First thing's first, let's explain exactly what the advanced seeding means so everyone's on the same page. 32 players in Wii U and 64 in Melee are being automatically promoted from R1 of pools. Our phases are very similar to how EVO's are: 3 rounds of pools to get to top 8. We have 128 pools in Melee. This means that the 64 players start in R2 winners with one bye. Compared to what they would face in an ordinary bracket, this means that we're taking seed #64 beating seed #193 as granted, #63 beating #194, etc. The Wii U bracket is roughly half as big, so promoting the top 32 is basically isomorphic. This means that 32 players in Wii U are automatically in the top 96 of the event (if they go 0-2, they will get 65th place tied with 32 other players); it doesn't mean that 32 players automatically get top 32 (if it did there would be no space for anyone else). For Melee, it's 64 automatically in the top 192. So why did we do this?

The primary reasons for this were scheduling constraints and the recognition of player efforts over the last year. What were the scheduling constraints? We have planned for Friday the TMG finals, Wii U Regional Crew Battles, Melee Crew Battles, and potentially a Brawl Tournament of Legends. These in total take up the entire day for each game (especially the crew battle brackets, which will be ~8 hours out of a 12 hour schedule). We also wanted to allow for top players to have a 1-2 hour friendlies + 'meet-and-greet' session during Friday. This already is 4 or 5 of the 6 available waves during the operating day. Therefore we were forced to make some choices: scrap the meet-and-greet, scrap a side event (which we had already promised and which presently people have donated to the compendium towards), limit registration to 2 events only to all players participating in these side events (also presents some difficulty as we don't necessarily know who all these players will be ahead of time, and has the hazard of discouraging participation in those side events), or automatically award promotion to R2 brackets in singles (of course, maybe there were other alternatives that we didn't think of). Why do these have to all be run on Friday? We do not want to run these on Saturday because our Saturday schedule is even more impacted: we have to finish 2 teams events and get to top 8 in 3 singles events in a community where cross-registration is a -huge- issue. We also felt it was unfair to distract players from their more important R2 and R3 bracket matches with relatively inconsequential side events or meet-and-greet sessions.

Each decision has advantages and drawbacks and we weighed them out. In particular, the drawback for the auto-seeding was that we might be preventing some potential upsets from happening (however, some of the examples presented of upsets have more to do with poor seeding in the first round than anything else. I do accept that obviously there's a chance that it prevents some upsets from happening, and it weighed in the decision), that the selection would be too difficult/imprecise, and that some people would miss out on playing top players (we thought the appeal of a friendlies + meet-and-greet session would balance that part out, not to mention that the vast majority of people wouldn't get to play them anyway).

For Melee, we had an easy reference for the top 64 in the MIOM top 100 which is presently coming out to allay the second concern. So this was a relatively painless decision on that end.

For Wii U, there were more uncertainties from my perspective as a Melee player as to what the community would prefer and how they would react. So I came to a facebook group which comprises about 100 top TOs/leaders/players in the Wii U community, purely on a mission of fact-finding. I have no personal stake in whatever decision the Wii U community would make, I wanted to know what the best course of action would be, and I was concerned about potential controversy. I explained the comparison to Melee: "To compare to Melee, we're seeding out 64 players from 128 pools. This means that we're taking seed #64 beating seed #193 as granted, #63 beating #194, etc. That's a pretty large gap. Upsets of course aren't impossible, in fact I'd probably expect 1-2 upsets on average, but we feel that was a fair price to pay. For Wii U, seeding 32 people out of 64 pools means taking #32 beating #97 as granted, #31 beating #98, etc. A smaller gap to be sure, but I'm not sure what kind of skill levels and dropoffs you guys have. It may be appropriate for you or maybe you'd want to do less."

In particular, this post from Dec 11 illustrates my attitude: "Do you guys think this is possible? I'd like to hear from as many other people as possible whether they think this is a good idea or whether this would create more controversy and complaining than it's worth." and 2 days later: "So guys, we're going to go through with this unless we hear any objections. This has a good shot at being the biggest Wii U tournament outside of EVO, so I want to make sure that what we do at G3 sets a good example for y'all's following Wii U regionals/nationals. So please let's hear any possible concerns about this before we make this final!" These posts were 'seen' by 63 people at last count and there were literally -zero- dissenting opinions. Nor were there any posts -at all- expressing skepticism about this decision in the thread discussing seeding that followed the next couple weeks which was 'seen' by 66 people. This is too many people to be the work of top players looking for a free ride or the voices of a small minority. I think this shows I did my due diligence as a tournament organizer in asking the community about the best options, and that I made the correct decision based on the information available to me at the time.

To turn things around, I'd like to compare what we are doing to things that past events such as Apex have done, whereby players could earn "Road to Apex" points to get seeding by placing in sponsored events. The top seeding as a result of this system are largely fine, but as you get towards the margins, it becomes clear that people who found ways to attend more events, or were in events with disproportionately weak turnout, or who otherwise were able to 'game' the system had a clear seeding advantage over ostensibly similarly skilled players. I think you'd have a hard time convincing many people that the Apex points system resulted in more accurate rankings than that year's MIOMrank (itself an imperfect system), especially the bottom half or so. Yet there was basically no outrage over this (in my opinion) plainly inferior promotion scheme, which was transparently devised primarily for the benefit of Apex and its associated events (not faulting them for that, just pointing out the difference in motives). Which leads us to wonder, if there were a Wii U circuit over the last year with similar methods, resulting in a promotion scheme like ours (which in my opinion would likely produce less accurate results than polling top players/leaders/TOs), would there be an outrage? I think it's unlikely. So why's it happening?

I think the outrage is due in large part to our latency in communicating this to the general public, which contributes to some idea that we intended to be secretive and mislead people. We thought it might be wiser to wait until the list was finalized to avoid rampant speculation and petitioning of our staff and community leaders for seeding and whatnot. Apparently that was wrong, and I'm willing to take full responsibility for that decision. Many complaints have come about the fact that our seeding would be subjective; I think people have the mistaken belief that just because something is superficially "objective" means it is fair/acceptable. Of course, the Apex system is not totally objective, because subjective value judgments go into creating the system in the first place; our system of ranking by general informed opinion is not extraordinarily more subjective than the Apex system. In my opinion, only when esports can produce a large base of players that can afford to play in major events regularly (like tennis) will those systems approach real utility for seeding beyond what any human could already tell you, and Smash is not there yet.

As far as failing to understand the values of the Wii U community, as I've said, I think any reasonable person in my position, seeing no complaints from 60+ respected people in the community for 2.5 weeks, cannot be blamed for making the same decision. Consulting the general player base is generally fraught with teasing out which subsegments of the base (stream viewers? casual players? people going to G3 only? people who have played for X amount of time?) they are whose opinions have weight and merit as far as what goes into a tournament, and does not reliably produce actionable results, much as we might wish it could. I'm also not going to apologize for perhaps different values about the nature of tournaments/competition from others, which I'm willing to defend if necessary. You might personally think this is an 'unfortunate experiment' but it's because of experiments that Melee has the solid ruleset that it does today and that we learn new things about how to construct Wii U rulesets with every tournament, including EVO which conducted an experiment of its own which I'd regard as more adventurous than ours.

Depending on the response to this post from the public and Wii U community leaders, and from our own assessment of alternatives, we're not ruling out the possibility of changing this for Wii U.