Community eSports Sky League (March 22nd - April 26th)

[Imagine]

I wouldn't use the word random, more like unexpected, and it just makes it more exciting. 

Whichever team makes the fewest mistakes wins, it's just as much about mental toughness as skill, allows teams to surprise their enemies with unexpected tactics and tests them to see if they can react and adjust.

It's not just that, it's also maps. Some teams are going to be better on some maps than others, so having a bo3 will give you a lot more accurate count of who's better skilled overall, and not just better at a particular map.

Well that can lead to the same result as a best of one.
The first map is random-ish, and the team that's better at that map is going to win.
The losing team picks the next map, something they're generally better. They could lose it here or win it.
If the team that won the first match loses the second match, they get to pick a map they're better at and have a high chance of winning.

That first map plays the largest role in determining the overall winner. That's not saying teams can't come back and get two wins, but I would suggest that it's very uncommon. I'd have to go and check the past match videos to verify that however.

Two things:
1) We've never had teams picking maps afaik.
2) While most games end in 2-0, you're still more likely to have the team that's actually better (if for nothing else, that particular day) win out.

What you're not going to have (for the most part) is some team be really good at only sniping (or the other team being just bad at playing against it), and getting lucky with having their map being on Dunes and winning that way in a one and out.

[Captain Smollett]

I just kind of like the suspense where every match is make it or break it.  It also leads to shorter casts and shorter wait times for the teams.

[Imagine]

I just kind of like the suspense where every match is make it or break it.  It also leads to shorter casts and shorter wait times for the teams.

Don't disagree. Just saying I like to see double elimination if bo1 is the case, something which at least these group stages provide

[Thomas]

Ah, I was thinking the Leviathan where the loser picks. But after going back through the CeSports videos, the 20 or so I looked at, the loser of the first match only came back to win twice. So 90%+ of the time winning the first match means you're probably going to win the best of three.


Overall I think having a best of one is more exciting and intense for both participants and viewers, and it would be awesome to adjust the bracket portion of the tournament to reflect this, but re-apply the double elimination condition, allowing the teams who get eliminated in the Group stage a chance to come back.

By announcing the map ahead of time, like in the group stage, you remove that 'luck' factor of getting maps you're good at. You know what map it's going to be, and you can prepare for that, instead of hoping you get something you're good at and something the enemy isn't as good at. Essentially there's no "Well, they're better at these kind of maps" excuses, since both teams knew what they were getting into well in advance.

[Frogger]

I for one vastly favor the Best of 1 approach. Having one's Saturday or Sunday eaten up by 3-4 hours of Bo3s is exhausting. It's more fun both to play and to watch.

[Mr. Ace Rimmer]

I too like the idea of Bo1 double elimination. However, perhaps that is best for another tournament as to ask Urz to redraw the brackets now for a double elimination just seem a little bit selfish of us considering the work already put in.

On that note though, I'm struggling to see why we have two top seed matches in round one with a second seed exclusive match. Surely it was possible to have only one set of top seeds facing each other in R1.

[Imagine]

I mean as for group stages you already had double elimination. No need to bring the teams back for fairness sake.

[N-Sunderland]

On that note though, I'm struggling to see why we have two top seed matches in round one with a second seed exclusive match. Surely it was possible to have only one set of top seeds facing each other in R1.

Fully agreed. To me it looks like E2 is actually a better seed than E1 because of all that.

[Thomas]

I wouldn't mind sacrificing some time to use my sub-par GIMP skills to redraw the brackets. Actually might do it for fun anyways now. xD

Not sure what you mean by the top seed and second seed stuff.

As far as I can tell it would be to cut down on the number of matches in R1, since they're currently slated for best of three. Which could mean up to 12 matches, with only two or so being steamed at a time. If you bring in all the teams at once, that'd be up to 18 matches. Already it's stretched pretty thin.

I think it was set the way it was for the time constraints as well as some fairness. the X2 teams have already lost one match, where the X1 teams haven't. However, the E group fought more matches than the other groups (or at least E2 will).


Now if it was switch to best of one, you cut down the possible number of matches by 3. So we'd only have 4 matches in R1, and only 2 in R0. Which should conclude fairly rapidly, and opens up the possibility of adding back in the 'lower bracket'  to keep the party hopping.

That being said, with some teams already dropping out, and some more likely to drop out, and the weird size of group E; creating a lower bracket would be both a hassle, and probably have to be done after the next group matches. I'm still for it, and will probably draw up some concepts after next weekend, but I can completely understand not including for the sake of simplicity.

[Urz]

Bracket seeding was created with the following conditions:

"1" seeds get seeded into round 1
"2" seeds get seeded into round 0
"E2" is upgraded to a "1" seed since they potentially have to play additional matches to advance
No two teams from the same group will play in the bracket before the finals


As for best of ones in the bracket stage, there are two primary reasons that will not happen. The first is that I will not change the structure of a tournament after it has already started (an exception would be made if the tournament would otherwise be cancelled or not finish). Teams have played matches with the understanding that they will either advance or be eliminated during the group stage. To change that after such determinations have already been made would compromise the fairness of the event; in this case it would be a decision biased towards the teams who were eliminated last weekend.

Secondly (and something which has already been brought up in this thread), a best of one can easily be decided by a game bug, advantageous map roll, or a single mistake. "Best of three" format mitigates those scenarios, none of which are good for the state of competition. This is the reason all my events have been Bo3 in bracket stage despite Bo1 being the precedent prior to CeSports, and since Sky League is intended as a high level competition to determine who is the best in the game, matches in bracket stage would be Bo3 regardless.

That being said, a thought I had after Saturday's matches was that if the smaller weekly events return (eg Sunday Rumble), they could be best of one. As they are lower stakes and you can just sign up for the next week if you're eliminated, it would be a reasonable compromise to increase turnout and reduce the fatigue of all involved.

Feedback is definitely welcome, as if there is another season of Sky League, it will certainly go through some changes then.

[Thomas]

In response to those reasons:

-The tournament was already dramatically changed. Although it hadn't gone into the matches yet, the sign-ups were closed and it was less than a week before the matches began. It had already 'started'. Essentially the format and rules we agreed to when everyone signed up were tossed out. That's like buying a car and instead having the dealership give you a truck. We do understand that it did have to happen because of the circumstances, but it does make that first line of reasoning fall apart.


-Most of the second points were debunked. Almost always if you win the first match, you're going to win a second one. Very very rarely does a team lose the first match to turn it around for a double win. An advantageous map has nothing to do with best of three or best of one, that's just from having random maps. Instead they should be announced ahead of time, as this removes the luck factor entirely. Teams can then prepare for that map and their enemy, leading to a much higher level of strategy and gameplay. Bugs and lag are treated the same in Bo3 and Bo1, and they effect these matches the exact same way. Having more matches does not necessarily mean that they'll have a smaller impact. Big issues result in a pause or reset, and there's not much you can do for lag. You can see how small their impact is on the outcome by remember that whoever wins the first round is 90% more likely to win the overall match regardless.




I think the big points in favor of best of one is that it's a lot more fun to watch, and a lot more fun to participate in. Double elimination was the original intent of the tournament, and what a lot of teams were looking forward to. I really don't mind double or single, but if it was to switch to something more exciting than Bo3, there would be some time to fill, and it would be a lot of fun to see those great teams try to work their way back up to ultimate victory. Instead of watching the same teams fight 2 or 3 times in a row.

[Urz]

The distinction between changing the tournament structure before or after matches have been played is important. Once matches have already been played, we know which teams will be affected more positively than others, and thus any change would be potentially biased towards the teams who benefit the most from it. The parenthesized exception is also important here, as if I hadn't made those changes prior to the tournament starting, it would have been cancelled outright. Comparing participation in a free tournament run for your enjoyment to a motor vehicle purchase is not a particularly apt analogy either.

The rest of your points are primarily based on flawed data interpretation, but I will indulge you with a more detailed explanation. Those '20 or so videos' you pulled from the CeSports channel were likely from the more recent weekly tournaments. Those are events in which any team can sign up, show up, play, and be knocked out in a couple hours. With Sky League, by contrast, before we even get to the bracket half the teams have been filtered out. It is also a much longer event where each match has more significance. A more apt comparison would be the last major tournament, Anvalan Conflict, where in four of the eleven matches the team who lost the first game came back to win the set (37%).

Now to address your points specifically...

On maps: your argument is that allowing both teams equal time to prepare means both teams will be equal on every map. That is incorrect and discounts how certain maps favour certain styles of play.

On bugs: if somebody disconnects during a key engagement and their team wipes because of it, I'm not sure how you can say that has the same impact on both a Bo3 and Bo1. To try and explain this in the simplest way I can: 2 deaths out of 5 total versus 2 deaths out of 10 or 15 total (or 1 game out of 1 versus 1 game out of 3).

[Thomas]

The thing about this tournament is that it's split into two sections, group and bracket. It's too late to change the group stage, since that's already happened. However, changing the bracket stage from bo3 to bo1 shouldn't be capable of giving one team an advantage over another. I'd say that the purchase of a vehicle is a decent analogy. Instead of money, we're committing our time and effort to participate in the tournament and make it successful, which provides entertainment for your viewers. If you want we can change it to signing up for a marathon only to have them change it to a 100-meter dash.

The data I used did come from the more recent videos where teams like the Ducks, Sac, The Gents, CSR, etc participate. Or 'the same exact groups that will be in this tournament'. There's also a lot more data to pull from. But it really depends on what data is chosen to be used and chosen to be ignored that really determine which way the statistics will go. And of course one of the biggest factors is the map luck.

maps: If you know what map is coming, and you know your playstyle doesn't work well on it, this gives you ample time to change or improve your playstyle for that map, which leads to a better match. This might actually be the biggest contributor to the belief that we need a best of three to determine who should win. If both teams know how things are going to go down, there's no excuses for 'well, my playstyle doesn't work well on this map', since they had the time to fix that.

bugs: It really depends on the bug. Most bugs and issues generally result in a pause of the match; and occasionally need a restart. Disconnecting in the middle of an engagement isn't all that common, at least as far as I'm aware. More importantly, bugs affect matches, not just kill counts. And even in Bo3, they try to have the most balanced matches possible. Pausing, resetting, etc when someone suffers a bug on that scale. This causes the matches to go much longer, and generally happens more than once. With a best of one, the total match time is much shorter since you only have to play one match. Making adjustments for critical bugs easier to accommodate.

[Urz]

At this point we just disagree, so your opinion has been noted.

[Omniraptor]

I would agree that knowing the map in advance would make the game more fun because it allows for deeper planning.