OGS team league proposal

My personal preference on clock is long main time and byo-yomi, something like 60 minutes+3-5x1minute.

  1. 30 seconds byo-yomi is really scary to me, I typically just use the last 10 second to place stone.
  2. my own experience is also, this is also commented by pro, long main time is just to ease pressure, and does not necessarily cause the long game times. The reason is once we enter byo-yomi, we tend to use all of the time allocated even when we have decided the next move. we are taking time to think others. This is also true that I tend not to resign in byo-yomi cause i don’t have time to count to confirm i am hopeless. I have been playing quite some tournament games this year with 45-60 minutes main time, most of the time, I used 30-35 minutes before the games ended.

However, I’ve been play live games only this year and am ok with any time settings. Just want to share some of my thoughts.

No canadian byo-yomi please, that seems to be very confusing. I played only one game and thought I was right on time, then lost the game by time out. Only then I realized I was supposed to make 10 moves instead of 1 within that time frame. lol

3 Likes

I am wondering if H = min(ΔR - 3, 9) (for ΔR≥ 3) would make matches more interesting. I am thinking in particular about the 8k vs. 21k matches.

1 Like

I don’t think I’ll well versed enough in mathematics to understand this expression.

Could you translate it to layman for me please?

1 Like

If difference of ranks = 0, 1, 2 or 3 : no handicap.
If difference of ranks = 4, handicap 1 stone.
If difference of ranks = 5, handicap 2 stones.

If difference of ranks = 12, handicap 9 stones.
If difference of ranks = 13 or more, handicap 9 stones.

2 Likes

My intuitive feeling is that a difference of four ranks deserves two stones.

A one-stone handicap is just a komiless game, after all.

I don’t think I’d have a good chance of beating a shodan, for instance, at just no-komi.

1 Like

My very limited preference is increase/ decrease komi instead of extra stones on the goban.

But of course I don’t insist, whatever is decided is OK.

4 Likes

Another good idea!

I’m pretty sure Samraku prefers that, since he is the founder of the Lagrange reverse komi tournament.

I didn’t even consider it.

3 Likes

I’ve never played with more than 7.5 reverse komi so I don’t know what is best.

Isn’t stone better understood/played than reverse komi as black? At least from the beginning, you feel the advantage?

When I play 0.5 komi on either side, I don’t know any difference and play no differently. Interestingly I was watching a game between a top Chinese amateur player and a 7P with .5 komi. The amateur controlled the game from the beginning. The 7P commented if they play even games, he would win 90% of the time.

For me, it’s stones that I have to take into consideration without having actually having thought out and placed them. It feels more complicated, even though it looks like an advantage.
With “score handicap” I play as usual and I just have extra points for help. I still lose, just by less :blush:.

2 Likes

What would be fair reverse komi?

Something like three points reverse komi / rank difference?

eg. drifterwolf would take 12 points from terrific, you’d take 18 from Arclaw, etc.

reverse komi = 3ΔR should be about the same as handicap stones = ⌊ΔR / 2⌋.

1 Like
3 Likes

If you want to maintain an advantage for the stronger player, then you would need less than 2 * komi per rank. Don’t forget to add komi (7 in NZD) back in for White at the end regardless of ranks, as well.

2 Likes

First the OGS rating to rank conversion formula from the new ranking update is used to convert the player’s overall OGS rating to a rank, which is then multiplied by 14 (subject to change if this proves to be too high as Sensei’s implies) to convert to points rating. This is performed anew each round to maintain correspondence with OGS ratings. Each player has a “rating offset” which is persistent between rounds and increases by one for each win, decreases by one for each loss, and can be manually adjusted by request (subject to approval). Before pairings are made and reverse komi values generated, the points rating and rating offset are added together.

I appreciate that Lagrange was centred on reverse komi, but I’m not running a system like that simply to provide handicap as an auxiliary tool.

If that’s how reverse komi has to be then I’d rather use handicap stones.

I guess 1 Handicap stone is worth more than six points.

Score Estimator on 9 HC

That’s only 9 points short of what would be estimated for komi 7 games, and 0 points short of what would be estimated for 6.5 komi games, rounding to the nearest point. Not bad.

Here are the systems I’m willing to consider:

  1. handicap stones = ⌊ΔR / 2⌋
  2. reverse komi = 3ΔR (ignored for one-rank difference)
  3. reverse komi = ⌊3.5ΔR⌋ → nearest multiple of 7

I can’t run any system more complicated than that.

1 Like

The analogue of handicap stones = ⌊ΔR / 2⌋ would be reverse komi = 7ΔR.

2 Likes

How about

reverse komi = 14ΔR
… ?

I could be wrong but that’s how I understand the formula in use by Samraku, and it doesn’t seem so complicated.

//Edit: I have no problem with normal handicap either, or free placement.

Or how about free placement and additionally the weaker player may trade handicap stones for reverse komi, just to give them even more options before the game even starts :laughing:

For comparison:

Rank dif. H = ⌊ΔR / 2⌋ RK = 3ΔR (unl. R = 1) RK = ⌊3.5ΔR⌋ → nr. 7
1 0 stones 0 pt 0 pt
2 1 stone 6 pt 7 pt
3 1 stone 9 pt 7 pt
4 2 stones 12 pt 14pt
5 2 stones 15 pt 14pt
6 3 stones 18 pt 21 pt
7 3 stones 21 pt 21 pt
8 4 stones 24 pt 28 pt
9 4 stones 27 pt 28 pt

I am pretty ambivalent, but I need the system to be simple.

1 Like