2016 NBA/ABA Tournament of Champions: ABA Seeding Tournament

It's time to get the 2016 Tournament of Champions underway!

If you read my intro post, you know I'm going bigger this year. I didn't feel that it was enough to just bring in the 2016 NBA Champion. This time, since I had the resources, I wanted to include the nine ABA champions in the mix as well.

I have no idea how good these teams are in relation to the NBA champions that have participated in the last two tournaments, and I wasn't sure at first how I wanted to integrate the ABA teams into this. Originally, I was going to have seven groups of nine teams from the NBA for pool play and have the ABA be its own group. However, that guarantees four ABA teams advancing to the Elimination Round, and I'm not convinced those teams deserve four guaranteed spots.

Thus, the fairest way to do this is to pre-rank the ABA champions, and to do that, I am having those nine teams spend the early part of the month of May playing a round robin tournament. This one is only a single round robin to minimize the mileage on these players, though they'll have a month after this to rest up and prepare for the big time. In the grand scheme of things, these games don't really mean anything other than pride and the best ABA champion getting to avoid the '96 Bulls until the Elimination Round.

So, here's ultimately how this is going to work. The placing by each ABA team in this portion of the tournament will correspond to which pool that ABA team will join for the filling of the eight groups for Group Play. My simulator doesn't do neutral courts in the NBA, and normally given the schedule this would be a problem, as I usually play with the left team on the road at the right team. However, my scheduler decided to put the '73 Pacers on the left hand side of every round, which would have them playing eight road games while four teams get five home games. I went through and remedied this issue on the schedule page, with the Pacers' home games getting a special note to signify it. With this in place, every team will play four games at home and four games on the road, all determined randomly (including the case of the changed games, where I just switched home teams for the '73 Pacers).

In the event of a tie in the standings, head to head record will take care of any two-way tie. In the event of a three or more-team tie, head to head record will remain the top tiebreaker, but in case there's a rock-paper-scissors tie, I will go down to record against teams above the tie, then record against teams below the tie, then point differential. All simulations are once again provided by WhatIfSports.com.

Good luck to these ABA teams, and welcome to the Tournament of Champions!

2013 Death to the BCS Playoffs: Official Bracket

Saturday was the BCS' final gasp. As far as drama goes, it went well for the "Cartel", what with the couple upsets and some compelling games. I maintain that having this 16-team playoff would do a lot of favors for the sport in that yesterday, basically every game would have had truer postseason impacts. It would also reduce a lot of the sheer hatred NIU suffered. I saw one person on Facebook who just lambasted them for no reason other than they were the modern day Boise State and dared tread on what elitists consider the sacred turf of the BCS. It's ludicrous.

Ultimately, that's why I started compiling stuff for a playoff system outlined in the book seen here. I'm sick of people talking crap about conferences like the MAC just because they're not the SEC, Pac-12 or what have you. So, with the regular season officially done (unless the Army-Navy game intrigues you to no end) it's time to outline what a truer college football postseason looks like.

A lot of people are arguing for either the 4-team playoff that gets implemented next year, but more people want either a four or eight-team playoff, with even some arguments for 10 or 12-team brackets. I'm arguing 16 because that gives every conference champion the opportunity to play for a national title. At that point you truly have every game really counting for something. To fill out the field, I will take 6 teams that didn't win a conference title but still had great regular seasons. Most years, this will likely mean most, if not all, one-loss teams and two-loss teams will also have a good shot.

The 16 teams will be seeded by a selection committee (read: me) with the help of some number crunching. The main reason for this is the other aspect of this playoff that really makes every game count: for the first three rounds, higher seeds get to host the games. Consider it a reward for a really good season.

Seedings are determined by kind of an organic process. I will look over schedules to try and decide who had some signature wins, how margin of victory looked, how bad their losses were, etc. I will also employ some math to help quantify things. Non-Conference Schedule Strength (NCSS) tries to analyze how well a team scheduled its out of conference slate. While it doesn't look at record, it rewards teams for playing major conference opponents and/or on the road and punishes scheduling cupcakes. It's not entirely a fair formula, but it can help a little bit. The more significant one that I created is the Playoff Points (PP) metric, which is basically the combined number of wins of each team a given school has beaten. It's a concept I borrowed from the IHSA. I also use a pair of computer rankings created by current or former people whose formulas were used by the BCS. Current computer guy Jeff Sagarin (SAG) has a "politically correct" version he sends them, but has a better ranking that I use. The late David Rothman (FACT) also created a formula that the BCS stopped using when they decided margin of victory wasn't important (or more so, decided it went against sportsmanship, which is something of a bogus argument). His formula is public domain and I found someone at UCLA who compiles it at the given link.

So without any further long-winded explanation or grandstanding, here's how the 2013 Death to the BCS Playoffs will be arranged.

2013 Death to the BCS Playoffs: Week 14 Seedings

Happy December! Hope you all had a good Thanksgiving weekend full of family (first and foremost), food and football. And really, the football part (Green Bay getting smoked aside) was amazing.

I didn't watch much football this weekend with a lot going on, but I was thrilled Saturday to see my North Central Cardinals earn a trip back to the quarterfinals of the D-III playoffs for the first time since 2010. And the Death to the BCS Playoffs get quite a shakeup as well. I didn't watch much of the big-time action this weekend, but I did see the end of the Iron Bowl. Holy crap.

With Alabama and a couple other key teams losing, there is a definite shakeup in the overall standings, which impacts the playoff seedings compared to last week. As a refresher, the Death to the BCS Playoffs are based on the system in the book whose cover is seen here, where a 16 team playoff is built from all conference champions plus at-large bids. I'm acting as a one-man selection committee who is giving tentative conference titles based on records and filling in the at larges from there. Seedings are based on a review of each team's resume, with help from a few formulas. I include my Non-Conference Schedule Strength (NCSS, for which you can see the most up to date info based on conferences here), Playoff Points (PP, basically showing how many wins the teams you've beaten have as something of a "strength of schedule" metric), and then computer rankings compiled by Jeff Sagarin (SAG, found here, using the rankings in order of the list) and the late David Rothman (FACT, found here, though his formula is public domain and rankings based on his algorithm is compiled by someone at UCLA) which use margin of victory as a factor.

So with all the chaos and math in mind, let's see how things unfold.

2013 Death to the BCS Playoffs: Week 13 Seedings

Saturday was my 24th birthday, and that morning I tweeted out a short list of things I wanted from the sports world for my birthday. Of those, a couple came true (including North Central's football team moving on) while others didn't (I'm kind of mourning D-Rose's latest knee injury). But the one I forgot to throw out there was for the BCS to undergo a little chaos, just to eff with people. Sure enough, it happened.

Of course, the drawback for this is that I need to reevaluate my playoff seedings, which takes a little bit of work. Granted, I've been updating my numbers weekly, so that part isn't new, but needing to shift teams around is also a responsibility I need to take. Every week I've had to do some shifting around, some weeks more than others. This might be one of the more chaotic weeks. And last week was pretty chaotic.

As a refresher, the 16 team field is made up of the ten conference champions, plus six teams selected "at large", then seeded in regular fashion. The official criteria to select at large teams and seed the field isn't concrete, but is based off of a mixture of Non-Conference Schedule Strength (NCSS, most recent data for conferences can be found here), Playoff Points (PP, basically the total number of wins of the teams you've beat), and computer rankings based on formulas created by Jeff Sagarin (SAG, found here, but I'm retweaking based on a closer reading of Sagarin's header and using the official "rating" column as his ranking, as it is a synthesis of his two other score-based rankings) and the late David Rothman (FACT, compiled by someone from UCLA, which can be found here). I also peer over schedules and try to decide as best I can how things should be seeded So without further ado, here's how I would seed a playoff if the season ended today.

2013 Death to the BCS Playoffs: Week 12 Seedings

After a busy night on Saturday involving 7 hours of basketball and some personal chasing sandwiching it, I headed home and listened to the end of USC-Stanford. Oh man. Great finish, and it throws a major monkey wrench into the Death to the BCS Playoffs.

I think I've said it on here before, but I'll say it again: for all its flaws, big time college football does seem to have the best regular season. And I still think an expanded, 16 team playoff wouldn't diminish it in any way. After all, the other divisions (D-III, for example) all have playoffs, and that same idea of "Every game counts" holds true, more so than in the big time. The NCAA unveiled its D-III bracket yesterday, and like in this system I've been outlining here, basically all conference champions get an automatic bid (except for a couple that don't want to hand one out), and everyone else has to battle for 5 at large bids. It makes more games meaningful for everyone. Isn't this how it should be in the big time?

So with all the games from this past weekend done, here's how I would seed the 16-team playoff. All ten conference champions make it, plus six at-large teams. Selection criteria isn't set in stone, but involves Non-Conference Schedule Strength (up to date conference averages can be found here), Playoff Points (basically, you get points for however many wins the teams you beat have), and computer rankings that factor margin of victory done by Jeff Sagarin (I use the "BCS Predictor" set found here) and the late David Rothman (his formula is public domain, and someone who uses it compiles rankings here). There's no exact formula for combining these factors, just me as a one-man "selection committee". For last week's seedings, click here.

2013 Death to the BCS Playoffs: Week 11 Seedings

Eleven weeks are in the books, and another unbeaten has fallen.

Granted, that unbeaten falling is somewhat old news, as Oregon lost to Stanford Thursday night, but it's still kind of a big deal. At this point, we are down to only six undefeated schools at the highest level. Glancing at standings before I even start to crunch playoff point numbers, there will definitely be a shakeup again this week. And to think, some idiots believe a playoff would damage the regular season.

As a reminder, this playoff is set up based on the playoff outlined in Death to the BCS by Dan Wetzel, Josh Peter and Jeff Passan. 16 teams get in: the ten conference champions and six at-large bids are awarded to the remaining pool. Criteria aren't set in stone, but I use a combination of metrics outlined in last week's post about seedings. In the meantime, if you want to see how I've ranked conferences based on their overall scheduling to date, you can view the most recent rankings here. Below are the new Death to the BCS Playoff seedings with all games through November 9th taken into account.

2013 Death to the BCS Playoffs: Week 10 Seedings

You'd think before writing last week's seedings I'd look closer at upcoming matchups.

I mentioned the ACC title game being between Miami and Florida State, not bothering to notice that they were playing each other on Saturday. This does in fact bring us down to seven unbeatens left in big time college football. Odds are at least one of these will fall, if not more, but I'm kind of hoping they all win out just to cause a mess at the top.

So in the meantime, we've played 10 weeks of the college season, and we just have the month of November left to get through. Looking a bit at schedules, I can see why BCS proponents make the regular season argument, because there are some really compelling matchups left on the docket, though I still think those matchups would be enhanced with Wetzel, Peter, and Passan's playoff system.

I remain a one-man selection committee, but like the first two weeks of seedings, I'm not going in totally blind. To aid my selections, I've been crunching numbers and pulling other numbers from other sources. Next to each team I have their seed from the previous week and a few metrics. "NCSS" stands for "Non-Conference Schedule Strength", and tries to objectively capture how well a team schedules its non-conference slate. A higher score correlates to (but isn't perfectly indicative of) a tougher schedule. "PP" stands for playoff points, and is basically the win total of all the teams a school beat. (Note: a school's win total includes wins over FCS teams, but you don't get any playoff points for actually beating them). "SAG" and "FACT" are computer rankings done by people who either currently or used to do them for the BCS and either were kicked out for using margin of victory (for FACT, created by the late David Rothman) or still use margin of victory in another formula, but have a "politically correct" version they send to the BCS (for SAG, created by Jeff Sagarin; the Death to the BCS Playoffs use the "PREDICTOR" rankings after using other, less accurate Sagarin rankings the first couple weeks). As a reference as well, current conference "champions" are based on conference record if the season ended today, with the tie being broken by overall record.

If you want to see how last week's seeds fell, you can see those here. Below are this week's updated seedings.

2013 Death to the BCS Playoffs: Week 9 Seedings

One thing I will credit the current system for: they do do a pretty good job of creating some drama in the college football regular season. I still think more could be created with a playoff, since more games would have implications, but what do I know? I'm not a money-laden, easy-to-be-bought sheep who believes every lie the big conference commissioners spit out for a long time. Hold on, let me get off my soapbox.

A couple of the unbeatens fell from the pedestal this week, leaving us with only 8 teams at that mark through 9 weeks of football. Of these, amazingly only one is guaranteed to fall (and at the latest, that would be during the ACC title game between Florida State and Miami). Could you imagine if we have 7 undefeated teams at the end of the regular season? How much anger would that drum up against not only the current system, but even the "College Football Playoff" that will debut in 2014? If this year were the first year of that and those seven unbeatens held firm, you'd have to snub three teams. Obviously we know Kirk Herbstreit doesn't have a problem pissing off fan bases of schools from mid major conferences, but Fresno State's fan base would be upset too. On a bigger scale, you'd have to keep one of the following teams out: the winner of FSU/Miami, Baylor, Ohio State, Oregon, or Alabama. This is why 4 teams is not enough. Of course, knowing the way college football works, 4 of these teams will probably end up losing and we'd have to discuss the merits of their wins and losses, which is why we need a 16 team field. Give every conference champ a berth, add in at larges to fill the field, and let them play to decide who's the best. Isn't that the way football was meant to be?

But enough of my grandstanding. With Week 9's results taken into account, here's how I would seed the 16 team field if the season ended today. If you want to see last week's setup and a key to the acronyms next to each team, you can view it here.

2013 Death to the BCS Playoffs: Week 8 Seedings

I've decided that for this year, this series of information needs its own set of posts.

If you've been reading COAS for most of the season, you've seen me update how well I think teams are scheduling their games. But you may also remember I alluded to a couple other formulas I was going to use. Looking through standings, with a lot of the unbeatens now fallen, I actually have few enough that I should be able to get them all in, and still fulfill the requirements to set up the field of 16 for the Death to the BCS Playoffs! And it's much less complicated then last year! Maybe that has something to do with the fact that this time, I know what I'm doing ahead of time.

So if you've never read Death to the BCS (and really, you should; it exposes all the flaws of the bowl system that we're switching from next season), it proposes a 16 team field made up of the conference champions (though we're down a conference from years past) and at large berths fill the bracket. The idea here is that elitism rampant in the power conferences gets tempered somewhat for the benefit of giving every team a legitimate shot at a championship by deciding it on the field, not in biased polls or algorithms. In an ideal world, I'd have a selection committee do this, but I'll make do with myself and a wealth of information. I also want to take into account a couple of the computer formulas. Jeff Sagarin, who does one of the computer rankings for the BCS, does a more correct version that accounts for margin of victory. The late David Rothman had a formula that used to be incorporated into the BCS until they took it out under claims of "sportsmanship". His formula is in the public domain, and those rankings can be found here. These are here to serve as something of a guideline to help confirm what my numbers and the "eye test" are giving me.

There's no exact method by which I'm seeding everyone; it's more of a "Take all the information available, and try to make an educated decision." So without further ado, here's the first set of playoff seedings. As a clue for what the abbrevations are, PP stands for "Playoff Points"; "NCSS" is the "Non Conference Schedule Strength", a school's score that gets averaged into the conference rankings I've been doing the past few weeks; "SAG" is Jeff Sagarin's "ELO_SCORE" on his webpage linked above which incorporates scores of games and is likely a more perfect system than mine; and "FACT" uses David Rothman's old formula.