Re: Pente Masters 2020: Solving the seed problem
Posted:
Jul 22, 2020, 1:00 AM
If rich isn't reseeded and we use pente_gon's pot sorting semi-random selection, then it is almost certainly the case that zoey will be placed in Pot C and richard in pot D. If I thus understand it correctly, there will still be a 25% chance that both richard and zoey are placed in the same section with a top 4 seed and a top 8 seed. Unlikely I know, but we then have a 1 in 8 chance of:
pente_gon/watsu/be_water/karlw spavacz/madhippo zoey richard
in the same section in Round 2, much stronger than the other sections.
Anyway, I still think it's an improvement over the current system and we could sit here complaining all day about any outcome--that is not my intention. I just wanted to get the information out.
It's no use going back to yesterday, because I was a different person then.
I'd suggest if provisional rated players are playing the tournament than their rating is to be treated as established. Not for this tournament, ofc, just for the future.
Re: Pente Masters 2020: Solving the seed problem
Posted:
Mar 17, 2025, 2:45 PM
first of all, the seeding is explained in the FAQ, for the single elimination format at least, the lowest-seeded faces the highest-seeded player, the 2nd seed faces the 2nd lowest seed. A special consideration is done for Byes.
If you allow non-established players to play, you need to see them for what they are, non-established. A different formula is used for their 1st 20 games so their rating converges faster to their actual ability, but this also means high volatility at the start, with their variance converging to that of established players as they reach 20 games. My point is there is no point in treating them as established.
I don't think seeding is what you're looking for, if you're looking for more chances to advance to the next round, you'd need to use one of the other formats.
Re: Pente Masters 2020: Solving the seed problem
Posted:
Mar 18, 2025, 7:43 PM
Just to mention it, I know what seeding is, I've used to be a referee a lot in both online and offline renju tournaments. I know how to pair swiss, round robin, single elimination, double elimination, mcmahon (not sure about spelling) and some other systems.
> if you're looking for more chances to advance to the next round
If you are talking about me personally, than the answer is no, I don't care much about my result in the tournament, I play here for fun and to connect communities somehow.
My motivation was to use seeding for what it really is, I've wrote it in the previous message.
How do you think, being 2nd rated in the tournament (or 3rd if you include provisionals), how should he feel?..
So the question persists. Provisional rating, despite having higher volatility, still predicts results better than a constant set to 0 (or 1200, or 1600, or whatever, but 0 is the worst imo). Why do you think 0 is better?
Re: Pente Masters 2020: Solving the seed problem
Posted:
Mar 19, 2025, 1:40 AM
Can you point me to some documentation on the topic? We are still using the algorithms from before my time (I took over in 2013), and the website is just 2 years shy of turning 30 yours old (it became live in 1997 I believe, when I was a wee lad).
I've only made changes to the round robin pairing since then, and I'm keen to learn more. If you want to discuss the topic more, please do make a new thread, I doubt the people in this one are all interested and may receive email notifications for each reply (kindly send me a pm to point my attention to the new thread if you do).
