Time and Time Again

It is a question that many chess players have had to repeatedly deal with throughout their chess careers:  How much time should I spend on a move?  Yet, no matter how many answers they are given, not one of them seems to be the one answer that solves everything.  There always seems to be an exception, and more often than not, another instance where the situation occurs again.  I, for one, am no exception to this – I cannot recall the countless instances where I have misjudged what I believe was the correct amount of time to spend on a specific move, whether it was too short or too long.

Now, it very well may be impossible to know exactly how much time should be spent on a move based on the position.  There are just way too many factors that contribute, including position, comfortability with the position, material imbalance, tournament situation, player ratings, and so on.  Perhaps a computer may one day solve the puzzle; however, at the moment, we humans are not able to give a definitive answer to the posed question.  Yet, there exists a select group of people that is definitely more qualified to give one; those people are the game’s grandmasters.  Their high general knowledge of the game has allowed them to reach the level that they currently play at, and this means that they also know a thing or two that the rest of us haven’t come to realize yet.  So, with the help of these grandmasters (and a very helpful website!), I am challenging myself to try to get as close to solving the problem as is humanly possible at the moment.  The purpose of undertaking this challenge is to understand the thought processes of a grandmaster’s mind as he or she plays through the game; hopefully, that will allow me, as well as the other readers on Chess^Summit, to gain a more thorough understanding of the topic at hand.

I plan on accomplishing this task by sifting through the games of a super-tournament (the 2016 Sinquefield Cup, to be exact) and noting down the amount of time that each player spent on each move of each game in each round.  In this case, with nine rounds and five games per round, that will add up to a total of 45 games.  Then, with the ample amount of stats I have learned in past years at school, I hope to run a detailed statistical analysis of the times spent on each move.  Assuming that the times are normally distributed, I hope to fit a normal density function to the data as well as calculating the basic stats (mean, standard deviation, standard error, etc.) on the time spent throughout the games from move 1 to 40.  I chose this interval because the majority of tournaments have a first time control of 40 moves within a certain time frame, whether that be 90 – 120 minutes.  Furthermore, the addition of the second time control after the 40th move would clearly change the dynamic of the game from that point on, and it would be difficult to generalize the findings from the relatively smaller sample size of games that actually last that long.  For example, if a game of the tournament lasted until 70 moves, then data and statistical calculations would still only be present from 1 – 40.  I am quite confident that the means hereinafter will be normally distributed, so that should definitely aid my efforts in this quest.  If time permits, I also hope to delve even deeper by conducting second-order analysis over localized series of moves and seeing if the grandmasters tend to make relatively quick moves following a single long think.

My goal for this challenge is to be able to collect enough data and analyze the results thoroughly enough so that we all can further understand the complicated realm of time and use them as supporting knowledge in our future games.  For example, the ultimate goal is to have a general understanding of how much time to spend for the first 10 moves, moves 10-15, moves 15-20, etc.  Obviously, the game position at a given move could change the numbers found here, but these would act solely as a guiding principle that we should not veer too far off from.  Also, if you do end up spending a relatively long time on a certain move, we should know how quickly to play the subsequent moves to compensate for the lost time.

Since this idea was formulated relatively recently, I did not have much time to prepare and analyze the data before today.  I had only partly collected the data, and though I considered trying to finish before today, I ultimately decided that it was better to wait and finish the job thoroughly than to do a half-hearted job in a short period of time.  Since I am set on trying to achieve this goal, I wanted to take this time to share with you the challenge I am undertaking over the next few weeks and a general summary of what should be expected.  The next two weeks should give me adequate time to fully complete all data analysis, and the hope is that I can share the results that I found in my next article.  But, until next time, goodbye and good luck in your future games!


One thought on “Time and Time Again

  1. Pingback: Time and Time Again (Part 2) – chess^summit

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s