Hi Everyone
I'm making this thread to reflect back on the rands swiss tour and try to collect all the lessons I've learned from it.
I'll start by saying that for the large part, the tour ran as I was expecting it to, and that I'm overall satisfied with it, however I'm not making this to pat myself on the back & talk up about what a good job I did. This is so I can write down all the mistakes that have been made, and what ways that I can improve on them.
Because of that, I'd like to turn to all of you. If you participated in the tournament & weren't happy with any of the decisions I made, methods I used, or actions I took, then I want to hear your thoughts.
A few requests before you post though:
Thank you for your time, thought, and opinions. They truly mean a lot to me.
I'm making this thread to reflect back on the rands swiss tour and try to collect all the lessons I've learned from it.
I'll start by saying that for the large part, the tour ran as I was expecting it to, and that I'm overall satisfied with it, however I'm not making this to pat myself on the back & talk up about what a good job I did. This is so I can write down all the mistakes that have been made, and what ways that I can improve on them.
Because of that, I'd like to turn to all of you. If you participated in the tournament & weren't happy with any of the decisions I made, methods I used, or actions I took, then I want to hear your thoughts.
A few requests before you post though:
- Keep it civil. If I did a bad job then I want you to tell me, but remember that I did all this in my own time with the best interests of the website in mind. I'm rather thick-skinned and will be quite lax with what counts as too far, but don't take this thread as an excuse to call everyone idiots
- limit your feedback to myself and the tournament itself. This is not a discussion about how good/bad my cohosts performed, and I will not condone any dogpiling/insults laid towards them. This thread is about what I can do better.
- If you didn't like something/thought something was done wrong, thats great and I want to hear it, but it would be even better if you could provide ideas/suggestions for what could be done instead. Have a think about potential solutions to the problems you saw. Even if your idea isn't fully fleshed out, let me know!
The Challenge: a hard top-cut induces uncontrolled luck mechanics (in the form of Resistance) which feel bad for players caught up in them.
My Thoughts: I've had a lot of experience with swiss-system tournaments during my time in the Yugioh community, however there are some distinct differences between how they run their tournaments compared to how VGC is done. One difference is that they do what can be referred to as a "hard cut" - the top 8/16/32 players make it to the single elimination rounds, and if you got 9th/17th/33rd well tough luck. I remember it always being a pain point, being "on the bubble" and just hoping your tiebreaker points were enough to give you placement. It's not a good feeling.
Some people have suggested replacing Resistance with a game-wins scenario, (IE count how many 2-0s vs 2-1s you got) to reduce that luck feeling, but I think this creates a bunch of extra admin overhead without much benefit - especially if we do an asymmetric cut.
I'd heard that asymmetric top cuts were a thing that VGC did, however I'd never experienced one myself. The idea intrigued me as a potential solution to this hard cut problem, but I was unfamiliar with how they worked, and with all the other preparation for this event mounting (writing the points system for the circuit, building the standings spreadsheet from scratch, and many other things) I put the thought aside and decided to stick with what I knew.
Since then, I've read up and learned more about how to run an asymmetric top cut, and the system is much easier to implement than I thought it would be. In hindsight, I should have simply said I would be doing one, and then figured out how later, but it is what it is.
My Solution: if I run a swiss tournament again, I will do an asymmetric top cut
My Thoughts: I've had a lot of experience with swiss-system tournaments during my time in the Yugioh community, however there are some distinct differences between how they run their tournaments compared to how VGC is done. One difference is that they do what can be referred to as a "hard cut" - the top 8/16/32 players make it to the single elimination rounds, and if you got 9th/17th/33rd well tough luck. I remember it always being a pain point, being "on the bubble" and just hoping your tiebreaker points were enough to give you placement. It's not a good feeling.
Some people have suggested replacing Resistance with a game-wins scenario, (IE count how many 2-0s vs 2-1s you got) to reduce that luck feeling, but I think this creates a bunch of extra admin overhead without much benefit - especially if we do an asymmetric cut.
I'd heard that asymmetric top cuts were a thing that VGC did, however I'd never experienced one myself. The idea intrigued me as a potential solution to this hard cut problem, but I was unfamiliar with how they worked, and with all the other preparation for this event mounting (writing the points system for the circuit, building the standings spreadsheet from scratch, and many other things) I put the thought aside and decided to stick with what I knew.
Since then, I've read up and learned more about how to run an asymmetric top cut, and the system is much easier to implement than I thought it would be. In hindsight, I should have simply said I would be doing one, and then figured out how later, but it is what it is.
My Solution: if I run a swiss tournament again, I will do an asymmetric top cut
The Challenge: As some of you remember, back in round 2, one set of pairings were posted, only for them to be walked back and replaced with a new set just 2 hours later.
My Thoughts: This was a mistake on my behalf - I prepared a draft OP with the pairings in place when I made my closing act decisions post, but between when I wrote that draft and when the round was ready to post, players had dropped, extensions had been granted, and some games had their results changed. This was simply a teething issue that was caused by the host team not yet getting into the swing of things. Since then, we made a system where whoever was posting the OP would verify that everything had been properly recorded, and if any changes were needed, they would know how to generate a new set of pairings before posting.
This to me could also be solved by having a dedicated program that generates pairings, rather than my cobbled-together spreadsheet, and there are discussions happening in the background to hopefully make that a reality one day, but for now it's simply a matter of being prepared and familiar with how to generate the pairings, and having a team of people who can help cross-check for anything the main host might have forgotten about.
My Solution: I will be working with some friends in order to build a dedicated pairings app/webapp. I can't guarantee this will happen, but especially if tournaments like this become more common, I'd like to try and pursue it
My Thoughts: This was a mistake on my behalf - I prepared a draft OP with the pairings in place when I made my closing act decisions post, but between when I wrote that draft and when the round was ready to post, players had dropped, extensions had been granted, and some games had their results changed. This was simply a teething issue that was caused by the host team not yet getting into the swing of things. Since then, we made a system where whoever was posting the OP would verify that everything had been properly recorded, and if any changes were needed, they would know how to generate a new set of pairings before posting.
This to me could also be solved by having a dedicated program that generates pairings, rather than my cobbled-together spreadsheet, and there are discussions happening in the background to hopefully make that a reality one day, but for now it's simply a matter of being prepared and familiar with how to generate the pairings, and having a team of people who can help cross-check for anything the main host might have forgotten about.
My Solution: I will be working with some friends in order to build a dedicated pairings app/webapp. I can't guarantee this will happen, but especially if tournaments like this become more common, I'd like to try and pursue it
The Challenge: Week-by-week tournaments are a big commitment, especially when they are as large as this. 10 rounds is a lot, and it can feel especially disheartening continuing on once you know there's no prize at the end of it for you.
My thoughts: I wanted this tournament to be inclusive as possible, for the best players down to those just wanting to meet new people and learn the format. That is why I made the decision not to do a auto-drop on x-3 like other smogon tournaments do. Unfortunately that comes with the downside of people not caring about playing and ghosting the event. I tried to mitigate this by advertising you can drop just by asking, and while I think that did work a lot, it seems that it clearly wasn't enough, as some people have had multiple back-to-back act wins where the opponent didn't even give them the dignity of a response.
My Solution: this problem is a very difficult one to solve, but I think the only solution that doesn't involve more or less rewriting the Swiss format entirely is to change the incentive structure. If the event were pure Swiss (ie no top cut) would players still complain about burnout? I'm not sure. Therefore, I think there are 2 parts to this. Firstly, making dropping out as easy and front-of-your-mind as possible. And secondly, making top cut less meaningful (but still worth pursuing) - the specifics? I'm unsure
My thoughts: I wanted this tournament to be inclusive as possible, for the best players down to those just wanting to meet new people and learn the format. That is why I made the decision not to do a auto-drop on x-3 like other smogon tournaments do. Unfortunately that comes with the downside of people not caring about playing and ghosting the event. I tried to mitigate this by advertising you can drop just by asking, and while I think that did work a lot, it seems that it clearly wasn't enough, as some people have had multiple back-to-back act wins where the opponent didn't even give them the dignity of a response.
My Solution: this problem is a very difficult one to solve, but I think the only solution that doesn't involve more or less rewriting the Swiss format entirely is to change the incentive structure. If the event were pure Swiss (ie no top cut) would players still complain about burnout? I'm not sure. Therefore, I think there are 2 parts to this. Firstly, making dropping out as easy and front-of-your-mind as possible. And secondly, making top cut less meaningful (but still worth pursuing) - the specifics? I'm unsure
The Challenge: Smogon tournaments have an inactivity problem. Players will sign up for an event, and then forget they signed up & never participate. In elimination events this isn't too bad as they vanish in round 1, unless you get 2 inactive players face each other and you give them a coinflip, giving one of them a win and pushing them into the next round. At a critical mass of inactive players, this can cascade 2, 3, and even 4 rounds deep.
My Solution was to remove the coinflip mechanic entirely, and perform manual activity checks, dropping people out if there were no signs of participation. This was modelled off how it is done in live yugioh events, where after a player is 3 minutes late to a game, they would recieve a game 1 loss, and after 10 minutes of tardiness they would lose the match & get kicked from the event - it was then their responsibility to come back to request the Tournament Host reinstate them.
My Thoughts: Overall I'm really happy with how this policy turned out. out of 602 players, 244 were automatically dropped by the activity sweep, and a further 76 requested to be dropped manually. In an ideal world, the voluntary drop number would have been much higher, but there is very little that can be done to incentivize people to take that step. Rules in the OP are not sufficient as many people don't read them, and punishing players by forfeiting their circuit points if they get picked up by the activity sweep seems unfair in my opinion.
My Solution was to remove the coinflip mechanic entirely, and perform manual activity checks, dropping people out if there were no signs of participation. This was modelled off how it is done in live yugioh events, where after a player is 3 minutes late to a game, they would recieve a game 1 loss, and after 10 minutes of tardiness they would lose the match & get kicked from the event - it was then their responsibility to come back to request the Tournament Host reinstate them.
My Thoughts: Overall I'm really happy with how this policy turned out. out of 602 players, 244 were automatically dropped by the activity sweep, and a further 76 requested to be dropped manually. In an ideal world, the voluntary drop number would have been much higher, but there is very little that can be done to incentivize people to take that step. Rules in the OP are not sufficient as many people don't read them, and punishing players by forfeiting their circuit points if they get picked up by the activity sweep seems unfair in my opinion.
The Challenge: Unexpected scenarios throughout the event popped up, where I made a few manual adjustments to pairings, reversed the recorded result of some games, granted some extensions, and other things that kept the tour running smoothly. I don't expect any of these to be things an average host being handed my sheet (or given access to a pairings manager) and told "good luck" could (or even should) do, so doing things like that aren't exactly sustainable.
My Thoughts: Since I was the one who built the spreadsheet, I was intimately familiar with how it operated, and therefore knew what I could/couldn't do. This made making these changes possible. I think that many of these problems I created for myself by being lenient with extensions despite saying they would be granted very rarely, by taking responsibility for clerical issues when the rules could've easily dictated that it was player responsibility to verify, etc.
I think that someone who is more firm on rule enforcement than I am, coupled with players being educated better on said rules, will help mitigate problems, but they can't be entirely removed without the kind of manual fiddling that needs experience to do properly.
Once again I don't have a "solution" for this other than to more or less ignore the problem by telling future hosts to just not do what I did, but I feel like a good & experienced host should be able to make calls like this & do things that give players the best experience, rather than strictly adhering to book rules even when a solution that would make everyone happy is available.
My Thoughts: Since I was the one who built the spreadsheet, I was intimately familiar with how it operated, and therefore knew what I could/couldn't do. This made making these changes possible. I think that many of these problems I created for myself by being lenient with extensions despite saying they would be granted very rarely, by taking responsibility for clerical issues when the rules could've easily dictated that it was player responsibility to verify, etc.
I think that someone who is more firm on rule enforcement than I am, coupled with players being educated better on said rules, will help mitigate problems, but they can't be entirely removed without the kind of manual fiddling that needs experience to do properly.
Once again I don't have a "solution" for this other than to more or less ignore the problem by telling future hosts to just not do what I did, but I feel like a good & experienced host should be able to make calls like this & do things that give players the best experience, rather than strictly adhering to book rules even when a solution that would make everyone happy is available.
The Challenge: Extensions is not only a lot of work in Swiss, but it is also risky. If an extension is granted but the game is not played, then you've just added a pair-down into your bracket. If there was already a pairdown in this existing bracket, then you've now got an invalid group of pairings.
My Thoughts: as mentioned above, I granted extensions, and quite liberally in the early rounds, which I immediately regretted when one extension simply didn't play their match after all and I was forced to award a double loss, leaving me stuck with 2 pair-downs in a single round. I still think granting some extensions is doable, but the administrative work needed for it is tedious and risky, so I learned pretty quickly to only grant 1 extension per round, and only grant it if the players directly told me the time they were playing.
My Solution: I think simply denying all extensions is valid, but can cause players who are acting in good faith to get unnecessarily (but not unfairly) punished. I think that the best policy would be that extensions are only granted if:
My Thoughts: as mentioned above, I granted extensions, and quite liberally in the early rounds, which I immediately regretted when one extension simply didn't play their match after all and I was forced to award a double loss, leaving me stuck with 2 pair-downs in a single round. I still think granting some extensions is doable, but the administrative work needed for it is tedious and risky, so I learned pretty quickly to only grant 1 extension per round, and only grant it if the players directly told me the time they were playing.
My Solution: I think simply denying all extensions is valid, but can cause players who are acting in good faith to get unnecessarily (but not unfairly) punished. I think that the best policy would be that extensions are only granted if:
- both players request the extension (either by both posting in the thread, or by one saying "please get an extension" and the other posting a screenshot of it
- the time the game will be played is cited in the request
- the time is less than 48 hours from the deadline
- the host believes that both players were scheduling in good faith, and that the game will be played
The Challenge: Players who don't want to participate anymore can message their opponent "take the win" and therefore bypass the activity sweeps. Keeping them in the event even though they aren't participating.
My Thoughts: Early in the tour, we started marking wins whenever someone said "my opponent gave me the win" - I realised later on that this undermined the activity sweeps we already had set in place, causing people to come online monday, give their opponent a win, and then repeat the action again 2 more weeks in a row with no repurcussions.
My Solution: We switched to treating forfeits as activity losses from about round 6 onwards. This caused an uptick in act drops, which in my opinion is a good thing - after all, why are you in the tournament if you dont want to play?
My Thoughts: Early in the tour, we started marking wins whenever someone said "my opponent gave me the win" - I realised later on that this undermined the activity sweeps we already had set in place, causing people to come online monday, give their opponent a win, and then repeat the action again 2 more weeks in a row with no repurcussions.
My Solution: We switched to treating forfeits as activity losses from about round 6 onwards. This caused an uptick in act drops, which in my opinion is a good thing - after all, why are you in the tournament if you dont want to play?
Thank you for your time, thought, and opinions. They truly mean a lot to me.
Last edited:
