Posts:
1,049
Registered:
Mar 7, 2006
From:
Philadelphia, PA
Age:
38
Re: An argument against randomized sections
Posted:
Sep 15, 2026, 3:51 PM
Yes.
Exact same as current method, but after first assignment, count the number of matchups that are repeats of previous rounds (in this case, there are 4).
Then, starting with the two highest seeds (or two lowest, it doesn't matter), swap adjacent seeds in adjacent sections whenever doing so reduces the count of repeat matchups. Continue down the seeding until you reach zero repeat matchups or the end.
In this case, you would swap karl and zoey, which reduces the number of repeat matchups to 2. You would not swap tempest and owen, because that would keep the number at 2. You would not swap mumix and noah, because that would keep the number at 2. The sections would look like this:
karl owen mumix madmike
zoey tempest noah
With the only repeats being zoey vs. tempest and (unfortunately) owen vs. mumix. If this had been done for round 2, mumix and owen most likely would not have been repeat paired.
It's no use going back to yesterday, because I was a different person then.
Re: An argument against randomized sections
Posted:
Sep 15, 2026, 4:34 PM
You may miss global minimum with this algorythm (and why to minimize it at all?)
I'm happy to play mumix, btw. Of course, I could have a chance to earn some money with no risk if seeded in the same group with you, but it happens, no worries.
The system itself isn't fair at all, that's more important. For example, chances in group of 3 and in group of 4 are very different; not saying that 2/3 is greater than 2/4, also if you draw every set, in group of 3 you 100% go further, in group of 4 you may go out if unlucky. Also seeding process after Rd 1 should not rely on ratings but on the tournament performance as I already wrote.
For madmike, the seeding process is clearly unfair because he will surely play in group of 4 and it directly lowers his chances to advance to the next round. That is the thing that should be changed, not my chances to meet mumix or your chances to meet me.
Posts:
1,049
Registered:
Mar 7, 2006
From:
Philadelphia, PA
Age:
38
Re: An argument against randomized sections
Posted:
Sep 15, 2026, 4:58 PM
I agree with all of that.
Perhaps we can submit a proposal with mutually agreed upon tournament rules for the next iteration of Pente masters.
Unfortunately, the next one is not likely to have the same top-level attendance as the current one. Such is the state of modern pente. We'll have to work within the confines of what we have today.
At the very least, round 4 is almost certain to be free of any logistical problems, and will hopefully produce an indisputable champion.
It's no use going back to yesterday, because I was a different person then.
Posts:
1,049
Registered:
Mar 7, 2006
From:
Philadelphia, PA
Age:
38
Re: An argument against randomized sections
Posted:
Sep 15, 2026, 6:08 PM
Noah is anything but a bottom seed at this point.
His live rating is 60 points higher than mumix, and he has no set losses, unlike mumix.
Mumix is the true bottom seed.
And mike, I think I've done a fairly thorough job explaining how the seeding/sorting process works in this and other threads. Seeds are recalculated at the start of each round based on live rating, then assigned to sections based on rules explained in detail here:
Re: An argument against randomized sections
Posted:
Sep 15, 2026, 7:30 PM
> if i take my personal feelings out, bottom seed (me) doesn't really belong in the next round.
C'mon, why not! Zoey goes with 100% winrate, me 80%, you and Karl 75%, that's the current top-4. Bad thing is, seeding of Rd. 3 ignores that fact. So Karl plays with 50% guys and you have to keep up in a tougher group. But still you can manage it, keep fighting!
====== Some time ago I wrote a suggestion. The main thing is, if groups are unavoidable (why not swiss for 7+ players?), they should have the same size in the round and seeding should use tournament performance as a main parameter, rating should be secondary. I wrote it in detail before.
Re: An argument against randomized sections
Posted:
Sep 15, 2026, 7:51 PM
nah, i think i truly the bottom. i cant reliably beat mumix, and not enough data to support anything but probably loser to all the rest. maybe i can beat tempest but he has much higher rating and i not in the same group.
Posts:
1,049
Registered:
Mar 7, 2006
From:
Philadelphia, PA
Age:
38
Re: An argument against randomized sections
Posted:
Sep 15, 2026, 7:58 PM
Nice try, owen. A more accurate mid-tournament appraisal would not include timeouts. I would've thought you'd be above lying with statistics.
My own preferred (biased) metric is looking strictly at set losses. Of the remaining seven competitors, only two have set losses (excluding timeouts in won positions, lol) and they are both against zoey. People seem to drop a level in skill when they play zoey, or perhaps his opening prep is truly next level.
So I rank current performance as zoey at the top, then a four-way tie for 2nd (with mumix at the bottom, cuz he's mumix), then tempest and madmike.
Finally, not that it matters due to many of the games being decades old, but I have a winning TB record against everyone in the field. I'm not going to post the stats because it might verge on bullying.
It's no use going back to yesterday, because I was a different person then.
Posts:
2,316
Registered:
Mar 4, 2007
From:
San Francisco
Age:
45 Home page
Re: An argument against randomized sections
Posted:
Sep 15, 2026, 11:18 PM
Ahh shucks. You're making me blush karl. Im sure I'll start losing eventually here. Just got lucky so far
I only put in between 10 to 30 minutes of prep per opponent (my p2). But I use methods in that short time, which i discussed with you over private message.
I guess you can coin it "the zoey effect".
Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
