Admittedly this is somewhat dated, but I still found it to be worth a blog post.

Say you work for a software team and GH2 comes out and you want to facilitate some musical smackdown? Within the group, you've got people that have never played before and people that can accurately play GH songs without a console being turned on... what do you do? Do you make it a winner takes all or do you attempt to mix it up a bit, to get as many people involved as possible?

As the tournament director, I went with the notion that I wanted to give everyone a shot at advancing, so I purposely mixed it up. A lot.

Really, it all depends on what your goals with the tournament are. I wanted something fun and focused on team building. Besides, with the new titles and features - Battle Mode in Guitar Hero III and the four-player action with Rock Band - this tournament layout can certainly be tweaked... also, with the number of people involved - it worked out for us in that we had a nice even number of people, but there would be other tweaking needed if an odd number showed up.

Preliminaries
I couldn't figure out a way around this... after all, how do you build a tournament bracket and keep people interested if you don't know what their skill set is? Also, there's different types of skills in GH2 that you can look for... if someone plays a song half-assed on Expert, does that make them a better player than someone that nails a song on Medium? Depends, right? And on the song too! So for the qualifying round, people were invited to come in and play whatever song they wanted and on whatever difficulty they wanted.

Scoring for the Preliminaries: Ranking = value star rating * multiplier for song difficulty

Star scoring:

Song difficulty modifier:

Made sense to me. If you play a very hard song from Face Melters and three star it on Hard you get 3 * 2.0 = 6. No matter what song you play, if you only three star it on Expert your are given the same score as someone that five stars that same song on Medium. And if you think you can just go with something on Expert and muddle through it... well, if you fail while playing a song from Relentless Riffs on Expert, your score is -5 * 1.6 = -8. Oh, one other thing - with this type of scoring is it very common for people to have the same score. In the event of this, I used the % of Notes Hit as the second ranking mechanism (higher % is a higher rank) and if those also caused a tie, GH2-based score was the deciding factor.

First Round
Out of the Preliminaries you will have a ranked list of all of your contestants and you can now build a bracket. For our purposes, I went with the "standard" method of seeding... had 20 players, so Rank #1 was matched up with Rank #20, #2 with #19, etc.

But here's where things got interesting: rather than having #1 play against #20, it turned into #1/#20 playing co-op versus #2/#19 playing co-op. As part of this, the song was picked completely at random, but I opted to restrict it to collections 3 through 7. Also, with the generic version of co-op, each person is allowed to pick their own instrument (guitar vs bass) and level of play, which also helps things along. An alternative could have been having #1/#20 go against #9/#10 and #2/#19 against #8/#11 but I found some perverse pleasure is knocking out one of the two best players in the first round *g*

This round is based entirely on GH2 scoring between paired opponents.

Second Round
The field has been halved - this is where you might have to start tweaking things again based on numbers but for us it was still OK: this is the round where the top plays the bottom. Basically the winners of the First Round play against each other, but this time, on their own rather than at the same time. The song is picked at random again and each player chooses their own level of play; since the round is based on GH2 scoring again, players will want to play at the highest level they can play best at, stringing together streaks. In most cases, the player with the higher score of the two will also have the best % of notes hit and the highest streak count.

Final Round
Truth is that we didn't get this far... these types of tournaments take time and we got tagged by a Holiday Season crunch and re-org - then other games came out that stole our attention! However I had a plan:

The remaining people match up for Pro co-op (meaning that both people play at the same level) and the lower ranked of the two contestants picks the level. The song is selected at random - whoever is still standing at the end advances. Because there will typically only be a handful of people left, people at the highest ran can be the odd player out and sit out the first rounds - otherwise you should be able to crown a winner by continuing to play until one player is left.

And that's pretty much it. It worked out well for us, which is why I'm sharing it... and I thought it was a decent piece of work, considering that it kept people very engaged. True, the very best player of our team was knocked out in the first round, but then does that make him the best all around game player or just the best guitar player? Since we are a team, I didn't want people to feel like the best ax wielder would be an instant winner but I also didn't want to just give the weakest players an automatic pass either... Difficult line to walk there, but I think this arrangement walks it well.

The next step of course is how to incorporate the newer features of the newer titles... maybe next Holiday Season!