When playing the game of go online, a common time control setting is for players to have X (e.g., 10, 20, or 30) seconds per move, with the clock simply resetting after each move.

(There are variations to this byo-yomi theme, such as first having Y minutes of main time, and only entering the per move time controls once that main time has run out. Another common thing is to have a set of (Z) per move timers. For example "10 (5)" would indicate that you only lose on time after failing 5 times to play a move within 10 seconds.)

Advantages to this system are:

  • Easy to mentally keep track of how much time can be spent thinking on a move.
  • Games are less likely to be "ruined" by either of both players entering a "time crunch".
  • Reduces the annoyance of players letting their clock run out.

There are perhaps drawbacks too though that didn't occur to me yet. So with that in mind...

What are the reasons that per move time controls are seen as relatively unattractive for the game of chess?

BTW, I've of course noticed the relative popularity of Fischer style time increments in Chess, but personally I find these much harder to keep track of. Often I find myself racking up way more time on the clock than intended, turning what was intended as a fast game into a rather slow affair. Or with an increment of 3s or less the reverse is true. The resulting volatility makes me lose on time way more often than I'd care to admit.

  • 2
    The main reason is it takes away a players ability to allot different amounts of time to different positions. If 4 out of 5 moves are easy moves that I can decide on quickly, but 1 out of 5 is a difficult position that requires some thought, then I want to be able to spend much of my time on that 1 move and make up for it by making the other 4 moves quickly.
    – koedem
    Dec 30, 2021 at 5:37
  • @koedem Sure, but you can apply a similar strategy with a per move timer: don't play immediately after you've decided your move, but use the remainder of your move's time to think about upcoming moves.
    – Will
    Dec 30, 2021 at 5:41
  • 1
    Yes you can do that, but then it depends on what your opponent plays. Why not just make the move and then think that on your opponents time? Deliberately not moving when you already know what your move will be seems like an arbitrary hindrance to me.
    – koedem
    Dec 30, 2021 at 5:49
  • 1
    To give a specific example of a time usage pattern: what often happens is that some position holds a deep strategical decision for one player, who will spend a lot of time, then after they move the opponent in turn spends a lot of time considering the implications of that strategic decision and how to counter it. If you want to allow that in a fixed time per move setting you would need very long times per move, defeating the purpose. Or you have too short time per moves and then your quality of play will be much lower since you never have time for longer big picture decisions.
    – koedem
    Dec 30, 2021 at 5:53
  • @koedem It's not a hindrance at all: it's what hundreds of professional go players do worldwide. It might be hard to reason about intuitively without experience playing with both systems though. And really, it "depends on what your opponent plays" either way: with increments you gamble just as much that your oppent plays as you expect. How else would you evaluate whether upcoming moves are likely to require more or less thinking time?
    – Will
    Dec 30, 2021 at 5:58