EPTLet's get the numbers out of the way first before we talk about anything else.
Total donations towards the next tournament - $3904.66
Money left over from last time $1131.23
SCI5 = Fully Funded.
In fact, we're overfunded at this point. Due to the gap inbetween SCI3 and SCI4, there's been a few months of streaming and monthly subscriptions inbetween that. Basically SCI5, SCI6 and some of SCI7 are all fully funded now based on my estimations and projections. The only other costs we're incurring right now are Apollo's expenses (since we paid for his flight and such) and paypal/wire fees to pay players their prizemoney. Our graphic designer is paid a monthly salary anyway for the work he does on my main channel so that's not coming out of the cash we raise for SCI. So what does that mean?
Well it means we could easily do at least one tournament where we do not take any donations at all. That potentially means more cash for the players via the tip-jar and also ensures that the burden of paying for these events is taken away from donors. We're not yet at the point where we can guarantee a monthly tournament that is fully funded by the channel. The subscriptions count is healthy, but as you're well aware, CPM and ad revenue are all over the place on Twitch right now. In fact, January is the worst month for CPM across the board. Myself and other Youtube partners always dread January, because it flat-out sucks for CPM. Nobody wants to advertise because people don't have cash to spend after Christmas, so we all get hit pretty hard. While the donation count this tournament was incredible, the ad revenue was not. Factors that contributed to this are as follows.
1) January CPM
2) Twitch's continued issues with getting "fill", meaning the eCPM (effective CPM) is lower because less people see ads
3) System problems that caused ads to not actually play during parts of the tournament, regardless of whether the command was given.
Twitch is not to blame for most of this. They can't change the way CPM varies on a predictable yearly basis and they are trying to sell more advertising and get a better fillrate. It's in their best interests to do that so I have no doubt that they're not slacking on the matter. The system problems that occurred are of course something that need to be looked at. I have no idea what caused the issues, but it hurt us quite a bit. The full figures aren't in yet but ad revenue from this tournament is noticeably lower than previous events, even though this one blew the number of viewers we had completely out of the water. As far as I'm concerned, one of Twitch's priorities should be to ensure the ad system is rock solid and doesn't break down in the middle of a live event.
Anyway, it's not that big a deal, the money we lost from the ad issues was more than covered by the extra donations and additional paid subscriptions we gained so I'm not going to stress out over it. I'm still not 100% sure that I can fully fund every tournament in future myself based on Showcraft streams and subscriptions, but it's getting to that point and that's very good news. Severing our reliance on community funding permanently would be the ultimate goal of this thing.
Speaking of numbers, let's talk tips! Among the many records broken during SCI4 is the amount of money given as tips to players.
Socke - $187.50
Ret $252.50
ClouD $133.01Grubby $545.50Stephano $371.94ThorZaIN $570.44DarKFoRcE $793.45NightEnD $224.78------
Total - $3079.12
Well, that almost triples the amount that was given during SCI3 when the system debuted. It also breaks the record for most-tipped player. TLO gathered $455 in tips last time, this time Darkforce blew that out of the water with almost $800. What I really like about Darkforce getting that much is that out of our lineup, he was one of the least well-known. He has not had prominent results lately and even though he went out in the group stage he showed 2 of the most incredible games on Taldarim we've seen in a very long time. His astonishing comeback against Nightend and absolute refusal to give up was inspiring to watch. I want to talk more about the tipping system and the ethics of giving money to players that isn't proportional to their tournament results a little later on.
Now the big numbers.
* 14617 SCI2 Day 1 average concurrents
* 15386 SCI2 Day 2 average concurrents
* 189730 SCI2 Day 1 Total views
* 299415 SCI2 Day 2 Total views
* 14789 SCI3 Day 1 average concurrents
* 14109 SCI3 Day 2 average concurrents
* 156531 SCI3 Day 1 Total views
* 255443 SCI 3 Day 2 Total views
* 23376 SCI4 Day 1 average concurrents
* 33601 SCI4 Day 2 average concurrents
* 449405 SCI4 Day 1 Total views
* 445978 SCI4 Day 2 Total views
Twitch has also significantly improved their analytics system so for the first time I can bring you some new statistics.
* 198120 SCI4 Day 1 Unique viewers
* 184462 SCI4 Day 2 Unique viewers
* SCI4 Day 1 hours cast - 13.5
* SCI4 Day 2 hours cast - 8
* SCI4 Day 1 Maximum concurrent viewers - 32152
* SCI4 Day 2 Maximum concurrent viewers - 49337
* SCI4 Day 1 total hours of SCI4 watched - 309734
* SCI4 Day 2 total hours of SCI4 watched - 254804
Stats are fun, we like stats. Well I think I can safely say SCI4 crushed every previous SCI tournament in terms of raw numbers. The tournament was a run away success. Views-wise it equalled HSC4 at times and came close to beating it's maximum concurrent views. From memory, the only online tournament to manage more live viewers than this was TSL3 and our event gained more viewers than many large off-line tournaments. Yeah, it's safe to say I'm happy with how things turned out in that respect and greatly humbled by the support the community has shown SCI4. You might think it's a little odd that Day 2 had fewer unique viewers than Day 1 when we had more concurrent viewers during Day 2. That's simply down to the fact that Day 1 was over 50% longer, allowing it to reach more timezones over the course of it's duration. We had more Australian viewers than usual for instance because we started earlier and gained the American viewers as the day went on. SCI's appeal is unquestionably global at this point. Perhaps next time we will get a Korean commentary team too.
Onward and Upward
It's safe to say that after SCI3 I was not happy. I felt I'd compromised the tournament by inviting too many "fan favourites" rather than focussed purely on recent results and performance. This resulted in a lot of 3-0 games and while every player had the potential to win, only a fraction of them realistically had a shot at it based on their current form and results. After SCI3, I was determined that the next tournament would be a massive step up in terms of game quality and skill on display. I immediately went into planning the SCKI, SHOUTCraft Korean Invitational. Going through a contact in Korea I started making contact with various teams and gathering a roster which included foreigners training in Korea (including HuK, as we teased at the end of SCI3) and competitive Korean Code-S level players. Unfortunately we ran into a lot of problems. Communication is obviously an issue and was even worse back then than it is now. Korean teams have a broader global outlook now than they did several months ago and it's easier to talk to them now. Some teams flat-out refused to take part, others offered their lowest profile and least skilled players. Some teams did play ball and we have a few extremely talented players ready and willing to play in a future SCI, but as we were continuing to try and setup an already delayed tournament, our point of contact left Korea and we lost our link to those teams. After that we entered the "month of hell" where myself and Apollo were constantly travelling and casting and there was absolutely no time to even consider an SCI tournament.
I abandoned the idea of the Korean tournament until 2012 and built a roster of strong foreigners. We ran into problems there too because a couple of those guys ended up going to Korea to train (IdrA being one of them as many of you suspected) and the IEM Kiev qualifiers pretty much killed any chance of us getting SCI4 done before the New Year.
Eventually after a lot of consideration we came up with our SCI4 lineup which ended up being all-European this time around. It consisted of high profile players, recent successes and competitive players who we knew could give a good fight against anyone in the line-up. I'm happy to say that I think our choices were correct. The group stages showed that we put together the most balanced line-up yet. Every single player won at least one series in the group stages, there was not a single white-wash where it was obvious that one player had a massive upper-hand against the other. Even Stephano, the eventual victor, dropped a map against EVERY player in his group. When we said "every group is a group of death", we meant it. Game quality was, in my opinion, the highest it's ever been. There are so many memorable series I don't know where to begin. Darkforce vs Thorzain was an unbelievable clash where Thorzain wore down and eventually broke the spirit of Darkforce over the course of 50 minutes, resulting in a second game in which Thorzain quickly destroyed his opponent with little effort. Darkforce vs Nightend played host to one of the greatest comebacks and underdog stories I've ever seen. On the verge of defeat, Darkforce executed a series of tactically brilliant defenses that resulted in the destruction of his opponent. Neural parasite, clutch transfuses, that series had it all. What about Thorzain throwing the TvP metagame rulebook out of the window and demonstrating the viability of late-game reaper compositions to break a zealot-heavy maxed deathball? How about Grubby's creative finesse on display time and again as he defeated Ret, who up until that point looked unstoppable, going 2-0 twice in his group. What about Thorzain vs Ret, a matchup Thorzain was expected to lose based on his past history with the King of Drones, which resulted in a dramatic tense final game? The list goes on and on, every player deserved to be there and every player proved that in their play.
I wouldn't even call the grand final disappointing. Despite a 4-0 sweep for Stephano, the games were not one-sided. Game 1 demonstrated some of the things we love most about long macro games as well as some of the things we hate. One way or the other, it was memorable. At the end of the day though, Stephano showed masterful ZvT once again, even getting the opportunity to go back to his original style which the Koreans have been trying to pick apart (and succeeded at times). Stephano shows us what you can do with lings, don't be surprised to see a lot more of that on the ladder over the next few weeks.
Overall I feel my choices were correct in terms of the roster.There are always people we want to invite but can't because we only have 7 spots (plus winner). I do feel somewhat guilty for not bringing DDE back, however we had a lot of feedback that indicated bringing back 1st and 2nd place from the previous tournament was not welcome and resulted in more stagnant and less interesting rosters. Hopefully we'll see DDE again in future tournaments, he's been a bit off his game lately due to various RL issues.
Korea?
When are the Koreans coming? The most frequently asked question. SCI5, is the answer. We are dedicated to making it happen this time around. We have impressive stats to show the Korean teams which will draw interest. We have a good format and a solid pedigree. The plan for SCI5 right now (subject to change just like every other time) is to have a "Welcome to Korea" tournament where 4 foreigners and 4 Koreans battle for supremacy. We want to take the level of skill up a notch once again by including the best we can get. No doubt some you will be shouting "MVP! NESTEA! MMA!" right now and while we'd love to do that, you've gotta bear in mind a few things. Those guys are important, really important, the most important as far as their teams are concerned. They are EXTREMELY selective about the tournaments they play in and as regards to money, they aren't short of it. You can't tempt players that have earned between $135k-255k with a $5000 prizepool and being an online tournament based outside of Korea puts us in a position where we don't have a lot of bargaining power. That doesn't mean we can't get great players and we've got a few in store already. It should be easy to guess some of the players we're inviting for SCI5, though as to how many are able to accept is a different matter. It's an open secret that Naniwa wants in, because he posted in the thread about it and we're certainly not going to refuse his request. Expect Naniwa in SCI5 even if we don't end up getting the Korean-server tournament together, assuming he doesn't have a sudden change of heart.
Format
While the first day was very tiring since we cast for 13 hours with no scheduled breaks, myself and Apollo are happier with the group-based format than the single elimination BO5 bracket. As far as I'm concerned it shows people more of the matches they want to see, generates a more accurate result and is more player friendly. There is one thing I need to address and that is how we determined who went through in a tie-break situation.
Based on the rules we gave out to players, we used the following priority order to determine the 1st and 2nd seeds from the group stages.
Series won (obviously)
Performance ratio (percentage of maps won vs maps lost)
Head-to-Head score
Head to Head is the lowest priority, it's a last resort in-case we have absolutely identical scores otherwise. I really dislike head-to-head as a determining factor, which is why it's lowest down the priority list. This is common in large tournaments, they tend to consider map-score more important than head-to-head. The difference in our tournament from some of the majors is how we approach map-score. There are a few ways to look at it. One of the most common is simply "maps won", which takes priority over "maps lost". If we used this system, we could have ended up with the following result in Group B.
If Grubby had beaten Ret 2-0, he would have ended up with a score of 5-3, having conceeded a map to Socke and lost 2-1 to ClouD. This would have left Ret with a score of 4-2, beating Socke and ClouD 2-0 but losing to Grubby 2-0. In a "maps won" priority system, Grubby would have advanced as 1st seed to play Thorzain and Ret would have advanced as Seed 2 and played Stephano. However, that is not the system we used. In our hypothetical example, Ret earns a 66% performance ratio based on his 4:2 record in the group. Grubby earns a 62.5% win ratio with his 5:3 record. In this scenario, Ret goes through 1st seed regardless of his result against Grubby. Grubby ended up winning but dropped a map against Ret, resulting in a 5-2 record for Ret and a 5-4 record for Grubby, a clear 1st seed group victory for Ret.
Why am I even going through those hypotheticals? Because there was some controversy regarding our statements on the stream about who would go through as what seed in what scenario. Based on the system we used, we were correct. As to whether or not our system is better or worse than the "maps won" priority system, I honestly don't know. I feel that performance ratio rewards the more solid, consistent player. 2 2-0 victories are significant, they show a clear dominance of one player over another. I think that a player who can achieve that should have an edge when the scores are tallied, as opposed to a player who wins his games, but drops maps every time. We are open to feedback on this issue though, as I feel it is contentious and worthy of discussion.
Tipping the Balance
The tipping system of SCI is unique. No other tournament does this. We devised it because SCI originally came about to help the pros while also providing a great tournament for the fans. The tipping system allowed fans to directly say to a player "I enjoyed what you just did, let me show you my appreciation". It is a morale boost to players to see fans directly willing to give them a contribution. SCI3 saw TLO win big in terms of tips, but this time around we saw one of the least known players in the line-up, Darkforce, walk away with an astonishing amount of tip money. A thread began on Teamliquid about the tipping system and whether other tournaments should look into implementing it. Opinions are strong and tempers have run high, with some outright condemning the system as anti-competitive, open to abuse and match-fixing and downright stupid. I believe the words "hurting/killing esports" may have been used at some point. While I do not think that this system is suitable for every tournament, it is certainly something that smaller invitationals should be considering doing. Being a pro-player is not an easy life as some people would have you believe. Depending on where you live, holding a sustainable income to the point where you can pay your bills and focus on practice is very difficult. The UK's pro-scene is effectively dead as a result, whereas countries where living costs are lower such as Poland and the Ukraine have a strong roster of pro-players. One way or the other, the only way the foreign scene is going to be able to compete with the Korean way of doing things is to ensure pros can dedicate 100% to their practice and treat it as a career. The tipping system assists in this. It also, along with our bounty system, largely removes the "meaningless game" from the equation (though we still arguably had one in the groupstage battle between Stephano and Thorzain, since Thorzain could not advance as first seed, but needed to win one map to go through to the RO4 instead of Nightend). I guess you can't get rid of meaningless games entirely in that format, but for the most part, as Darkforce displays beyond a shadow of a doubt, you can encourage players to pull off outstanding performances even when they have no chance of going forward in the tournament, rather than simply throwing their games resulting in a sub-par experience for the audience.
Regardless of what anyone may think about whether or not a last-place player should earn more than the 3rd place finisher, the fact remains that the tip system is personal choice. People can give to whomever they want for whatever reason they wish. We are merely enabling them to do so easily. Results are not, in fact, the only thing that matters in competitive Starcraft 2, they never have been and they never will be, regardless of whether or not we continue using the tip-jar system.
The Technical Side of things
Oh thank god, a tournament without major technical problems. We had no sound issues, no dropped stream, no overlays left on obscuring the game. Every mistake we made was minor (though still not excusable and will hopefully not happen again), in the form of things like getting the player race wrong on the scoreboard, once leaving a player nameplate on the bottom bar for most of the game and having the "next game" text header be out of date on multiple occassions in-between matches. The text issue is easily solved with the use of Xsplit's remote title system (basically means someone who isn't the guy on the computer that's doing the streaming can update the text. This is important because Xsplit has no preview feature to allow you to make changes to a scene without putting it live on air, a glaring issue in the software that needs resolving). Everything else is simply a case of keeping an eye on everything and not fucking up. This'll be easier once I get a second monitor and a bigger desk. We were also able to use Twitch's beta broadcast delay system which is much more reliable than Xsplit's, so we had a smoother stream throughout and eliminated the ever-present threat of stream-cheating once again (as we have every SCI tournament). As far as we're concerned, replay based tournaments are for the most part irrelevant now. The biggest argument for replays being used in tournaments was to prevent stream-cheating that could occur during a live, online event. Broadcast delay prevents that from happening and as viewers crave the live experience more and more, I believe replays should only be a last resort in future tournaments.
Aside from that we were able to use Dropbox to live-update our bracket and group image directly into Xsplit, a big improvement over previous tournaments where we've had to try and overlay a webpage into the screen-capture area which often ended up being the wrong size. The site also kept going down so we couldn't always put up the up-to-date brackets. We also implemented a snazzier overlay which answers the feedback given in previous tournaments that a simple video loop was not sufficient. We added a caster camera for the first time which we incorporated into the scene. We don't feel that a caster camera is appropriate during matches, but in pre and post game it does add just a little bit more engagement for the viewers since they have something visual to focus on. Overall we pushed the production value higher this time around.
Thanks
The thanks list is always large and I hope once again that I don't forget anyone.
Genna, my loyal wife and manager, who did the legwork in organising this event, booking Apollo's travel arrangements, managing the finances (and paying out the prize-money to all the players within an hour of the tournament ending), coding and maintaining the website at shoutcraft.com as well as putting up with my nonsense.
Cristian, our full-time graphics designer who is responsible for the new overlay, all the tournament art assets, promotional materials and live-updating the bracket images which we used. The tournament was a hell of a lot slicker thanks to your talents.
Apollo, for travelling back from Sweden to commentate this event and once again providing his peerless analysis and personality.
Twitch.tv and it's staff, who were kind enough to put us in their regular newsletter, frontpage the event on Twitch.tv and also keep an eye on the stream, resulting in a more stable and enjoyable tournament experience for the viewers.
Teamliquid, for their support and promotion as well as all the forum regulars who contributed to be a very active and lively thread.
Reddit, for it's support and promotion, which contributed a great deal to the tournament's momentum and build-up over the course of the week prior to the event.
The Teams, for sending us their players, in no particular order, Alternate Attax, Fnatic, Mousesports, Teamliquid, Millenium and of course, Team Grubby.
The Players, Socke, Grubby, Nightend, Thorzain, Cloud, Darkforce, Ret and Stephano for putting on a fantastic show.
The Viewers, for their support and time, as well as financial aid, resulting in a record breaking amount given to SCI and the players who participated within it.
Colin Smith, we named this tournament after him and once again he showed his generosity, donating $1337 to the next tournament.
We will see you in the next tournament, which we hope to hold in March.
