators.

 

CC is different, due to the timescales involved the indicators that can be used for OTB all indicate a loss of form in CC typically a year ago!

 

So how can the CC player detect a loss in form in a timescale that enables them to be able to take measures to “limit the damage”?

 

One technique I have utilised which may be of help is to:

• Go through my portfolio of running games
• For each one note:
  • My opponents rating
  • My estimate of the likely outcome of the game. I used to use the standard three outcomes but now use additional two scores of 0.25 and 0.75, along with 0, 0.5 and 1. The two additional scores represent worse but not losing (yet!) and better but not winning yet.
  • Any games still within a “book opening” phase are counted as 0.5
  • I then use a very simple rating calculation, for each game, of:
    • 0 : Opponents rating – 200
    • 0.25 : Opponents rating-100
    • 0.5: Opponents rating
    • 0.75: Opponents rating +100
    • 1.0 Opponents rating +200
• I then calculate an average of all these ratings to give a basic performance score.
• These performance scores I calculate once a month and any significant downward movement tends to indicate an issue of which a loss of form could be one!

 

So, if you discover a loss of form what can you do with about it? My experience tends to indicate the following are often the underlying causes

 

• Too Many Games/Too little time
  • I provided a method of calculating the optimal number of games but in general, most players will know when they are playing too many games. The solution is simple but can take time to implement of being more selective in the events the player chooses to enter.
• Unsuitable openings being used for the quality of opposition
  • Would/Should you the same openings against 2000 rated opposition as you would against 2400, probably not! Generally, whatever the common opinion of an opening is in OTB it is generally worse in CC (So OK in OTB translates to dubious in CC, dubious translates into unplayable!!)
• A lack of a system being used to select moves/openings/plans
  • If the player does not utilise a method, whatever that method is, they are more likely to suffer dips in form. If they have a system, the dip can still occur but the system can then be improved to reduce in impact.
• New openings being deployed without preparation
  • Here the obvious solution is to prepare new lines before playing them in CC
• Old opening lines reaching their “sell by date”
  • The is especially true for a certain type of CC player – “The opening stats nerd”. This breed of player will follow lines based solely on its % performance. This approach is vulnerable in several ways but this player can suffer many reverses as refutations are found but with the line still showing good statistics.
• Newly obsolete software or hardware being utilised.
  • This issue can happen as both hardware or software can become obsolete almost overnight. For example, if a player were relying on a dual core machine it would become slowly less competitive but then at some point, it would “fall off a cliff” in terms of competitiveness. They key thing here is that whilst a player may not wish to get involved in an “arms race”, they should look to keep their set up are current as possible at minimum cost. For example, in terms of engine Stockfish 8 is free and cutting edge, yet I still see people using older versions.
• The player reaching their competence ceiling
  • This is not really a loss of form but it is possible for a player to reach a point where they are not strong enough to compete with their opposition, yet are too strong for the players below them in the rating list. This situation is analogous to Football teams which yo-yo between the Premiership and Championship in the English Premier League. Here the player needs to work on their game to attempt to become competitive.

 

In summary, if you think you have lost form, Don’t Panic! take a deep breath and play your way out of it! If you can talk to a chess friend or a stronger player (or both!) who may be able to help you diagnose any problems!