Staging the Showdown – Part 2

This week we continue with the second part of Loki’s 8 part series detailing Shadowloo Showdown 2012 from a tournament director’s perspective.

Last week Loki provided us with details of his history with the Shadowloo Showdown tournament and how he got things rolling for SS2k12. In part 2 Loki walks us through the initial stages of estimating the tournament schedule, including competitor forecasts and how those impact the pool structure that will be employed.


Approaching Shadowloo Showdown 2012 we knew we had to put some pretty high expectations on ourselves. We needed to cope with not just 256 but the next logical step of 512 players for both Super Street Fighter IV: Arcade Edition and the newly released Street Fighter x Tekken (which was sure to be everyone’s favourite game by then right?). For the slightly less popular games like Ultimate Marvel vs. Capcom 3, Soul Calibur V and so on we needed to be able to support up to 256 players.

This might sound elementary so far but deciding this early on set a bunch of important things in motion for the event. Based on these numbers we were able to work out how much time we needed to allocate for each game depending on how many setups we had access to, and consequently how many volunteers would be required to actually run the tournament. There are only so many hours in the weekend to get through the games, so depending on how many games we were going to run we needed to work out these numbers as soon as possible to ensure we could make everything fit.

Watching the Clock
Now comes the math; it’s pretty simple but it can be a bit confusing if you’ve never done it before. I’ll explain the method I use to work out how long a tournament should take to complete.

The general rule of thumb I’ve been using for years now is that on one setup, a full single elimination 32-player bracket takes approximately 2 hours to complete, while a double elimination on one setup would take approximately twice as long i.e. 4 hours. Obviously this is only an approximation as some games are faster and others are slower. For our purposes it’s a good starting point and you can always make minor adjustments at the end based on the game involved.

These numbers are then scaled up and down based on how many competitors and setups are available. For example if you double the players you double the time, but if you have an extra setup, you also halve the time. So while 32 players under double elimination would take 4 hours, if you have 2 setups it goes back to taking 2 hours again.

Sounds simple right? Well for the initial estimates it is, but even small deviations in how you decide to run the bracket can have major impacts on the actual runtime.

For example you might want to do run the top 8 on stage in front of the audience. This will slow you down considerably because you’re trying to run all 10 matches of the top 8 on one setup, and while in theory that should take just one hour, it always finds a way to take longer. My rule of thumb is to give closer to one and a half hours for a popular, hotly contested game like SSF4:AE. When there’s big money on the line players slow down, think carefully about the match, take their time between games to button check and most games go down to the wire. The grand final usually goes to the second set, almost every round is won by a magic pixel, and all your time estimates that rely on a mix of player skill levels in the early rounds completely unravel at this point. You need to schedule carefully for all that lost time.

If there’s one piece of advice I can give you when timetabling, it’s always give yourself more time than you think you need. If you finish a game early then people can relax, play casuals, play money matches and enjoy the event. However if you finish late, then the next game starts late and the entire weekend starts to fall apart one lost hour at a time. To avoid this you must overcompensate for time. Trust me, you’ll never be sitting there late on Sunday evening cursing yourself for making the schedule too relaxed and easy to handle, but if you’re still trying to make your way through pools to get out of the venue on time, you’re going to wish you trusted that one piece of advice.

Pools Aplenty
Ok, so how do we apply to all that to Shadowloo Showdown 2012? Coming up to the weekend we had 9 official games on the cards, and Sunday was already allocated to be the day of finals. So how do you fit the pools for 9 games into one day and still accommodate over 500 people?

We knew that we had 12 hours in the venue on Saturday and, in theory, our biggest games were going to be SSF4:AE and SFxT, so we started with those. The overlap between those two games was likely to be very large, so if possible we wanted to make sure they didn’t run at the same time. This meant giving ourselves a maximum of 6 hours for each game to work down to the top 8 for Sunday. Based roughly on the previous formula I mentioned, to get 512 players through double elimination brackets in 6 hours we needed at least 11 setups (4 x [512/32] / 6 = 10.67 setups). Even throwing in an extra setup for good measure that would leave no wiggle room at all for downtime or individual pools taking too long. On top of that depending on how many pools we ran we might also need to run semi final pools, which would add extra time of their own.

The point about semi final pools brought us to another dilemma: those pools would probably have most of the best matches of the whole weekend in them. If we wanted to be serious about getting those matches on stream we needed to give ourselves even more time where those pools don’t overlap with each other. But with 512 players, how many pools would we have, and how would that affect our 12 setups?

This brings us straight to the other big problem of Shadowloo Showdown 2011, the 64 player pools. Let me go on record here and say that you should never, ever use 64 players pools. Firstly it’s way too many people to be managing at a time, and secondly they all have to stand around the same small space waiting for their match which clogs up your pools area, no matter how big it is.

– Brendon “Loki” Watson

In part 3 Loki will provide the final decisions on the pool structure for SS2k12 and get down to the business of working out how many setups will be needed to make the tournament work in the time available, including how the SS2k12 team sourced the setups they required.

Feedback and Future Articles
If you have any feedback about this article, or would like to request a subject for a future Bracketed article, you can send me a PM or an email, or make a post in the Bracketed Feedback thread on the OzHadou forums.

]– Ziggy –[