ELECTRONIC T-NOTES


CHESSBASE USA'S WEEKLY ON-LINE NEWSLETTER


FOR THE WEEK OF DECEMBER 30, 2001


UNDERSTANDING TIME CONTROLS

by Steve Lopez

As I was mentally preparing an article on Fritz7's online chessplaying functions for this week's ETN, the realization hit me that I get a lot of e-mail and phone questions concerning time controls in chess. Those of us who are experienced tournament players often forget that we were once beginners and didn't understand chess clocks and the various settings used to control the players' time usage. So, although this week's article may seem elementary to some of us, it's worth a review before we delve into the world of online chess in a future ETN issue.

Prior to the mid-1800's, chessplayers were allowed to use all the time in the world to plan their next move. When it was your turn to make a move, you could use all the analysis time you wanted before actually moving a piece. While various means (such as an hourglass) were sometimes used to time a player's move, such devices were the exception rather than the rule. In fact, it was not unheard of for a player to take an hour or more before moving -- there have been a few notorious cases in which a player took seven or eight hours to think before making a move!

The introduction of the chess clock changed all of this. A chess clock is actually two clocks connected internally by a bar or lever mechanism:


Pictured is Fritz7's display of a typical analog chess clock. Missing from the display is a small hanging "flag" which is usually connected to the clock's face at the 11 o'clock position (which we'll discuss in a few minutes).

Operation of the clock is simple. Your clock runs when it's your turn to move. After you've made a move, you press the button on top of your clock. This moves a lever which stops your clock and starts your opponent's clock. His clock runs until he makes a move and hits his button, which then restarts your clock. Thus your clock shows only the time you've spent considering your moves, not the time spent by your opponent -- that's displayed on his clock. The chess clock was a really handy invention and tournament/match play was considerably shortened by its introduction.

Initially, time controls were pretty long -- players in the 1924 New York tournament were required to make just 15 moves an hour. As the years passed, time controls got shorter. By the 1990's, the "standard" tournament time control at the top levels of chess was 40 moves in 2 hours.

The way this works in practice is simple: you have to complete your 40th move by the time two hours are up. If you don't, you lose the game, no matter what the position on the board is like.

Typically, though, there was a secondary time control added on. Your primary time control was 40 moves in two hours, followed by something like 20 moves per hour afterwards. This is actually pretty simple to understand, but is really hard to explain properly. Typically, you'd set your clocks to read 4 o'clock at the game's start. You had until the clock displayed 6 o'clock to make your 40th move (the two hours given for the primary time control). A little flag is attached to the clock at about the 11 o'clock spot on the dial (as described earlier). When you had five minutes of time left, the minute hand of the clock would start to raise the flag. The flag was just long enough to cover the area between 11 and 12 on the clock's face. When the minute hand reached 12, the flag fell back to the hanging position and you were out of time. You had to make that 40th move and hit your clock before the flag fell.

If a player made his 40th move before the flag fell, he'd "made the time control" and the secondary time control was added on. For example, in an event in which the primary time control was 40 moves in 2 hours and the secondary control was 20 moves in two hours, the players just (mentally) added the secondary time onto the clock. In this example, let's say that the White player made his 40th move with 20 minutes left on his clock. He now gets an extra hour added on to his available time and, since the secondary control states that he must make 20 more moves in an hour, he now has until the clock reads 7 o'clock to play his 60th move. So he now has the extra hour plus whatever time he had left over in the first time control -- ergo he now has 1 hour 20 minutes to make another 20 moves.

Let's look at how to set these kinds of time controls (i.e. moves per hour) in Fritz7:


The dialogue displayed to the right is the "long game" dialogue in Fritz. In the Game menu, go to "Levels" and then select "Long game" from the submenu that appears to the side. You'll see this dialogue appear on your screen.

Fritz7 allows you to set three time controls. In the graphic (the defaults in the program), the primary time control is 40 moves in 2 hours. Assuming the player makes this time control, he now gets the secondary time control (i.e. another hour to make it to move 60). If he makes it to that control, he then has to finish the game in another 30 minutes. Remember that when you make the time control, any time you have left is added on to the time given for the next time control.

This is a handy way for limiting the time for a game. In this example, each player has a maximum of 3.5 hours of time, so a game can last a total of seven hours. To set a game of (theoretically) unlimited duration, set the third time control for zeros in all the fields. Then each player has 2 hours to make his first 40 moves and 20 moves every hour afterwards, until the game ends by checkmate or draw.

Sometimes you come across a really oddball time control. For example, I used to play in Baltimore tournaments that had the unique time control of 40 moves in 80 minutes followed by "sudden death" in 30 minues. To set this in Fritz, you'd leave the secondary time control set at zeros -- so you'd set the first time control for 40 moves in 1 hour 20 minutes, set the second time control for zeros across the board, and set the third time control for 30 minutes.

But many players these days find these time controls to be impossibly slow. In the late 1980's, "sudden death" time controls became increasingly popular. This type of time control gives both players a set amount of time to make all of their moves. If your flag falls, you lose.

You set these kinds of time controls in Fritz by going to the Game menu, selecting "Levels" and then choosing "Blitz game" from the submenu:


Note that in Fritz7,"Blitz game" isn't a reference to five-minute games exclusively. The "Blitz game" dialogue allows you to set up any kind of sudden death time controls (including Fischer increments, which we'll discuss later).

Sudden death controls are frequently referred to as "Game in x" controls, with x being the number of minutes allotted to each player. In the late 1980's, when I first joined the USCF, a popular time control was Game in 60 -- each player was given 60 minutes at the game's start. That's the time control I've set in the graphic to the left. When you play with this time control, you have an hour to finish the game. It doesn't matter whether the game lasts ten moves or a hundred; sixty minutes is all you get.

Today there are a wide range of popular sudden death controls. Game in 30 minutes became really popular for rated tournaments in the 1990's. The guys in my chess gang are fond of ten to fifteen minute time controls. A very popular form of chess is blitz, typically 5 minutes for the whole game. And in the last 3 or 4 years, bullet games have become extremely popular in online games -- usually one or two minutes allotted for the whole game.

While not exact, a few definitions should help you get familiar with the terminology:

OK, so far so good -- that's not been too difficult, has it? But now we come to the time setting that drives many people batty: Fischer time controls, also called increment time controls. I hate to self-plagiarize, but I explained this form pretty well back in the January 9, 2000 issue of ETN, so please allow me to repeat that explanation verbatim:

The idea is that for every move you make, a set amount of time is added to your clock. This is not a new idea -- players of the Asian games Go and Shogi have been doing this for many years. The idea behind an added increment is to ensure that (in theory) a player should never lose a game by running out of time.

Let's look at an example. My preferred time setting for computer games and Internet chess games is 20/10. This means each player starts with twenty minutes on his clock and receives a bonus of ten seconds each time he makes a move. Let's assume I'm playing Black. My clock reads 20:00 at the start of the game. My opponent plays 1.e4. I play 1...c6. It takes me two seconds to physically make the move. Normally, my clock would now read 19:58, to reflect the two seconds it took me to move the c-pawn. But by using the Fischer increment of ten seconds, my clock now reads 20:08. I made the move, taking two seconds, so my clock reads 19:58 (for a fraction of a second), but then ten seconds are added to my remaining 19:58 once my move is completed, making my total remaining time 20:08.

Then it takes me, say, three seconds to make my next move, dropping my time down to 20:05. But ten seconds then get added to my time once my move is completed, bringing my new total to 20:15. As long as I make a move in under ten seconds, I actually gain time on my clock. This will become crucial to me later in the middlegame and endgame. If I can play the obvious moves (opening "book" moves, recaptures, or forced moves) quickly, it buys me some time for later in the game when I really need it.

Let's jump ahead to the early middlegame. I have 16:27 showing on my clock when my opponent makes his move. My clock starts and I have a tough decision to make. It takes me an entire minute to decide on a reply, dropping my time down to 15:27. I make my move, and ten seconds get added to my clock, making my time 15:37. In effect, I just spent 50 seconds on my move instead of a whole minute.

So, in theory, as long as a player moves quickly enough, he need never run out of time in a game using the Fischer increment. I can (proudly?) state that I have never lost a Fischer-increment game by my time running out (I have lost a few by being mated in under twenty moves, but let's not talk about that). And I've frequently played against computer programs which finished a game with more time than the initial allotment (I once played a long game at 30/30 against a program which finished with over 45 minutes on its clock).

Hopefully, that nails it. It's a bit complicated at first, but if you remember that the time you spent thinking is deducted and then the pre-arranged increment is then added on, it's not too hard to grasp.

It took a few years (nine, as of this writing) for it to happen, but the Fischer increment style of time control has more or less become the de facto standard in online play. Face to face tournament players have been slow to embrace this form, but that too is changing. The big hangups to its widespread acceptance in over the board tournament play have been:

But once you get the hang of how the Fischer increment works, it's actually a lot of fun to use. Here's how you set it up in Fritz7:


The illustration to the right shows the Blitz game dialogue set up for the 20/10 increment I talked about in the italicized example above. "Time (min)" refers to the main time control. "Gain per move (sec)" is the number of seconds you get back whenever you make a move and whack the clock. It's as simple as that.

As a short aside, the "human bonus" fields are for players who think that they get robbed because they have to move a piece via mouse while a computer program can just fire off its moves instantaneously. The "min" field allows you to set an extra number of minutes that the human player receives as compensation for time lost via mouse manipulation. For example, in a straight 5 minute blitz game, setting this for "1" gives the human player an extra minute on his clock at the game's start (for a total of six minutes instead of five). The "sec" field works exactly like a Fischer increment; the human player gets that number of seconds added to his clock whenever he completes a move.

Although this article has shown how to set time controls in Fritz7, the main purpose has been to show newer players (or those who are new to timed games) how the time controls actually work. You'll need to know this information if you plan to play online chess on the Playchess server (or any other Interrant chess server, for that matter) or even if you plan to play in face to face chess tournaments. But there are a lot of other neat things you can do with time controls. For example, time controls can be used to offset a stronger player's natural advantage. Back when my kid nephew was first learning chess, we'd play blitz games with time odds. He'd start with five minutes while I'd have two or three. That really put the heat on me to think and move quickly, and he won more than a few games because my flag fell. We had a lot of fun playing this way because he loved having a fighting chance to win despite my experience, and I (at the time) enjoyed the adrenaline rush of constant time pressure (I'm older now and a 1-1 Cup final with the referee looking at his watch three minutes into injury time is about all the stress I can handle anymore).

Back in the days when I was hustling chess in pubs, I'd sometimes hear potential opponents complain that I had an "unfair" advantage due to my experience. So I'd offer them time odds as a handicapping device. I'd take half the time that they started with -- but there was a catch. Since we typically played for a buck a game, I'd give them a choice: we could play with the same amount of time on the clock with a dollar at stake or I could give them double the time at double the monetary risk. I'd take, say, ten minutes to their twenty. If I lose, I pony up the buck. But if they lose, they'd owe me two dollars, since they had a 2-1 time advantage at the game's start. Either way, I didn't miss many meals because of cash shortage.

And Fischer time controls are great for players who constantly find themselves in time trouble, since you can accumulate extra time in the opening or whenever a move is forced. You can rack up an extra couple of minutes playing a recognized opening line and then put the added time to good use later when you're trying to figure out how to win a Queen vs. Rook ending -- aaaaaaaaaaaaaaaaaaagh! You can even set different added time increments for players of differing strength: the weaker player gets a ten second increment while the stronger player gets just three to five added seconds per move.

So have fun playing around with time controls. Learn 'em, live 'em, love 'em, and sometime in the next few issues we'll look at how to use them on the Playchess server. And for more information on time controls in Fritz, see the ETN issues for January 9th and January 16th, 2000.

Until next week, have fun (and may your flag never fall)!

HAPPY NEW YEAR!

You can e-mail me with your comments, suggestions, and analysis for Electronic T-Notes.