Move Theory
By Tom Braunlich   
September 16, 2008
Tiebreakslide.jpg Analyzing The New Modern Time Scrambles - and How to Protect Yourself Against the Madness

Moving pieces from one square to another is so basic to the concept of chess that players are often called woodpushers. Yet strangely when it comes to tournament play, the Laws of Chess have always been alarmingly unclear about how to actually do it!

Shortly after the invention of the chess clock there was the invention of the time scramble... and suddenly the previous calm and orderly process of taking turns moving the pieces broke down as speed of movement became paramount.

Recent videos of wild time scrambles in major playoff games highlight a rule problem that has essentially been left unresolved - the rules do not clearly discuss the "turn sequence," with the result that no one knows for sure when you are allowed to start your turn. In normal play it doesn't matter, but in time scrambles it is critical and can affect the outcome. Another recent example is this video from the ongoing Women's World Championship.

·        Can you start your move before the opponent has pressed his clock?

·        What exactly is the proper way to move, anyway?

·        What should I do to "defend myself" in modern mega time scrambles?

·        What can organizers and TDs do to mitigate time pressure madness?

·        What do the top tournament directors think about this?

All this is covered here - information all tournament chess players should know.

The New Time Scrambles

Muddled time scrambles are not a new phenomenon, and they do not relate only to 5-minute "blitz" games. Any game played without a delay or add-on on the clock (as all games were before the "Fischer Clock" became popular) can come down to a serious scramble where the broken movement rules become painfully obvious. In the old days, time scrambles were also common, but they tended to be brief because the time controls were not "sudden death" and only required you to make a certain number of moves in time, such as 40 in 2 hours. Often a player only had to make a few quick moves to reach the time control, so the scramble rarely got out of hand. (Even so, players like Walter Browne were infamous for leaving themselves just a minute for 15 or more moves!) The Fischer Clocks were originally conceived to help mitigate scrambles at the end of such time controls.

However, in recent years the situation has changed. First, the advent of strong computer programs required sudden-death time controls so that the game could be played in one session (without adjournment).  Second, attempts to broaden the appeal of the game have also pushed organizers toward quicker play.

Thus we have seen the advent of sudden death time controls, and they are surely here to stay. A major effect of this has been to stretch out the "scramble time" from a few moves to dozens, creating time pressure situations that are more common and last much longer than before. Add to that the recent ability to see such scrambles in slow motion on a YouTube video, and this dormant problem has now become obvious to everyone.

This unaddressed problem in the rules can no longer be tolerated. I was a professional game designer for over 20 years, working with all the major American game companies, and my specialty was writing rules. (See my book, The Art of Game and Toy Design for a discussion of it if you can find a copy; it is out of print.) Rules are tricky because one must write them not only like a lawyer (rigorously eliminating loopholes and covering all possibilities), but also must explain them with clarity and common language so that they can be easily learned and understood.

Unfortunately, the Laws of Chess for tournament play are disturbingly unclear when it comes to move rules. Perhaps this is not surprising, since they are complex and created by committee, and need to evolve as the sport changes. One would expect minor problems with constant refinement and updating needed. But even so, there are critical ambiguities in one of the fundamental aspects of the game - moving the pieces.

Mixed Terminology
The first serious problem underlying the current "movement rules" confusion is mixed terminology. The term "move" is used in many different ways: as a verb and as several different kinds of nouns. Look at all the different meanings for "move" we have in chess:
·        Pieces move and players move.

·        A "move" can mean:

o       "the transfer of a piece from one square to another...", or

o       the technical part of a turn when a player does that, or

o       the entirety of a players turn.

·        A player can be "on move" (i.e., when his clock is running).

·        A "move" can also mean what the computer programmers call "two ply" (a pair of moves from the player and the opponent, as in the fact that a "40 move game" actually has 80 moves made in it).

·        A player under the current rules can "make a move" without yet his move being "made" (i.e., his "move" (turn) "completed" by punching the clock).

·        There are also definitions for "touch move", "determining a move", etc.

There really should be more distinct terminology used for these different concepts. (See conclusions, below).

Incompleteness

Also, the current movement rules have a glaring hole in them. They do not discuss the issue of "turn sequence" explicitly. Turn sequence is an essential concept in any game and needs to be rigorous or you have chaos. The chess rules discuss fairly well (although annoyingly scattered about in various places) what you can do on your turn, but not what the opponent can do. Being "on move" does not mean that the opponent can do nothing until it is his turn. There are several things he can do (such as call your flag), but they are not enumerated. This is critical because one of the most important questions in the chess rules - whether or not the opponent can begin his move before you have finished yours by pressing the clock - is left unaddressed.

Normally common sense prevails during chess play. The players take turns making their moves with no problems. But time scrambles bring out the importance of these technicalities.

Before we get to the key issue - turn sequence - let's quickly review what the rules say about what you can do on a turn to get a mental "big picture" of the move theory. 

MoveChartcolor.jpg

I hope I didn't leave anything out! But I wouldn't be surprised if I did; it's complicated. Keep in mind that there are some further rules extending even all this, such as some particulars about how to make a castling move (move the king first!), how to correctly promote a pawn, etc.

Turn Sequence - The Big Unresolved Rule Issue

In time scrambles players tend to "overlap" their move onto the opponent's move, in order to stop their clock as fast as possible after it was started and keep the time from running down. It is not unusual for skilled players to rattle off four or five moves in less than a second this way, especially if they can get into a good rhythm. The actual moves being made become unimportant. It's all about the clock. Many players move the piece that is closest to the clock, back and forth, to diminish hand travel time. Sometimes it can get very exciting, but unfortunately it soon devolves into a mess as pieces get knocked over and roll ridiculously. Players continue making instant moves, often illegally. It would be hilarious if there were not so much at stake for players and fans.

Watching the videos in slow motion you can see what is happening. As Player A moves a piece and reaches for the clock Player B is already starting his next move by moving his piece, etc. Player A does the same thing to B, and the game is now about manual dexterity. Although this kind of scramble normally occurs when there is sudden death and no clock increment, it can also happen if the increment is short enough for the players to feel the time pressure.

It is important to realize here that this mess is essentially a natural consequence of the sudden death time controls, the existing rules, and the situation the players are in - not something nefarious and underhanded the players are doing to cheat. The circumstances force them to play as fast as they can. Essentially here the players are taking their turns at the same time, overlapping the move sequence very closely.

The rules do not address whether or not you can do this. The only rule that seems to apply is rule 6.8:

Rule 6.8.a (FIDE)

"During the game each player, having made his move on the chessboard, shall stop his own clock and start his opponent`s clock. A player must always be allowed to stop his clock. His move is not considered to have been completed until he has done so, unless the move that was made ends the game. (See Articles 5.1, and 5.2)
The time between making the move on the chessboard and stopping his own clock and starting his opponent`s clock is regarded as part of the time allotted to the player."

This is, surprisingly, the only rule that explicitly discusses turn sequence, and all it specifically says is that the moves have to be "completed" in sequence (by pressing the clock), that you cannot interfere with the opponent finishing his move by blocking his access to the clock or by making your move so fast that he never gets the chance to press his clock at all before you have already pressed yours in reply.

The Arguments

The two sides in this rules interpretation argument look at Rule 6.8.a differently. Those who believe it is legal to "overlap" moves often use the above rule to support their position, because the rule wouldn't seem to be necessary otherwise. For example, Geurt Gijssen (one of the most respected FIDE arbiters in the world), said in a recent ChessCafe.com article:

"In my opinion, this part of Article 6.8 only makes sense if a player makes a move before the opponent has pressed his clock. It means that even when a player is not on move, he is allowed to press the clock in the given situation."

But this is very debatable. Those opposed to overlapping use the last sentence, (that the time after moving before the clock is pressed is "part of the time allotted to the player,") to assert that it is logical and common sense to assume that the opponent can't move during the player's time.

Similarly, anti-overlappers have stated that the concept of being "on move" (USCF Rule 6B, similar to FIDE Rule 1.1) seems to imply that you can only move when you are "on move" (i.e., your clock is running). It says, "A player is said to be on moveor to have the move when the opponent's move has been completed." So how can you legally begin your move before you are "on move?"

What the Experts Think

These aren't the only arguments. To help understand this issue I polled about a dozen top arbiters and TDs (FIDE and USCF) to see how they would interpret this issue. They included such well-known directors as Bill Goichberg, Frank Berry, John Hillary, Ernie Schlich, and Mike Atkins, as well as Stewart Reuben of England, (chairman of the FIDE Organizer's Committee and author of The Chess Organizer's handbook). Each of them has decades of top-level tournament direction experience.

I started off the email thread discussion among these experts by framing the two sides of the argument as follows:

Geurt Gijssen, Mike Atkins, and others have stated that it is not illegal to start you move before the opponent has pressed the clock, as long as you don't interfere with the opponent pressing the clock at the end of their move. The rules don't directly discuss the issue. The gist of Gijssen's argument was that the specific rule about not interfering with the opponent pressing the clock (FIDE Rule 6.8) would not be necessary (or make sense) if you had to wait until the opponent hit the clock before even touching a piece; and therefore the rules don't intend that interpretation. ...

On the other hand, others have stated that the concept of being "On Move" (USCF Rule 6B, similar to FIDE Rule 1.1) seems to imply that you can only move when you are "on move"... So how can you legally begin your move before you are "on move?"

Here, in heavily edited form that I've tried to arrange to give a sense of the discussion, are the responses of these experts:

Bill Goichberg:

"I agree with Gijssen. It's clear that you can begin your move before your opponent presses the clock, because this is commonly done and permitted, and would be very difficult to outlaw.  If a player neglects to press the clock and the opponent moves and releases the piece, I don't think many TDs would allow the opponent to then make a different move.

However, there are other issues.  If a player moves and has not yet pressed the clock, while the opponent may move, the opponent may not do other things that require being on move.  For example, claiming a draw by triple repetition requires being on move, and a player who has released a piece but not pressed the clock might then decide to make such a claim.  In that case if the claim is upheld, the opponent's "move" would not occur, because he is not yet on move.  In other words, you can determine a move while not on move, but you cannot complete a move while not on move."

John Hillary:

"I continue to believe that moving a piece before one's clock has started is illegal. Only the "Player" may move a piece, the "Opponent" does not become the "Player" until the move is completed, and the move is not completed until the clock is stopped. (It's "determined" but not "completed.") I agree that this could be tricky to enforce in time pressure, but so are a lot of things. That doesn't excuse ignoring them."

Frank Berry:


"Nowhere in the rulebook is this question addressed. ... [If I had to make a ruling on it I'd have to] hand the player the rulebook and ask them to show me what rule is being violated."

Mike Atkins:


"I think if you get too technical with words and too legalistic in interpretation here, you run the risk of making the game unplayable in some situations. For example, in the classical situation in which a player has moved and forgotten to push his clock, by one of the interpretations here it would be ILLEGAL for me to move because I am still "the opponent" and not the player, when in reality I AM the player and someone simply FORGOT to push his clock. Would it also be illegal for me to disturb or distract the player and tell him he forgot to push his clock just so I could move??  I had a player in a tournament a few years ago who was using this whole scenario to play mind games with his opponents. He would make a move, not push his clock, and when the opponent would eventually begin to move he would say "nah nah nah, it is still my move, you can't move yet..." When he did this a second time he stopped when I told him he would be forfeited for doing it again. But, see the problem if you get too legalistic?"

Ernie Schlich:


"I believe that it is not proper to move before your clock is started. I was taught early on (by my wife Joan) that the player on move - the entire time his clock is running - owns the board. The opponent does not adjust pieces, touch the clock, speak except for a problem, or otherwise disturb the opponent. In addition to the previous concept, moving before one's clock is started is unethical ... Saying that it is proper as the clock is punched does not deal with the gain in time and also leads to the game losing its serious aspect."

Stewart Reuben:


"I think most arbiters (including me) agree that requiring a player not to move before the opponent has pressed the clock would be a considerable change in common practice, especially with blitz and very hard to enforce. ..."

"Rule 6.8a states inter alia that aplayer must always be allowed to stop his clock. Thus the fact that the opponent has made his subsequent move makes no difference. It does not matter as the player was still entitled to press his clock. Indeed with cumulative or delay mode this is essential. Otherwise the player is deprived of legitimate thinking time. ..."

"Rule 4.6 of the Laws explains when a move is made. The clock plays no part in this. The opponent is then free to move. When the clock comes in, in rule 6.8, it refers to the move being . There is an intended distinction between made move and completed move. Thus the player makes the moves on the board, but does not press the clock. He can lose on time even though everybody knows and there is video evidence that he had made his move in good time."

John Hillary:


"Only the ‘player' may legally move a piece. (1.1: ‘A player is said to `have the move` when his opponent`s move has been made.'). One does not become the "player" until the opponent's move has been ‘made.' The question is, at what point has the move been ‘made,' and is this the same thing as ‘completed' (4.6: "The move is considered to have been made when all the relevant requirements of Article 3 have been fulfilled.")?"

"If you consider only 1.1 and 4.6, it is possible to argue that the move has been ‘made' when the player's hand releases the piece on a square. However, that part of the Laws of Chess does not consider the use of the clock. That's covered in Article 6. The phrase ‘... His move is not considered to have been completed until he has done so...' states that the move is not completed until the player has stopped his clock and started his opponent's. Reading this to mean that the ‘opponent' (who is not yet the "player") may make his move on the board as soon as the piece has been released seems like a real stretch. It is certainly possible to argue that this reading is undesirable (or unenforceable) in blitz/rapid games, but now you are talking about what the rules should be, not what they are."

What are the rest of us to make of this discussion? Perhaps we should evaluate it as "unclear": .

There was also much discussion among these experts about other related issues, such as, "If there is a time scramble where illegal moves are being made, then should the arbiter step in?" There was disagreement about that too!

Conclusions

  Recommendations for Rule Committee Members:
USCFRulebook.jpg
The only thing clear is that the issue is confusing, harmful to the game, and in dire need of some clarification. Both the FIDE and USCF rules committees need to take this aspect of the game under consideration and try to address it in the next rules update (which I understand could not happen until late next year at the earliest). That's plenty of time to discuss it - and they will need it because it is sure to be contentious and controversial. They could start by trying to clean up the terminology ambiguities discussed above. (I'd suggest utilizing the term "turn", for example.)

Until the revamping of the rules is complete, the preface of the rulebook should be kept in mind, which begins, "The Laws of Chess cannot cover all possible situations that may arise during a game, nor can they regulate all administrative questions."  

However, I have a different suggestion.

The rule writers should consider the possibility that there is no workable way to satisfactorily resolve this issue with rules alone. Overlapping move turns is probably an inherent aspect of tournament chess play when the players are in a time scramble, and trying to outlaw it with rules will never really work. You are dreaming if you think that even an arbiter watching every time scramble will solve the problem. Are we really going to have an arbiter decide whether or not a GM jumped the clock by a 10th of a second with his move? There is an inherent problem with the way chess works relative to chess clocks, at least in over-the-board play.

Instead, perhaps the rules should acknowledge this and attempt to attack the problem from another direction. Overlapping move turns only become an issue in time scrambles. Therefore, mitigating time scrambles will make the move issue moot. I thus propose the rule committees should make a requirement for all major tournaments (and playoffs) the use of a digital clock with a sufficiently long delay or add-on function per move.

Furthermore, I (and Bill Goichberg and others) strongly believe that the function used by these clocks should be a "delay" and not an "add-on". Delay is the function that best works to inhibit time scramble craziness. With delay, you are encouraged to move slowly and use up your full delay time for thinking before you move (otherwise that time is lost), thus encouraging an orderly turn sequence. With add-on, however, you are still encouraged to bang out your move as quickly as possible, in order to accumulate time to use later.

Add-on is preferred by many top players, as they like the ability to accumulate time. However, logically this is a dubious thing to allow in tournaments. Time management should be an aspect of tournament chess. Allowing a player to confidently let his time dwindle down to one second because he knows he can later build it back up with a lot of repeated moves and pointless maneuvering hardly seems in the best interests of the sport. No other sport allows that kind of abuse of its clock timing mechanism.

Bill Goichberg is a strong supporter of using delay only, with massive experience in large tournaments like his World Open informing that opinion. He says,

"Five seconds delay has worked well and is enough to avoid the massive time scrambles we used to see with pieces and clocks flying and other players disturbed by all the commotion.  It's odd that shortly after 5-second delay was introduced in the US, FIDE began using add-on increment without ever even trying what was working well here (perhaps this was because at some FIDE meeting there was no US representative)."

Stewart Reuben's response to that (from a European perspective) was that 5-second delay was perhaps "too fast for wrinklies" and suggested a minimum for the purpose might be 10 seconds (which Goichberg thought would be fine), or the FIDE time control using 30 seconds which is growing more popular. At our large national events in Oklahoma this year that I organized (the FKB U.S. Championship, the FKB U.S. Championship Qualifier, and the Okie Chess Festival) we used the G/90 + 30 system. The feedback I got about the long increment was generally very positive, as it really smoothes time pressure anxieties. People still feel time pressure and make blunders - but there is no wild display of illegal moves, etc. We were allowing both "delay" and "add-on" for the increment. I don't see any problem with going to delay exclusively.

One aspect of a long delay like 30 seconds is that players can still be required (and indeed are required) to record their moves (and can do so easily), and therefore disputes are further minimized because you can prove threefold repetitions and 50-move claims, etc. Thus fewer game scores are "lost" to posterity due to unrecorded time scrambles. This is good for the sport when spectators aren't cheated from seeing the end of an interesting game.

Use in tournament games of a mandatory delay of at least minimum duration (whether it is 5, 10, or 30 seconds; whatever the authorities decide) would also help mitigate other bizarre features of the modern chess rules, such as the awkward and unnatural "insufficient losing chances" claims - a rule that might easily be improved or eliminated in this case.

Note that for Blitz tournaments, Mike Atkins (who I'd say is perhaps the most experienced blitz director in the world) adamantly opposes increments for pure blitz tournaments.  The time element is considered part of the game there, as are the resulting scrambles, and there are additional rules relating to them. However, note that for regular tournament playoff games that come down to rapid or blitz or blitz/armageddon play, I did recommend in a previous article that they include a delay on the clock of at least 3 seconds to help prevent time scramble madness in such important games.

Recommendations for Organizers

Until there is new guidance on this issue from USCF/FIDE, I would recommend all organizers help themselves by specifying large delay-only time controls such as G/2+10  or G/90+30. For some tournaments this will cause initial problems due to having insufficient numbers of delay-capable clocks available. But such problems exist already and there are several well-known ways one could deal with that situation until analog clocks get essentially phased out for use in important events.

How to Defend Yourself - Practical Advice for Players

There is a lot you can do as a player to shield yourself from exposure to exasperating time scrambles. First, own a good digital clock, know how to set it, and bring it to the tournament. If you must use an analog clock and get into a messy time scramble as a result, you have no one to blame but yourself. I think it is a bit nutty to try to play a modern sudden death time control using an old analog clock. It's asking for trouble. Also, ask the T.D. before play starts what his policy is on time scramble issues.

Defensive Play - Rule Options You Can Use

Most players do not take advantage of the rule options available to them during a time scramble. You do not have to be at the mercy of the clock! There are things you can do to defend yourself, or even to take advantage of the situation.

Use Rule 11d to Punish Illegal Moves

Know how to stop (pause) every clock you play with, and be prepared to do it! If your opponent makes an illegal move in a time scramble, shout "Illegal!!," stop the clock, and call the director. According to rule 11d, if the director upholds your claim you will get two minutes added to your clock!

One of the most common illegal moves in time pressure is a violation of rule 11c:  "If, during the course of a move, a player inadvertently knocks over one or more pieces, that player must not press the clock until the position has been reestablished...." So again, if this happens take advantage of it: stop the clock and call the TD immediately and you may get a 2-minute compensation that could turn the tables for you. (Conversely, don't give your opponent this advantage by knocking over pieces yourself!)

Another very important time pressure rule to know is USCF rule 11d1:

"A director should not call attention to illegal moves in sudden death time pressure. If either player has less than five minutes remaining in a sudden death time control and the illegal move is not corrected: (a) before the opponent of the player who made the illegal move completes two moves, or (b) before either player resigns, or (c) before either player is checkmated with a legal move, or (d) before either player is stalemated with a legal move - then the illegal move stands and there is no time adjustment if the game is still in progress."

This means you have two moves to call "Illegal!" and get your two minute bonus. Otherwise you forfeit that ability if you keep playing. This is important because it may be that you are moving so fast that your brain didn't have time to register the opponent's illegal move before you punched your clock in reply - but you can still make the claim if you do it before you complete your second following move.

(The above rule 11d1 is USCF. It's my understanding that this is somewhat different for FIDE. The FIDE rules largely assume that there are more arbiters available at an event than we normally have at a weekend swiss in the U.S., and gives them more latitude.)

Players should also defend themselves with a good understanding of the "Insufficient Losing Chances" rules (14H), which can save you if you are in a time scramble with an analog clock. (You might be able to get a clock with delay to be put on your game, for instance.) These rules are complex and TDs tend to administer them haphazardly, and I believe there are also lots of changes in the works for them that are yet unpublished.

Similarly, know the "Insufficient Material" rules (14D and 14E), but beware that there are major differences between the USCF and FIDE approach to insufficient mating material claims that can trip you up - just ask Josh Friedel.

This brings up another point. It is probably a good idea to ask the director at the start of the tournament (or for him to announce) whether the event is going to use the USCF or FIDE rules on these issues. It's my understanding that American tournaments can use USCF rules even if the event is FIDE-rated (and indeed it is assumed to use USCF unless stated otherwise). But it wouldn't hurt to clarify that. Similarly, if you are involved in a blitz playoff game you should ask beforehand whether it will be played under normal tournament rules (which I'd recommend) or under USCF "blitz rules."

The 50-Move Rule is also important to know for a practical player in time pressure. Many of the worst time scrambles feature one player trying to win with, for instance, K+R vs K+N, a theoretical draw which can go on forever if not for the 50-move rule. Rule 14F4 also allows a player in "time pressure" to "... stop both clocks, declare to a director an intention to invoke the 50-move rule when possible, and ask for assistance in counting moves. ..."  When you do that, there are various options listed for the director to choose from, including assigning a deputy to count the moves or putting a clock with a move counter on the game, etc. I suppose he can even refuse to do anything. There are lots of aspects to this rule, so be familiar with it as well.

An experienced tournament director could probably point out some pitfalls when using the above "defensive" rule strategies, or some other rules that a player might be able to take advantage of in time trouble situations. But these I've listed here seem to be the basic tactics available. Time to get out that rulebook and relearn the rules. (USCF and FIDE!)

Or if you want to avoid all this entirely, don't get low on time!

Tom Braunlich is the 2008 USCF Organizer of the Year. He also wrote
Playoff Theory, an in-depth discussion on the best way to break ties.