Prepping for South Sound Go Club’s upcoming tournament on Sunday, I find that I’m pretty psyched! It has been a while since we’ve had tournaments in Tacoma, and I’ve been missing it more than I realized. I’m about to make sure the game clocks all have working batteries and are pre-set to the correct time controls, so it seemed like a good time to follow up on an earlier post about tournament time controls.
|AGA Minimum Time Limits|
|Canadian||30 minutes||20 stones in 5 minutes||Good with analogue chess clocks|
|Japanese||30 minutes||1 stone in 20 seconds||Also called “byo-yomi”; more info below|
|Fischer||3 minutes||+15 seconds per move||For each move played, 15 seconds is added to your clock|
|Absolute||45 minutes||None||Also called “Sudden Death”|
So — four different approaches to time limits on AGA rated games, deemed to be the minimum necessary for players to have a chance at performing well. Time limits longer than these also are acceptable for rated games; but games with time limits shorter than these are not appropriate for submission to the ratings system. (“Lightning” games should not be submitted to the ratings system.)
The way Canadian Overtime works is, when your main time ends, you pause the clock and count out 20 stones, and put 5 minutes on the clock. (This assumes you’re using the established AGA minimum for Canadian overtime — 20 stones per 5 minutes. You could use a less severe Canadian overtime, like 20 stones in 6 minutes, or 15 stones in 5 minutes; but rated games should not have a more severe overtime. For example, 20 stones in 3 minutes would not work; and 30 stones in 5 minutes would not work.)
Start the clock again, and play your moves from the pile of 20 you counted out. As long as you play all 20 before your 5 minutes elapses, you can continue (count out another 20, and set your clock to 5 minutes again). But if your 5 minutes ends and you have not finished your pile of 20, then you’ve lost the game.
Canadian overtime works well with mechanical analogue chess clocks — the kind powered by a wind-up spring. Pro: Simple, easy to understand, easy to do; flexible — you can spend more time on some moves and less time on others, as you see fit. Con: Pausing to reset the clock to 5 minutes (and count out the specified number of stones) can feel distracting; and it’s important to play stones from the counted-out pile. If you accidentally take stones from your bowl, and you run out of time with some of your counted-out stones left (that you were supposed to be playing), there’s not much the tournament director can do to help you. This is why we would put the lid on the bowl, when this type of time control was prevalent at Seattle Go Center.
Many electronic clocks have a way of handling Canadian OT that doesn’t require you to pause and count out stones. But those clocks generally also handle Japanese byo-yomi, which Go players seem to prefer for historic / cultural reasons.
I’m told that “byo-yomi” means “seconds reading”, which is why Ing clocks mark the button for setting overtime with “RS” (for “reading seconds”). (But I digress — I’ll talk about specific clocks some other time.) What would happen is, a player would run out of time, and the timekeeper would start counting off the seconds — “reading the seconds” — as the seconds hand would sweep around the clock. I have the impression this was done mostly for Japanese championship matches. (Wikipedia has a good article on byo-yomi.)
Obviously, our American amateur tournaments can’t have so many officials (a referee, a timekeeper, and a recorder marking the moves down on paper) for each game. The cost would be prohibitive. But we have electronic clocks that replace the timekeeper.
With Japanese-style overtime, called “byo-yomi”, you typically you get 5 overtime periods, and each period is usually 15 seconds or more. Seattle Go Center plays with 5 periods of 30 seconds, which clearly is more than the AGA minimum of 1 period of 20 seconds. In the past in Tacoma, we’ve played with 3 periods of 30 seconds, which also clearly complies with the AGA standard; but for South Sound Go Club’s upcoming tournament, we will use 5 periods of 30 seconds.
When your main time period ends and you enter overtime, you have x number of byo-yomi periods. For this example, I’ll assume x = 5, which is pretty common at tournaments. So, you have 5 periods, each of 30 seconds. As long as you make your move before 30 seconds elapses, you keep all 5 of your periods. When it’s your turn again, you’ll have 5 periods of 30 seconds.
But suppose you’re thinking hard, and you don’t move within 30 seconds. Then, you’ve lost one period (but not the game — at least, not yet), and you have 4 periods left, each of 30 seconds.
Play continues like this. If you play your move before your current period expires, then you keep all your periods, and your current period resets. If you don’t play before the current period elapses, then you lose that period. Eventually, you may end up with only one period left; and if that period elapses before you play your move, then you lose the game.
There are a few players who routinely play slowly, and end up on their last byo-yomi period, playing each move just a second or two before they would lose on time. Heavens help the tournament director who pairs two such “byo-yomi artists” against each other, because that game might drag on an hour after everyone else has finished.
Pro: Flexibility; cultural connection to tradition. Con: Tricky to explain to neophytes; requires clocks that can handle this time regime (but those are common nowadays); one or two games might thrash the tournament schedule.
This type of time control is named for American Chess legend Bobby Fischer. It’s also called “bonus” time, or sometimes “increment” time. Your clock starts with a relatively low amount of time — 3 minutes is the minimum for games that go into the AGA ratings system — and a “bonus” is added for each move played.
I don’t have any experience with this time system. I seem to remember that some Go players on the east coast were experimenting with it, some time around 2010, using an initial time of 15 minutes which incremented by 30 seconds per move. My guess is that turned out to be way too much time for most games, since the AGA minimum now is 3 minutes plus a 15 seconds bonus.
Pro: Familiar to Chess players; games tend to finish in a timely fashion, which is good for the tournament directors’ schedules. Con: Somewhat inflexible, because there is less time to think about moves in the opening, which traditionally are regarded as very important. And, as with byo-yomi, you need a clock that can handle this time system; but those are fairly common now.
Absolute Time / Sudden Death
This is the simplest time control. If you run out of time before the other player does, then you lose the game.
Pro: Can be used with old or new game clocks; and games finish on schedule. Con: Psychologically stressful for slow players, which may prevent them from playing their best. Also, when adrenaline is running high in the final minutes, the clocks take a beating.
In my opinion, it’s best to use a 5-second delay with Sudden Death (if you play your move within 5 seconds, you don’t lose any time from your clock), though that does mean you need a newer electronic clock rather than an old mechanical wind-up. It does tend to prolong the useful life of the clock though, because players have less incentive to smash down the plunger as rapidly as possible.
That touches on another type of time control — “delay” or “Bronstein” — that I won’t go into because, though it’s common in the world of Chess, I’ve never seen Go players using it. And it doesn’t have a minimum time setting listed for inclusion in the AGA ratings system.
At a tournament in Oregon, I once saw an interesting variant of Sudden Death, that I guess you could call “Sudden Coma”. If you ran out of time, you didn’t automatically lose the game; but you couldn’t make any further moves — but your opponent could play moves(as long as he or she still had time on the clock). In theory, if you had secured your groups sufficiently well, you might still come out ahead when counting the score, even though your opponent could make moves which you couldn’t answer. In that particular tournament, the situation didn’t arise — all games finished with time still on the clock.
Thinking about it, years later, it’s probably best to use Chinese scoring rather than Japanese (which is a topic for another post; or read about it at Senseis’ Library) if you use Sudden Coma. But, I’m not sure what the tournament director would have done if both players ran out of time before finishing play. On reflection, I can’t recommend “sudden coma” as a time control — too much potential for mayhem.
Those are the kinds of time control you’re likely to encounter at an amateur Go tournament in America nowadays: Absolute (Sudden Death) — simple and effective; Fischer (Bonus / Incremental) — shiny and new; Byo-Yomi (Japanese) — traditional and popular; and Canadian — tried and true. Each has its proponents and detractors; and any of them can be chosen by tournament organizers, as long as the clock settings meet or exceed the minimums for rated games.
Hoping to see you across a goban soon, –Mike M.