Experimental Tournament Format Changes Discussion

(Sep. 08, 2016  12:45 AM)Cake Wrote: It appears that the Challonge API can only access tournament data for the user whose API key it's using, which means that if we want to make a tool for converting tournament data, either anyone who wants to use it will need to pass it their API key and username, or there would have to be a single account that all tournaments would be hosted on. Unless I'm misreading it, that is Tongue_out

@[Cake]
I can now confirm that in the case where you just want to retrieve tournament data (such as name, date, participant list and matches), no api key is needed for the user who created it. I've just tested this by retrieving the data for Kei's Challonge records (THE BEY-GINNING OF AN ERA : Experimental in fact) without his API key (just used my credentials).

Therefore, an automatic Challonge to spreadsheet converter is possible, and will be the subject of my free time for a while. If successful, it will remove the need to create spreadsheets after a tournament, as the hosts just need to interact with Challonge itself and everything else will be semi-automated.
Amazing news Ben! Thanks so much.
(Sep. 08, 2016  6:19 PM)Manicben Wrote: Therefore, an automatic Challonge to spreadsheet converter is possible, and will be the subject of my free time for a while. If successful, it will remove the need to create spreadsheets after a tournament, as the hosts just need to interact with Challonge itself and everything else will be semi-automated.

Awesome! This would be so great. Things like Winning Combos would still have to be manually entered, but this sounds like it would take out the vast majority of the work.
(Sep. 08, 2016  7:47 PM)Kei Wrote:
(Sep. 08, 2016  6:19 PM)Manicben Wrote: Therefore, an automatic Challonge to spreadsheet converter is possible, and will be the subject of my free time for a while. If successful, it will remove the need to create spreadsheets after a tournament, as the hosts just need to interact with Challonge itself and everything else will be semi-automated.

Awesome! This would be so great. Things like Winning Combos would still have to be manually entered, but this sounds like it would take out the vast majority of the work.

Yo, if hosts did a very cheeky edit of the tournament description afterwards with the Winning combos, I could automate that too Grin Just need to parse through the text till I find a token (say some kind of codeword), then start taking words in as input and sort them into combo names associated with usernames and places. Sounds tricky and will definitely take a day or two to implement, but is defo possible.
Woah, hey guys welcome to my thoughts!

I honestly didn't have a problem with this format either than the 2 point KO rule, it seemed to be more RNG than bursts at the event, and I absolutely hate any type of RNG in anything, RNG is SUCC. But yeah I just generally don't think the 2 point rule should apply because it's first to 3 and it really seems like you're just getting screwed over unlike in deck rotation where you could counter your opponent to make up for the 2 points you lost last round. You cant do that in the regular format which makes you feel cheated when you get bursted or KOed for 2 points. It really sucks and it's something you don't understand until it happens to you lol (mostly because it's so competitive to get into the finals, if it's a stamina match and you get an unlucky burst or KO then, yeah have fun never being top 1 again).
Thanks 1234! The 2-point system in deck rotation changes the flow of gameplay entirely and both incentivizes aggressive maneuvers and encourages players to keep effective defensive options at the ready and to be wary of using fragile combos, but since you can't really counter in the standard game, that flow is lost. At least we can put that to rest.
(Sep. 07, 2016  6:13 PM)Bey Brad Wrote: How about giving the loser the opportunity to switch, and only if they switch is the winner also allowed to switch (or not)? This gives the loser of the previous round some control over the tempo of battle, at least

Also, what should the rules for switching and re-attaching look like? When it's just the loser switching, they can just grab it and attach it. With both players, they will have to conceal and attach secretly.


Winner switches first (or chooses not to switch) without concealing their choice. Then loser switches in response. From what I can gather, the issue is that forcing the winner to carry-over their choice while allowing the loser to switch gives the players the feeling of rewarding the loser (for losing) and punishing the winner (for winning). Allowing the winner to switch removes this mindset because the winner retains control of the match, even though they are now on the defensive. I have never had anyone complain about the winner being allowed to switch, as above.

You know, I do prefer the mathematical perfection of the winner retaining their current choice (MarioKart, anyone?) but nobody else seems to appreciate this idea in practice. My sample size is about two-dozen, most of which are not WBO members and the results are for an only-similar game system, so obviously this might not translate perfectly. But the likeness is uncanny enough for me to comment. In any case, it's just information. I'm giving it to you for free. You obviously don't need to use it.
Quote:Winner switches first (or chooses not to switch) without concealing their choice. Then loser switches in response.

But doesn't this just allow the loser to counterpick anyway? I feel like that's why people dislike the system in the first place. (Though I would say again that it hasn't really come up when we've played deck so far.)
Link to post with more thoughts on this

Ok so I've still been pretty busy lately (I've been online a few times in the past few days, it was mainly just so I could reply to PMs because doing business with somebody), but I promised that since I had work off today I would reply to this, so here goes. Kinda just addressing points in the order I think of them as well.

Challonge vs Paper: I would strongly advocate the use of Challonge when possible, but due to the limitations that TSO and a few other people stated I think that making it mandatory would be a bad idea. Seeding should be more strictly enforced however. I disagree with the argument that "if Wifi is not available at the location, look up Ranks and create the brackets the night before" argument, simply due to the fact that it's not at all a rare occurrence that a new or unexpected player will appear and throw a wrench in things. This isn't as much of an issue for someone as obsessed with the ranking system as me (lol) since I have a general idea of where most people stand, but it could cause problems for other users who aren't as familiar with it.

This also brings up the (relatively minor) issue of seeding based on rank vs points: would a new player be seeded as "the lowest rank" or "1,000" (and therefore a higher seed than any negative players)?

As for the Challonge API thing, that would be interesting to see implemented if it can streamline the results process, but I don't know how accessible it would be some hosts (I'm kinda confused by it just by what people have posted about it so far tbh). I do know that each match has a "notes" section where I would think you could list the combos used for the Winning Combinations list.

Weighted Scores vs Tiebreak Matches: I understand the efficiency idea behind using weighted scores over tiebreak matches as a 3-way tiebreaker can loop through many times before being resolved. However, from a gameplay perspective there's a huge psychological element to the "second/third/nth encounter" during a tournament, and I don't think that should just be discarded. Weighing wins don't solve the infinite 3-person loop either, as all three players will have identical weights in previous matches.

Switching during Deck Rotation: Agree with Brad's point about the "winner switches, then loser switches in response" idea - I'm not really sure how it's supposed to fix the counterpicking issue we've been trying to address. The "winner can only switch if the loser switches" idea is definitely worth trying out though.

And yeah, switches should have some rule like that upon losing, the loser declares that they're going to switch, and both players turn thei backs on each other. The loser must switch if they declare they are going to (I see this as like declaring you are attached), and the winner may or may not switch.

Swiss in small Events/extending Swiss: While it would provide more matches and fix the "small Swiss events are too short" issue, I'm not too sure how I feel about extending the rounds. Ideally, Swiss is supposed to end when there is only one player left undefeated, with all the win records in a "binary" looking order (if 1 is a win, and 0 is a loss, then the bottom of the bracket will end at 0000 and increase in a binary pattern until it reaches 1111) and with no players having played more than one match against one another.

I haven't tried simulating extended Swiss yet, but I imagine pairing past the "ideal" round would require some questionable improvisation and the extra rounds might jumble up the scores too much. Like Kei said, given the short and dynamic nature of a Beyblade battle, I still think that (Block) Round Robin is the best options for smaller events. I think 16 is the magic number where you should switch to Swiss.

Will probably edit this post later with a simulation of the extended rounds and an example of the binary thing.

2 point KOs/Bursts: Were random KOs really frequent enough to seem unfair compared to random Bursts? Isn't the reason that people don't use KO-based combos in Burst is because KOs are so unlikely to happen?

I'm totally on board for this rule not applying to the preliminary 3-point matches though. I actually probably didn't read closely enough and must have assumed it only applied to Deck Rotation.

SE in finals: Now that the third place match has been added, I don't really see a problem with this anymore.
(Sep. 09, 2016  2:26 PM)Bey Brad Wrote:
Quote:Winner switches first (or chooses not to switch) without concealing their choice. Then loser switches in response.

But doesn't this just allow the loser to counterpick anyway? I feel like that's why people dislike the system in the first place. (Though I would say again that it hasn't really come up when we've played deck so far.)

Yeah, you're absolutely right. But this is precisely the point. It makes mathematical sense for the winner to not-switch, because this keeps each round fair. But as it turns out, artificially levelling the playing field by handicapping the winner is counter-intuitive to the normal psychology of winning. The winner should receive positive reinforcement - not negative punishment. This is why the winners complain.

Allowing the winner to switch gives the winner the proactive power to control the pace of the match. For example, if they notice the loser has a weak launch, they might switch their blade to suit - even though they just won with the blade they're already using. This limits the loser's ability to adapt to the scenario, without crippling them completely. Even if the loser is still given the mathematical advantage in this scenario (by switching second), the winner is given the power to influence the loser's choice... Which is, psychologically, just as gratifying as any other form of reward.

The winner switching will always be less mathematically "fair" than just letting the loser switch. But keeping everyone happy is far more important (to me, in a game about fun) than punishing the winner in the name of fairness. A subjective notion perhaps - and I originally felt the same way as you - but that's why I feel the test of time seems so important in this discussion.

(Sep. 09, 2016  8:57 PM)Wombat Wrote: Switching during Deck Rotation: Agree with Brad's point about the "winner switches, then loser switches in response" idea - I'm not really sure how it's supposed to fix the counterpicking issue we've been trying to address. The "winner can only switch if the loser switches" idea is definitely worth trying out though.

And yeah, switches should have some rule like that upon losing, the loser declares that they're going to switch, and both players turn thei backs on each other. The loser must switch if they declare they are going to (I see this as like declaring you are attached), and the winner may or may not switch.

I appreciate the idea of making switches dependent on the loser - but I think your notes on the consequences of that system illustrate how it might get messy. You basically just end up with double-blind again. The beauty of the deck rotation seems to be in revealing new layers of strategy for the game - which perpetual double-blind would negate. Particularly if you were facing a sore loser.

Edit: forgot to sign!
I get what you're thinking Beylon, thanks for explaining. I think it's a very interesting compromise.
Personally, i think KO's should leave and just leave Bursts a 2 pointer. Heck, if i got an attack build bey that's what i'm going to be aiming for.
(Sep. 10, 2016  2:52 AM)johnwolf Wrote: Personally, i think KO's should leave and just leave Bursts a 2 pointer. Heck, if i got an attack build bey that's what i'm going to be aiming for.
I think the problem with making burst 2 points in a 3 point to win format is even in stamina matches you can get the random burst, and if losing to that random burst sets you so far behind. In theory the idea seems good to try to get more attack types to be used, but the Canadian players who participated in the first and only test event for this format so far seemed to dislike this rule for the reasons I stated above, and maybe even some more reasons that I can't quite remember off the top off my head.
Right, and those same players had no issues with the rule in deck. So clearly the 3 points thing and inability to switch changes the dynamic a lot.
(Sep. 10, 2016  3:24 AM)Thunder Dome Wrote:
(Sep. 10, 2016  2:52 AM)johnwolf Wrote: Personally, i think KO's should leave and just leave Bursts a 2 pointer. Heck, if i got an attack build bey that's what i'm going to be aiming for.
I think the problem with making burst 2 points in a 3 point to win format is even in stamina matches you can get the random burst, and if losing to that random burst sets you so far behind. In theory the idea seems good to try to get more attack types to be used, but the Canadian players who participated in the first and only test event for this format so far seemed to dislike this rule for the reasons I stated above, and maybe even some more reasons that I can't quite remember off the top off my head.

That being said, it's a completely different bey system to be honest. It means you need to take a LOT into consideration when choosing your bey. From what i've heard, Defensive Stamina builds where the norm in other Tournaments. If you remove bursting advantage you remove the appeal again to Attack type beys. Bursting sucks if you're not prepared for it, but, you need to think these through too. It's also just as important to take into consideration that you can catch up just as fast if YOU get a burst finish over your opponent, so, really, you're not that far behind.
I think we're at the point now where we need to accept that we can't really incentivize Attack-type play in the state that it's in that much. However, deck rotation is one of our attempts at giving attack a chance to shine, and since you know what you're getting into when you use it it removes some of the fear.

Besides, I don't think bursts are actually that much more frequent with attack-types vs. other types. D2 Vs. D2 are stamina-defense mirror matches that burst like crazy.
(Sep. 06, 2016  4:46 PM)Bey Brad Wrote: I think this is obvious, though? Which is why I don't agree with your comments that losing two matches eliminating you from finals progression is "harsh." If losing two matches doesn't prevent you from advancing, what should? Ultimately, making it to the finals should not be considered a "default state" even for high-level players.

Beyblade does have a luck-factor — deck format in the finals mitigates this to some degree — but that's just the nature of the beast. While hardcore players may be happy to play 10 matches or more pre-finals, that amount of playing (and mostly, waiting) can wear down a lot of people.

It's hard to say because it really depends on the volume of matches played in total, but I'd say in general that amassing three losses pre-finals for any format is the point where any player shouldn't feel bad if they don't make the finals of any event. Considering the role of luck in the selection process–in spite of how well you can scout your opponents–and the role of luck that exists in any battle, I think it's fair to say that it is plausible for any Blader to lose a couple times largely due to luck-based reasons yet still ultimately be the best Blader of the tournament if they are given the chance to be (like they are in Round Robin). Not saying it is always viable regardless of format (especially Swiss; in huge Swiss events you have to accept that to keep the event at a reasonable length, your chances of advancing to the finals go down with even two losses), three losses is where I would personally draw the line in an ideal world.

Yeah, you might be right that it wears down some people. Although running a tournament efficiently with Challonge does help. 10 matches pre-finals is probably getting to the point of too much though, I agree. That's why Round Robin should probably be capped at 10 or 11 people instead of the current 12. An efficiently run RR tournament of that size can be completed relatively quickly.


(Sep. 08, 2016  4:01 AM)Bey Brad Wrote: If you don't know someone's BeyRank, you can just sort them in at the bottom, as the differences between scores are likely trivial. It's in high-ranking player's best interests to make sure they are actually seeded based on rank.

Yeah, this is the key point about seeding for me. It's among the highly-ranked players that this matters the most. Most players of this nature will have confirmed beforehand that they will be attending because they tend to be the more dedicated players, and if not, there's a good chance they will know their ranking or someone else will know at the event itself.

Precisely seeding players based on their actual rank would be the goal–it only took me a few minutes at the experimental event in Toronto–but as long as the highly-ranked players can be seeded at the top, that would be the most important thing to ensure. If there's a group of other players you're unsure of, just seed them near the bottom like Brad said.

(Sep. 09, 2016  8:57 PM)Wombat Wrote: This also brings up the (relatively minor) issue of seeding based on rank vs points: would a new player be seeded as "the lowest rank" or "1,000" (and therefore a higher seed than any negative players)?

Interesting issue, but I think it isn't worth thinking about too much. As I mentioned above, where this matters the most is among the highly-ranked players and making sure they are seeded at the top. In this scenario however, I would probably just follow the system of seeding based on rank. Players who are below 1,000BP are responsible for putting themselves in that position; they've "earned" that rank through poor play just like highly-ranked players have earned their ranks through good play. New players can't necessarily be assumed to be great or bad, so it would seem harsh to me to automatically assume they are bad players and seed them below any participating players who are under 1,000BP. It makes sense for them to be in between.

(Sep. 09, 2016  8:57 PM)Wombat Wrote: However, from a gameplay perspective there's a huge psychological element to the "second/third/nth encounter" during a tournament, and I don't think that should just be discarded. Weighing wins don't solve the infinite 3-person loop either, as all three players will have identical weights in previous matches.

Yeah, you have a point about being able to encounter an opponent multiple times throughout an event. However, as great as that can be from a competitive standpoint for those two players, part of what we're trying to do is limit or manage the number of BeyBattles that are played per event. Players who get to play in more matches–especially with highly-ranked players–are at an advantage over those who don't. Right now it's kind of all over the place because tiebreakers can infinitely loop and because many finals are often played using Round Robin style currently, adding even more battles between highly ranked players.

(Sep. 09, 2016  8:57 PM)Wombat Wrote: I haven't tried simulating extended Swiss yet, but I imagine pairing past the "ideal" round would require some questionable improvisation and the extra rounds might jumble up the scores too much. Like Kei said, given the short and dynamic nature of a Beyblade battle, I still think that (Block) Round Robin is the best options for smaller events. I think 16 is the magic number where you should switch to Swiss.

Yeah, I agree. The main thing we will need to determine precisely is the threshold between Round Robin, Block Round Robin, and Swiss. I mentioned earlier that 11 seems like a fair maximum for Round Robin. 12-16 seems right for Block Round Robin. 17+ seems like a better choice than 16 for when Swiss would be mandatory. In the Swiss Format rounds recommended for Magic: The Gathering, 17 is their threshold for moving up to five rounds (which to me seems like a fair minimum number of rounds for any player to be guaranteed in an event).

The tricky part is determining how we are going to determine the 4/8 finalists for our Single Elimination finals for every event without resorting to tiebreakers. You can use this page to run simulations. Tiebreakers seem inevitable in many scenarios, but we want to try and minimize those from happening. I'm not sure what the answer is yet because there's just so many different possibilities depending on the number of participants and if the paired down match-ups result in wins for the highly-ranked players or if they result in losses for those players.

(Sep. 09, 2016  8:57 PM)Wombat Wrote: 2 point KOs/Bursts: Were random KOs really frequent enough to seem unfair compared to random Bursts? Isn't the reason that people don't use KO-based combos in Burst is because KOs are so unlikely to happen?

Obviously our event is a small sample size, but it did feel like there were too many random KOs in relation to bursts, yes. Things like huge hits in the very first second of the battle when the Beyblades land in the stadium and one luckily hits the opponent out or silly taps from a Defense/Stamina type that end up making an Attack type get KOed. This is fine in Deck Format because you have a chance to make a comeback, but with three points needed for victory, this rule change was universally criticized at our event. Bursts didn't seem like as much of an issue, but I'd rather be safe and just keep things how they were before where everything is worth one point for our standard best 3 of 5 BeyBattles.

(Sep. 10, 2016  2:39 AM)Beylon Wrote: Yeah, you're absolutely right. But this is precisely the point. It makes mathematical sense for the winner to not-switch, because this keeps each round fair. But as it turns out, artificially levelling the playing field by handicapping the winner is counter-intuitive to the normal psychology of winning. The winner should receive positive reinforcement - not negative punishment. This is why the winners complain.

Allowing the winner to switch gives the winner the proactive power to control the pace of the match. For example, if they notice the loser has a weak launch, they might switch their blade to suit - even though they just won with the blade they're already using. This limits the loser's ability to adapt to the scenario, without crippling them completely. Even if the loser is still given the mathematical advantage in this scenario (by switching second), the winner is given the power to influence the loser's choice... Which is, psychologically, just as gratifying as any other form of reward.

The winner switching will always be less mathematically "fair" than just letting the loser switch. But keeping everyone happy is far more important (to me, in a game about fun) than punishing the winner in the name of fairness. A subjective notion perhaps - and I originally felt the same way as you - but that's why I feel the test of time seems so important in this discussion.

Really great reasoning and explanation here, Beylon. Thanks.

One of the reasons I love Deck Format is because–unlike the regular format where you are locked into one choice made in secret for the entire BeyBattle–it allows for losers to respond to their opponent. In Deck Format you should never run into a situation where every round feels insurmountable; you usually have at least some glimmer of hope if you've constructed your deck properly and can execute your strategy, even if you're down a couple points in the BeyBattle (I've even seen 4-0 comebacks LOL). But your point about how winners should also receive some sort of positive psychological reinforcement is great, and your explanation of how to implement something like that without taking away the ability for the losing player to respond in a meaningful way is great. I'd love to try this.

(Sep. 10, 2016  2:39 AM)Beylon Wrote: I appreciate the idea of making switches dependent on the loser - but I think your notes on the consequences of that system illustrate how it might get messy. You basically just end up with double-blind again. The beauty of the deck rotation seems to be in revealing new layers of strategy for the game - which perpetual double-blind would negate. Particularly if you were facing a sore loser.

Completely agreed. I think we want to avoid double-blind scenarios beyond the initial selection phase of a Deck Format BeyBattle.
Quote:It's hard to say because it really depends on the volume of matches played in total, but I'd say in general that amassing three losses pre-finals for any format is the point where any player shouldn't feel bad if they don't make the finals of any event. Considering the role of luck in the selection process–in spite of how well you can scout your opponents–and the role of luck that exists in any battle, I think it's fair to say that it is plausible for any Blader to lose a couple times largely due to luck-based reasons yet still ultimately be the best Blader of the tournament if they are given the chance to be (like they are in Round Robin). Not saying it is always viable regardless of format (especially Swiss; in huge Swiss events you have to accept that to keep the event at a reasonable length, your chances of advancing to the finals go down with even two losses), three losses is where I would personally draw the line in an ideal world.

In that situation, I might say … better luck next time? Unfortunately we can’t just compensate for bad luck by adding more and more matches. Sometimes it’s just not your day. I mean, losing 3 out of 6 or 7 matches is just not a great record. Getting to the finals should be a tense, difficult path.

Quote:Precisely seeding players based on their actual rank would be the goal–it only took me a few minutes at the experimental event in Toronto–but as long as the highly-ranked players can be seeded at the top, that would be the most important thing to ensure. If there's a group of other players you're unsure of, just seed them near the bottom like Brad said.

To avoid vague language, I would say that strictly this would be the priority for seeding:

– rank of registered attendees, collected from the website the night of/morning before an event
– existing WBO players with unknown ranks
– new players

The default “1000 BP” should be considered a non-rank; it’s just a benchmark with room to move up or down depending on a player’s real performance. It’s important to remember that BP isn’t like a currency you’re meant strictly to accumulate, but just a numerical representation of your performance compared to others. That’s why I find this language too harsh:

Quote:As I mentioned above, where this matters the most is among the highly-ranked players and making sure they are seeded at the top. In this scenario however, I would probably just follow the system of seeding based on rank. Players who are below 1,000BP are responsible for putting themselves in that position; they've "earned" that rank through poor play just like highly-ranked players have earned their ranks through good play. New players can't necessarily be assumed to be great or bad, so it would seem harsh to me to automatically assume they are bad players and seed them below any participating players who are under 1,000BP. It makes sense for them to be in between.

We need to remember that the changes we make should benefit players of all ranks and experience levels.

Quote:Yeah, I agree. The main thing we will need to determine precisely is the threshold between Round Robin, Block Round Robin, and Swiss. I mentioned earlier that 11 seems like a fair maximum for Round Robin. 12-16 seems right for Block Round Robin. 17+ seems like a better choice than 16 for when Swiss would be mandatory. In the Swiss Format rounds recommended for Magic: The Gathering, 17 is their threshold for moving up to five rounds (which to me seems like a fair minimum number of rounds for any player to be guaranteed in an event).

That would look like this:

7–11 players - Round Robin
6–10 round first stage / 4 matches deck rotation

12–16 players - Block Round Robin
5–7 round first stage / 4 matches deck rotation

17+ players - Swiss
5+ rounds first stage / 8 matches deck rotation

It kind of sucks that there are less rounds the more players you have, but at least the top 8 in Swiss is pretty forgiving.

Quote:The tricky part is determining how we are going to determine the 4/8 finalists for our Single Elimination finals for every event without resorting to tiebreakers. You can use this page to run simulations. Tiebreakers seem inevitable in many scenarios, but we want to try and minimize those from happening. I'm not sure what the answer is yet because there's just so many different possibilities depending on the number of participants and if the paired down match-ups result in wins for the highly-ranked players or if they result in losses for those players.

Tiebreakers are kind of an inevitability when you’re explicitly crafting formats that allow every player to play as many matches as possible. But the current system of just tacking more battles onto the first stage is frustrating for everyone. While I’m sure some people will be disappointed that they’ll fail to advance based on tiebreaker criteria, that criteria at least makes sense — if you’re tied with a player who beat you earlier, it makes sense that they would advance over you. We failed to properly account for tiebreakers when we first designed our tournament systems, so now we’re making it right.

So it seems this is what we want to try for the next text:

First stage

– Return to first-to-3 wins
– Use tournament format criteria listed above

Deck rotation

– Players present decks at the beginning of match, then select first bey in secret
– @[Beylon]’s suggestion: “First the winner openly switches, then the loser switches in response.” What if the winner chooses not to switch? Can the loser still? It must be that way, right? Or the winner could lock the opponent into a terrible matchup.
In response to your Deck rotation question, @[Bey Brad]:

Overall, I just think that anyone should be allowed to switch whether or not they have won the first battle. There should be a 30-second pick rule implemented in there, though. I feel as though trying to give an advantage to a person that has taken a loss isn't fair to the winner. Whether or not the person has won by luck or skill, they shouldn't be punished for winning. Anything that doesn't keep it fair in the sense of a single battle is not right.
It's not about punishing the winner, it's about forcing both players to constantly change strategies and think ahead beyond just the current battle. I've played this format quite a bit now and I've never felt, nor observed anyone seeming to feel, like they were being punished for winning. Of course, if someone does feel that way, now would be the time to say so haha
(Sep. 12, 2016  1:28 AM)Bey Brad Wrote: In that situation, I might say … better luck next time? Unfortunately we can’t just compensate for bad luck by adding more and more matches. Sometimes it’s just not your day. I mean, losing 3 out of 6 or 7 matches is just not a great record. Getting to the finals should be a tense, difficult path.

Absolutely! Maybe you misunderstood my post: I meant to say that it is having only two losses and not having a chance of advancing that would be a hard pill to swallow given the nature of luck in Beyblade. Losing three is not a great record, I agree (which is why I said "I'd say in general that amassing three losses pre-finals for any format is the point where any player shouldn't feel bad if they don't make the finals of any event"). Of course, that depends a lot on the number of matches played too; you're right that we unfortunately can't add in more and more matches.

I'm just saying that the ideal situation is something like Round Robin where you actually have a chance to make a comeback. As the number of participants grows however, the number of rounds that can reasonably be played becomes lower, which means the path to the finals becomes naturally more difficult and unforgiving. It's unfortunate that this is the case because the results inevitably become less precise in some ways with less total matches played by each individual player (thereby increasing the chances that someone who has a couple unlucky losses is eliminated from the event), but it's the only way to keep the events whether large or small at a similar length of time to complete.

(Sep. 12, 2016  1:28 AM)Bey Brad Wrote: To avoid vague language, I would say that strictly this would be the priority for seeding:

– rank of registered attendees, collected from the website the night of/morning before an event
– existing WBO players with unknown ranks
– new players

The default “1000 BP” should be considered a non-rank; it’s just a benchmark with room to move up or down depending on a player’s real performance. It’s important to remember that BP isn’t like a currency you’re meant strictly to accumulate, but just a numerical representation of your performance compared to others. That’s why I find this language too harsh:

Quote:As I mentioned above, where this matters the most is among the highly-ranked players and making sure they are seeded at the top. In this scenario however, I would probably just follow the system of seeding based on rank. Players who are below 1,000BP are responsible for putting themselves in that position; they've "earned" that rank through poor play just like highly-ranked players have earned their ranks through good play. New players can't necessarily be assumed to be great or bad, so it would seem harsh to me to automatically assume they are bad players and seed them below any participating players who are under 1,000BP. It makes sense for them to be in between.

We need to remember that the changes we make should benefit players of all ranks and experience levels.

Maybe my language here was a bit too harsh (there's probably a better word than "bad players"), but is my point about not necessarily assuming new players skill level by giving them the bottom seeding not valid? It's possible a new player could be a very skilled player, or it's possible they might actually be a newbie without much experience playing. We don't know either way, so the idea of placing them between people above 1,000BP and people below 1,000BP made sense to me personally.

It makes more sense to me to use the real performance of players below 1,000BP to place them near the bottom, just like we do for the real performance of players above 1,000BP to place them closer to the top. New players being in between the two because they don't have any real performance and because we can't assume their skill level makes sense to me logically.

Although, this does have an interesting effect on the match-ups that would be given by Challonge that does kind of sway me away from this perspective. For example (8 Player Swiss, just for demonstration purposes):

Tournament A - New Players Seeded at the Bottom
1. Rank 1 (1300 BP)
2. Rank 2 (1275 BP)
3. Rank 3 (1250 BP)
4. Rank 4 (1200 BP)
5. Rank 20 (995 BP)
6. Rank 23 (970 BP)
7. New Player 1
8. New Player 2

First Round Swiss Pairings:
Rank 1 vs. Rank 20
Rank 2 vs. Rank 23
Rank 3 vs. New Player 1
Rank 4 vs. New Player 2

-

Tournament B - New Players Seeded Between Over 1000BP and Below 1000BP Players
1. Rank 1 (1300 BP)
2. Rank 2 (1275 BP)
3. Rank 3 (1250 BP)
4. Rank 4 (1200 BP)
5. New Player 1
6. New Player 2
7. Rank 20 (995 BP)
8. Rank 23 (970 BP)

First Round Swiss Pairings:
Rank 1 vs. New Player 1
Rank 2 vs. New Player 2
Rank 3 vs. Rank 20
Rank 4 vs. Rank 23

What we can see from this is that if the new players are seeded at the bottom (Tournament A), the highest seeds will be paired with the players who are below 1000BP. This means that in the first round they get to play players which they could have potentially seen play before, which I would argue may be more advantageous than playing a complete unknown player who may be really good or who might be a newbie.

If new players are seeded in between over 1000BP players and below 1000BP players (Tournament B), the top seeds get to play the new players in the first round, which is arguably a harder match-up.

Not exactly sure where I'm going with this, but just wanted to point out that although I do think new players should be seeded in between over 1000BP and below 1000BP, it's possible doing this could generate a disadvantage for the top seeds in an event since they would have to play them first without any information.

It's also a question of whether we want new players to be immediately thrown into the fire against the top seeds or if they should be paired against slightly lower top seeds. Maybe won't make a huge difference a lot of the time, but there is a difference philosophically at least.

(Sep. 12, 2016  1:28 AM)Bey Brad Wrote: That would look like this:

7–11 players - Round Robin
6–10 round first stage / 4 matches deck rotation

12–16 players - Block Round Robin
5–7 round first stage / 4 matches deck rotation

17+ players - Swiss
5+ rounds first stage / 8 matches deck rotation

It kind of sucks that there are less rounds the more players you have, but at least the top 8 in Swiss is pretty forgiving.

Yeah, it does suck ... but I can't think of another way to ensure all events run for a decent amount of time while still being easy to organize.

(Sep. 12, 2016  1:28 AM)Bey Brad Wrote: Tiebreakers are kind of an inevitability when you’re explicitly crafting formats that allow every player to play as many matches as possible. But the current system of just tacking more battles onto the first stage is frustrating for everyone. While I’m sure some people will be disappointed that they’ll fail to advance based on tiebreaker criteria, that criteria at least makes sense — if you’re tied with a player who beat you earlier, it makes sense that they would advance over you. We failed to properly account for tiebreakers when we first designed our tournament systems, so now we’re making it right.

So it seems this is what we want to try for the next text:

First stage

– Return to first-to-3 wins
– Use tournament format criteria listed above

Deck rotation

– Players present decks at the beginning of match, then select first bey in secret
– @[Beylon]’s suggestion: “First the winner openly switches, then the loser switches in response.” What if the winner chooses not to switch? Can the loser still? It must be that way, right? Or the winner could lock the opponent into a terrible matchup.

I wouldn't say it's frustrating for everyone because of the "second/third/nth encounter" element that Wombat mentioned; playing the same player multiple times in an event is actually quite thrilling and challenging. But it does unnecessarily prolong events, unfortunately, so I do agree with the new tiebreak procedures (could have used them in the official event last week when I beat both of the people I was tied with in the RR LOL).

I'm not voting for the revealing of the deck just yet, but yeah that sounds right. As for Beylon's suggestion, I would assume that the loser can still switch.

Bey Brad Wrote:It's not about punishing the winner, it's about forcing both players to constantly change strategies and think ahead beyond just the current battle. I've played this format quite a bit now and I've never felt, nor observed anyone seeming to feel, like they were being punished for winning. Of course, if someone does feel that way, now would be the time to say so haha

Yeah, I agree. Should have mentioned this in my other post too; I haven't heard anyone feeling punished for winning in all of my experience with the format so far either. But I am definitely willing to try Beylon's proposal because it does make logical sense in spite of the lack of complaints from winners we've experienced so far.
I just think seeding things that way creates a lot of complication for an edge case, and I would really like to avoid introducing complexity where it can be avoided. Already asking organizers to seed based on rank is complicated enough, but having to tuck new players in between scores is an unnecessary wrinkle.

I don't know if revealing the deck will be better or worse, but I think we've done enough events with concealed decks to know how it feels, so it's perhaps worth trying the alternative now? I'm personally also in favour of keeping the decks hidden, instinctually, but I'd love to see how people actually find it in practice.
(Sep. 11, 2016  11:34 PM)Kei Wrote:
(Sep. 09, 2016  8:57 PM)Wombat Wrote: I haven't tried simulating extended Swiss yet, but I imagine pairing past the "ideal" round would require some questionable improvisation and the extra rounds might jumble up the scores too much. Like Kei said, given the short and dynamic nature of a Beyblade battle, I still think that (Block) Round Robin is the best options for smaller events. I think 16 is the magic number where you should switch to Swiss.

Yeah, I agree. The main thing we will need to determine precisely is the threshold between Round Robin, Block Round Robin, and Swiss. I mentioned earlier that 11 seems like a fair maximum for Round Robin. 12-16 seems right for Block Round Robin. 17+ seems like a better choice than 16 for when Swiss would be mandatory. In the Swiss Format rounds recommended for Magic: The Gathering, 17 is their threshold for moving up to five rounds (which to me seems like a fair minimum number of rounds for any player to be guaranteed in an event).

The tricky part is determining how we are going to determine the 4/8 finalists for our Single Elimination finals for every event without resorting to tiebreakers. You can use this page to run simulations. Tiebreakers seem inevitable in many scenarios, but we want to try and minimize those from happening. I'm not sure what the answer is yet because there's just so many different possibilities depending on the number of participants and if the paired down match-ups result in wins for the highly-ranked players or if they result in losses for those players.

(Sep. 12, 2016  1:28 AM)Bey Brad Wrote: That would look like this:

7–11 players - Round Robin
6–10 round first stage / 4 matches deck rotation

12–16 players - Block Round Robin
5–7 round first stage / 4 matches deck rotation

17+ players - Swiss
5+ rounds first stage / 8 matches deck rotation

That sounds more or less along the lines of what I was imagining, but the reason I put the cutoff point at 16 was because it's one of those binary type multiples (not sure of the exact term).

[Image: hHhGBFZ.png]

I made that example of the binary thing I was talking about, so it's easier to visualize the pattern I noticed with the win records. Also, for smaller Swiss events like the one depicted I'm not a huge fan of taking the top 8 for the finals - people keep saying that "the path to the finals should be difficult and not guaranteed for even experienced players", and the top 8 is literally half of the participants. I get that single elimination finals thins the crowd a lot faster, but you would need a seriously huge event to make the top 8 a true measure of the "best" players. I'm still a fan of taking the X-0 person, and everyone who is X-1 (in the cases of weird bye situations, the player with the greater amount of wins advances. 4-1 would advance over 3-1). And with many of the 2-2 players not having met and therefore having no "history" in the tournament, you would have to resort to an additional tiebreak round with the six of them, taking the three winners as well.

Disclaimer (Click to View)

(Sep. 12, 2016  2:19 AM)Kei Wrote: Tournament A - New Players Seeded at the Bottom
1. Rank 1 (1300 BP)
2. Rank 2 (1275 BP)
3. Rank 3 (1250 BP)
4. Rank 4 (1200 BP)
5. Rank 20 (995 BP)
6. Rank 23 (970 BP)
7. New Player 1
8. New Player 2

First Round Swiss Pairings:
Rank 1 vs. Rank 20
Rank 2 vs. Rank 23
Rank 3 vs. New Player 1
Rank 4 vs. New Player 2

-

Tournament B - New Players Seeded Between Over 1000BP and Below 1000BP Players
1. Rank 1 (1300 BP)
2. Rank 2 (1275 BP)
3. Rank 3 (1250 BP)
4. Rank 4 (1200 BP)
5. New Player 1
6. New Player 2
7. Rank 20 (995 BP)
8. Rank 23 (970 BP)

First Round Swiss Pairings:
Rank 1 vs. New Player 1
Rank 2 vs. New Player 2
Rank 3 vs. Rank 20
Rank 4 vs. Rank 23

Wait hold up is this how we are supposed to seed players? As in 1 vs 5, 2 vs 6, etc. instead of 1 vs 8, 2 vs 7, etc.? Either I missed this detail or it wasn't explained very clearly. In the example above I had it as the highest seed playing the lowest seed, and readjusted the seeding every round.

(Sep. 11, 2016  11:34 PM)Kei Wrote:
Bey Brad Wrote:It's not about punishing the winner, it's about forcing both players to constantly change strategies and think ahead beyond just the current battle. I've played this format quite a bit now and I've never felt, nor observed anyone seeming to feel, like they were being punished for winning. Of course, if someone does feel that way, now would be the time to say so haha

Yeah, I agree. Should have mentioned this in my other post too; I haven't heard anyone feeling punished for winning in all of my experience with the format so far either. But I am definitely willing to try Beylon's proposal because it does make logical sense in spite of the lack of complaints from winners we've experienced so far.

Ah, that makes more sense now. The only problem I can maybe see with this is that most people will probably almost always revert to their "safe" combo upon winning, which might cause some kind of matchup loop between only a few combos.
Quote:Wait hold up is this how we are supposed to seed players? As in 1 vs 5, 2 vs 6, etc. instead of 1 vs 8, 2 vs 7, etc.? Either I missed this detail or it wasn't explained very clearly. In the example above I had it as the highest seed playing the lowest seed, and readjusted the seeding every round.

Yep, this is how it works (and how Challonge does it). It's the kinda thing we would detail in our new event guide (that we're using this data to build).

Then after each round players with the same (or most similar) records during that tournament play against each other.
Yesterday I attended Let the Battle Begin Onto the Future! in Raleigh, North Carolina, and decided to share some of my thoughts after playing the experimental format.

Swiss Format
I actually enjoyed Swiss format a lot. I prefer it over Round Robin because I feel that Round Robin goes on a bit to long if you have 9-10 participant while swiss will still give you enough rounds to play, but it won't feel overwhelming. We had 9 people so we ended up doing 4 rounds of swiss and a cut to top 4. The only problem is uneven numbers with swiss gets kinda iffy with low amounts, because it appeared some people were slightly upset to get the bye and only ended up playing 3 matches. Other than that it seemed fine to me, it wasn't time consuming and it definitely makes it challenging to top when every round you go against someone the same score as you.

Seeding
The idea for seeding was awesome, it definitely rewards you for actively going to events and performing well. Definitely something I loved and really want to keep in.

Deck Rotation
Honestly in our finals we didn't see to much switching of beyblades in this format, but I really do like the idea of being able to change mid match if you got an unlucky guess originally. It definitely adds a different aspect to the finals and it will be awesome in the future. I also thought getting to 5 points was really fair, because it can eliminate variance which is what we want. Now what I don't think is correct is to have Bursts as 2 points. I played a stamina mirror match in the finals and a few random bursts take me from being up 3-0 to losing 6-3. I understand that it could promote attack, but it adds to much luck and honestly losing like that puts a bad taste in your mouth. Burst Finish definitely needs to be taken back to 1 point. I don't know where to place Knock outs though. We didn't really have much attack usage in our finals, but random KO's could happen in stamina match ups which we witnessed. Now what could promote attack (and this is just an idea) is to make KO's 1.5 points. We should try this because it doesn't punish you unless you get randomly knocked out a few times in stamina matches since the .5 wont make a difference unless it happens consecutively, but it still rewards someone for using attack types because 2 KO's would be equivalent to 3 OS's or Burst Finishes (if we ended up putting Burst Finish down to 1 point). It's just an idea, and I'm not sure how well it would work in affect, but I feel were so close to actually making burst a bit more skillful game and it could be worth trying.

No Banned Parts
Either due to the lack of players owning a Odin or after seeing EthanQ42's Deathscyther Heavy Defense burst Stormscorpio1's brand new Odin Heavy Defense 3 times in a row in Round 1 made an absence of Odin occur. We all felt it was definitely not what it was once made out to be during the tournament and the little testing we had before the event. I think unbanning Odin is definitely fine.

Tiebreakers
Although this didn't come up in determining which players entered the finals, I really like the idea of using Challonge's system to calculate who should advance to the finals. It makes our game actually resemble other games because we don't have all these matches we have to do when 2 players have the same record, which just look at the tie breakers and move on. It definitely saves time and is very efficient and rewarding to the player who overall had better opponents in Swiss.

Mandatory usage of Challonge
If were all being honest here to use the Swiss system Challonge is basically needed. You don't necessarily need it, but it will make using Swiss ridiculously easy. Also using Challonge makes makes every neater and easier, your giving your community a handicap by not using it to begin with. I love this system and I know so many other communities that love it, it definitely should be mandatory to make sure the swiss format goes through fine and there is no mix ups.

These are just honestly my thoughts, and I hope other members from that event could share their's as well, this format has so much potential!

@[Dark_Mousy] @[Stormscorpio1] @[Titan Tite] @[EthanQ42]