Aaron, I'm wondering why you're so vehemently against the hair rule? I'm not for it either, but it doesn't seem particularly egregious to me.

while i agree it is the most pointless, i don't know about it being the 'dumbest'. the nba age limit is just as arbitrary and arguably even dumber according to free darko:

http://freedarko.blogspot.com/2008/03/henry-chinaski-is-not-spider-man.h...

I guess I don't see a problem in the playoff seeding change. Overall I'm not very pro divisions. The artificial slicing up of the league can leave teams from strong divisions out of the playoffs while rewarding teams from weak divisions. If a team wins their division with a 9-7, 8-8 record I don't see why they should be more rewarded than a second place 10-6 team.

If you're going to re-seed and more or less eliminate the point of winning a division, then stop playing division rivals twice a year as there's no longer a point.

I'm actually strongly in favor of changing the division structure and the playoff seeding. Under the current system, old historic rivalries that are really meaningless today are preserved at the cost of giving some teams an easy ticket to a playoff (and a minimum of a 4-seed on top of that) if just three particular teams happen to have off years. Meanwhile, if you happen to be in a division with just one or two other strong teams, you'll probably miss the playoffs or wind up with a seed lower than you should have (see: the Giants last year). Finally, with the current structure, seeding ties have gotten more common, leading to the weird and complex NFL tiebreaking procedures needing to be used more.

Divisions were originally set up so that teams that were geographically close would play each other more, building local rivalries, requiring less travel by the players, and allowing fans to see away games, but now you have Miami and New England in the same division, and Dallas and the Giants.

I've actually been trying to work out what would make sense to realign. Click my name for what I think would be an interesting idea to fix the issues I mentioned above and let me know what you think.

As to the other rule changes:

* Hair length: dumb, but I can see some logic, so not super dumb. It will make life tougher on the game charters, though...

* Roster increase: I'm in two minds, as a fan. On one hand, the roster increase would decrease the chance of my favorite players getting injured. On ther other hand, it would also decrease the chance of seeing players I actually care about play in the pre-season games. Still, I think I lean a little more towards the preventing injury approach.

* Defensive radios: Yes yes yes. Typical arugments against this can all be defeated by technology. For example, the concern that equipment mix-ups might accidentally lead to two players on the field with radios can be simply fixed by setting each "wired up" helmet to a different frequency, and only allowing a coach to broadcast on a single frequency at a time. Theoretically, you could wire up every defensive player, if you had enough channels, and the coach could have a dial with each player's name on it that he picks the one he wants to talk to for that play. Just put a software check in so that only one player can be selected per play.

* FG replay: Why not? This is one of the least controversial replay rules I can think of...

Re: MJK
I just read over your proposal, it sounds good and all as the current division don't make much sense, but the problem is that a team from one division might not play a certain team from another division for 30 or so years. Imagine if the Packers and the Cowboys never finish in the same standing within their division. The current schedule system has it so they play every other team at least once every 4 years, having a system where two teams have to rely on chance to play each other would almost make it feel like 4 separate leagues that come together for the playoffs.

As regards to the rules, the hair thing is dumb, the forceout rule shouldn't be changed, should just be able to be reviewed, and the defensive audio link should be implemented.

also to MJK,
I'm not saying your idea is bad, I'm actually intrigued by it, It would just need a way to make sure that teams from different divisions play each other at least once every so many years.

re #5:

i'm from miami and live in philly. while the geography may no longer make sense, the rivalries in the afc and nfc are both real and thriving.

besides, i like the fact that winning the division is most important for getting to the playoffs. adds a little controversy and gives mediocre teams some hope at getting to the playoffs...the nfl, where everyone has a shot at the playoffs/superbowl etc etc

good chance if Chargers stay good Chargers and Raiders both host playoff games.
Gould be Chargers be 11-5 and Raiders be 11-5 and on tiebreaker one team get bye and other team get home game in Wildcard round. Woudlnt happen like that in old years.

2008 predicitions for playoffs-
*division winner
*Pates 10-6
Titans 9-7
*Colts 12-4
*Browns 10-6
Chargers 11-5
*Raiders 11-5

Wildcard round
Titans at Chargers
Pates at Browns

bye
Colts and Raiders

Re 3:

Name the last time there was an 8-8 division winner. It just doesn't happen.

I agree with whats been said earlier, if you eliminate division winners getting the higher seeds, you might as well get rid of divisions. That would probably be terrible for the NFL, since every division has at least one historic rivaly. Most have more like 3, and there are few new ones coming up (Jaguars-Colts). To me football would just not be the same with out 2 Bears-Packers games a year.

And Re 6:
Rosters are already huge in preseason, for the first two games I believe teams can have 75 players on their rosters, and for the 3rd it's like 63 or something. I would bet the starters play just as much even with larger rosters.

Put me strongly against any changes in divisions, against a change in the forceout rule, for increased training camp rosters, and for wired defensive helmets.

I don't care if 2 or 3 wired helmets are on the field at once as long as the time allowed for the coach to communicate is kept short (so as not to allow individual coaching to every player).

Is there a way to search past Comments by user? I'd love an opportunity to read everything Raiderjoe's ever written.

Re: 11. In 1984, I think (from memory) the Browns won the AFC Central at 8-8. With a month to go in 2004, no team in the NFC West even was at .500, though the Seahawks went on a late-season rampage against the likes of Arizona and finished 9-7. I remember a discussion on FO a while ago where Pat (IIRC) crunched some numbers which suggested that one division winner ought to finish at 7-9 every 20 years or so.

That said, I hate this proposal, because teams play nearly 40 percent of their games against division rivals, and that ought to mean something.

I think it is inevitable that the NFL is moving towards having radios in all the players' helmets. I think this will be great. It will speed up the game and improve the quality of play. Not to mention there wouldn't be anymore illegal pumping of crowd noise in domes.

The only argument I've here is that the crowds will feel like they are no longer a part of the game. But, since when did the NFL give a damn about the crowds at the game? This is a TV sport, has been for a long time.

Just imagine a world where all 22 guys on both sides of the ball know exactly what they are supposed to do on every play. Who is against this?

Re: 5

I looked over your proposal, and while I like the idea of re-alignment, I think you go overboard. I think your problem is that you have each team in the division play each other team twice, so that 14 games are played against 7 teams while only 3 are played against the other 24. This would be a huge imbalance, and if you happen to be in a particularly weak division (can happen even with 8 teams), mediocre teams will get lifted up. Meanwhile, good teams in really strong divisions will get pushed down.

Look at last season for example--take a look at what the Jaguars schedule would have been. There's probably 10 wins guaranteed in that division at least, and I wouldn't be surprised if they pulled off a 16-1 record overall. Do they deserve that? No, but since they don't face many challenges from outside their division they could easily do it.

Re 16:

"Just imagine a world where all 22 guys on both sides of the ball know exactly what they are supposed to do on every play. Who is against this?"

I am. If I wanted to see perfection, I would watch robots play. I want to see people play the game, and people have flaws and they make mistakes. Also, the crowd is a huge part of the TV experience. I love hearing the crowd get louder when something cool happens.

I would be against all radios, but since it seems they are here to stay, I think the defense should get one just to keep up with the offense.

The seeding changes don't sound too bad, but after reading that it originated from Goodell's office I'm dead set against it. I'd love for the forceout to be eliminated. It gets rid of a judgement call, and gives DBs some much-needed relief. Plus it would be awesome to see guys try to carry WRs OOB

#2, amen. The age limit is an abomination. Forces me to watch college basketball, which is a horrible game

RE: 14

Linked a Google site search under my name. You'll get Raiderjoe-esque commentary as well, but who doesn't want that?

Re. 6 & 7:

I'm not sure I agree with MJK about re-alignment, although it is an interesting thought experiment. Also, if divisions are supposedly downplayed in his scheme, why still play you division rivals twice a year? In that vein and agreeing with Zach that each team in the league needs to play every other team in the league every few years or so, here are some suggested tweaks to MJK's thoughts.

1. Every team plays its division rivals once per year, alternating between home and away every two years. (7 games)
2. Every team plays every team from one other division each year, 4 home and 4 away. The division played changes each year so that the team plays every other division's teams once every three years. The site alternates for each match-up, so that every team play every team outside of the division both home and away every six years. (8 games)
3. Based on last year's record, every team plays the equivalently ranked team from one of the other two divisions at a neutral site. (1 game)
4. Every team plays plays one team from the "unplayed" division. This becomes a little complicated because each division will have 4 teams that need a home game and 4 teams that will need an away game. However, it's likely that the equivalently ranked teams from the two divisions will either both need home games or both need away games. A way to work-around to this would be to sort the "home" teams and the "away" teams based on the previous year's standings while maintaining the relative rankings. Then the "best away game" team from division "A" would play the "best home game" team from division "B" while the "best home game" team from division "A" would play the "best away game" team from division "B". (1 game)

An example:
Year 1 - The west division plays all of the midwest division, has a home/away game with the southeast division, and has a neutral site game with the northeast division.
Year 2 - The west division plays all of the southeast division, has a home/away game with the northeast division, and has a neutral game with the midwest division.
Year 3 - The west division plays all of the northeast division, has a home/away game with the midwest division, and has a neutral site game with the southeast division.

These tweaks de-emphasize the divisions and ensure that every team plays every other team in the league at least once every three years. I haven't a clue how this affects the advantages MJK claimed for his proposal, though.

Correct me if I'm wrong, but wouldn't it still mean something to win your division? Wouldn't it still mean that you're guaranteed to qualify for the postseason?

Having a 9-7 division winner host a 10-6 (or better) wild card seems like too much reward for winning a mediocre division, IMO. For the most part, division winners will have better records than wild cards, and therefore will host as many playoff games as their overall record allows. But to have a division winner with an inferior record host a wild card with a superior record in a playoff game seems more arbitrary than allowing overall record determine playoff seeding.

For the record, I'm against dismissing divisions altogether (as I said, I think it's plenty to guarantee a division winner a postseason berth), and I'd be opposed to mixing conferences in playoff seeding (I don't know if the CC is discussing these options, but I know they've been mentioned by the media in the past). I think the same qualifying standards should continue to apply: four division winners per conference and two wild cards, and then seed according to overall record after the six playoff teams have been determined in each conference.

The question, in my mind, is if a seeding change would allow a wild card with the second best record in the conference (unlikely, but possible) to get the #2 seed and the accompanying bye week. I could see that garnering some additional opposition.

RE: 18

People would still react to what is happening on the field by cheering/booing. The players would still hear them. If anything, there would be more opportunities to react, since there would be more plays in a given game, more opportunities for exciting/excruciating things to happen.

They said free agency would make the games less interesting because fans wouldn't feel as connected to the players on their roster, that they wouldn't cheer for a bunch of "mercenaries". Crowds sound as loud today as they ever did back in the days of "Plan B" and before. Even the ones that aren't illegally enhanced by pumped in audio.

#23, there would still be some effect of home field advantage, but it would be diminished. The crowd forcing teams into timeouts and bad plays near the endzones would no longer happen.

I don't believe there would be more plays in a game either. Teams will always use what they safely can on the play clock unless the game situation calls for hurrying up or burning the clock.

I agree with the problem about re-seeding based only on record: different divisions play different schedules. For that reason, I'm OK with the wildcards being 5th/6th. If you want to host a game, win your damn division.

Thanks for the comments on my idea. I don't expect anything like this to be seriously considered, it's just a fun thought exercise. The point is not to de-emphasize division play; it's to make division play once more central to the NFL in a way that makes sense. Back in the "old days", there were a bunch of different clubs, and each club played all of it's other club members a lot. Then, at the end of the year, each club sent their best to face off against the best of each of the other clubs. That's kind of what I was envisioning.

Having so much intradivisional play, and rewarding each division winner with a bye, will quickly create new rivalries. It's kind of like what happens in the MLB--teams play other teams in their division A LOT, teams outside their division but in their conference a couple of times a year, and teams in the other conference hardly at all. As such we get great, intradivisional rivalries year after year.

But the main point is that, with bigger divisions, your chance of winning the division easily goes way down, because with seven division opponents instead of three, there is a MUCH greater chance that at least one or two of them will be about as good as you are.

I agree that a weakness is that there isn't much cross-division play, but is that a problem? Until the NFL switched to the new alignment, the Patriots had gone something like fifteen years without playing the Bucs--to the point that, growing up in New England, I didn't even know that Tampa Bay HAD a football team. I like the idea of playing each division rival twice--once at home and once away, since it gives each team a chance for revenge.

Maybe a good modification would be to rotate the three extra-divsional games, instead of choose them by record...but I kind of like making good teams schedules harder...

Oh, Kenneth, as to the issue with the Jags...

Wow, I hadn't noticed how bad so many of the teams located in the geographic southeast were. Still this is probably an anomally--if a team in any given year has a 40% chance of sucking and a 20% chance of being really good, then in an 8 team division, you have only a 0.16% chance of all seven of your rivals sucking in a given year (as opposed to a 6.4% chance with the 4-team division structure), while there is a 79% chance that at least one of your seven division rivals will be "very good".

For the pushout change, could care less about the hair, and for the possible reseeding (if applying in that year).
Why? It makes teams like TB (this year) play their last game/s for real. I am 100% sure that the Giants going all out in their last game gave them confidence once they reached the Bowl against the Pats. Had they rolled over in week 17, they prob. lose the SB because they think they will lose to the juggernaut. Having said that, I hope this is what the NFL decides on:
Seeds 1 & 2 (bye teams)= top 2 division winners.
Seeds 3-6= best record; if there is a tie, div. winner gets the higher seed. For this last year in the NFC, that means the Bucs try to win one of their last two games to get the 4 seed and its accompanying home game.

Joseph,

Theoretically, what happens if the team with the second best record is a wildcard?

For example, suppose Buffalo hypothetically was a world beater this past year, and lost only two games--to unbeaten New England. Would you then seed Buffalo third, not second (and hence not getting a bye) just because they happened to be in the same division as the first team that went undefeated in a 16-game schedule?

MJK,

As someone who goes to games, I would have been disappointed if the kind of heavy intradivisional play you suggest would have deprived me of the opportunity to ever see Barry Sanders or Jerry Rice or Brett Favre or Peyton Manning in person.

While there are always boundary issues, I would also hate to see you do what Art Modell couldn't and kill the Steelers-Browns rivalry. (I'm sure other NFC East fans would miss the Cowboys as well.)

And, of course, the biggest thing that the League would have to do is figure out how it all would work for TV; right now, each network gets one New York team, and I'm sure that's of paramount importance.

It's an extreme idea, but I love JKL's proposal on the PFR blog (linked in name).

I especially liked the part where a 7th and 8th team could make the playoffs in each conference if they can stay close enough in the standings to the 1st and 2nd seeds. Essentially, the top seeds would earn their first round byes by finishing 3+ games ahead of #7 and #8. If they don't, then they have to play them in round 1.

A good possible incentive to be playing your best players in all 16 games.

I think any plan to encourage NFL teams to play their stars in the last few weeks is a good one.

If the Raiders make the playoffs, can Raiderjoe do the preview article for their first game?

How about a scheduling system based on 2 factors:

1) You play each team at least once every 4 years. So, basically you would play 8 teams one season, and a different (pre-determined) 8 teams the following season, etc.

2) You play 8 games against the 8 teams nearest you in overall record the previous year. (In 2008 the Miami Dolphins would play the teams that finished 31,30,29,28,27,26,25 and 24 in 2007. They would also play the 8 teams they were pre-determined to play in 2008.)

I can't decide how important the rivalry games really are.

I can't believe that there are people who don't like the NBA age limit. It's good for NBA teams (for scouting), good for college (having top players), good for college fans, good for NBA fans who want to scout prospects or who want better rookies, and good for NBA veterans (since they can stay in the league a year longer).

The only people who it's bad for are the workout-wonder high school prospects who get exposed in college and get drafted lower than they would have. But before you feel bad for these guys, remember that they are essentially busts waiting to happen. Who feels sorry for busts, especially drafted by their favorite team? (I guess it also hurts the very few players who could be successful straight out of high school, but they will get paid a year later anyway and sometimes actually get drafted higher.)

My problem with giving a wildcard team a higher seeding then a divisional champ is what happens when the 2006 Jets get a 10-6 record against a whole heap of non playoff teams and they get a higher seed than the 2008 Browns who had to play the AFC South and NFC East and won the division with 9-7. 10-6 wouldn't make the Jets a better team or more deserving of a home play off game, their weak schedule would.
The point of the division format is that the winner is the best team of the four teams with the most comparable schedules. So under the current division format changing the playoff seedings would just lead to some other trouble down the road.

I think giving crappy divisions a playoff spot is enough when we are talking about four teams in a division.

Yes, the last time a team went 8-8 was 1984. The change to 4 team divisions increases the likelihood it will happen. In fact, it did happen in the old AFC central with 4 teams.

As to Stuart's point, yes, there will be occasions when a team with a tougher schedule has a worse record only because of the vast difference in strength of schedule (see 2002 AFC). But far more commonly, a division winner with a worse record is, in fact, a worse team than the top wildcard who has a better record. I went through 1995-2007 and looked at all situations where a division winner had a worse record than 1 or more wildcards. Putting them head to head by DVOA (and calling all teams within +/- 5.0% a tie), the division winners were 4-10-5 against the wildcards. The 4 that were "better" despite the worse record were Oakland 2001, Philadelphia 2001, NY Jets 2002 (over Colts) and Tampa 2007 (over Giants)

Aaron likes free flowing hair in order to more easily track players ^_~

"The four teams in a tough division could form a voting block to stop change."

If you are aligned in a division with consistently good franchises with owners that try, then you should actually want this change. Who would have benefited from it recently? Teams from the NFC East, who Clayton cites should form a block against it. Dallas in 2006 would have been playing to get a home game in week 17 over Seattle, instead of losing a meaningless home game to Detroit when they couldn't improve their seed. NY Giants in 2007, when Tampa Bay rested starters because no one else in their division was at .500.

Who should be actually be against it? Teams who get to compete against the Bidwells and Fords to gain an automatic home game. Seattle has it good right now, so they should lead the "No" charge. If they have a great season, like 2005, well, this rule wouldn't change a thing. But if they have a mediocre one but win the NFC West with 9 or 10 wins, well, they would have increased competition to keep that home game.

It was 1985 when the Browns won the division at 8-8...same year the Bears smoked the Pats in the Superbowl.

A realignment may be over-due, but what division would the Raider's join once Al Davis moves them to the Moon?

re40
Raiders stay in AFCWesy with Broncos Ciefs, Chargers. If relaignment happens maybe it mean AFCWest is 5 teams or 6 teams. Maybe Sputh division go bye bye and Texans come into West. Colts go to North with Bengals.
Titans go to West if divisions are 6 teams each. Jaguars go to East so to be in with Dolphins.

the Raiderjoe relaignment plan__--

AFc East
Bills, Pates, Dolphins, Jets, Ravens

AFc Central
Bengals, Browns,steelers, Jaguars,colts, Titans

AFCWest
Raiders, Chiefs, Chargers, Broncos, Texans

The playoff seeding proposal is decidedly NOT about improving the playoffs or even making the seeding more fair. It's about improving the NFL's product in the last week or two of the regular season.

The proposed change makes it less likely that a playoff-bound team would choose to slack off in or even throw a week 16 or 17 game. Yes, it's against the rules to throw a game, but it happens and it's impossible to enforce. (You know I'm talking to you, Co *cough * lts!)

There are other ways to make playoff-bound teams actually try to win in week 17. You could link the late-round draft order to winning the last game(s) of the regular season, for example. But as these ideas go, this re-seeding idea would work pretty well.

I loved it when the NFL went to 32 teams, a schedule that saw you play every other team on a regular basis and re-emphasised Divisions (twice as many Division winners as wild cards qualify). I can't believe they will be so stupid as to tamper with that.

You can't directly compare one division with another because of the schedules - the current system is fine, why tinker with it?

The Raiderjoe realignment plan is okay. It gets the Bob Price Seal of Approval!

Re: 40

IT is not yet possible to play NFL games on the moon. In time, it will be a reality. I believe there will be a domed stadium on the moon by the year 2050. There are many advancements being made in space travel and regular citizens will be travelling to the moon by 2025. It is only a matter of time until a football stadium is built on the moon. Moon football would be a tremendous revenue generator for the NFL. Also, teams that have to play on the moon will not have to worry about the excessive travel. By 2050, travelling to the moon to play football would be akin to driving 5 miles to the local convenience store to buy a bottle of milk.

I think the better team should have the better seed. Of course, it's impossible to determine that based on w-l record alone but I lean towards a wildcard team with a better record getting a home game over a division winner. True, schedule differences make this an iffy proposition.

My proposal is: wildcard teams must finish two games better than a division winner in order to host a playoff game, i.e. a 10-6 WC gets a home game over a 8-8 div champ but not a 9-7 one. This reduces the impact of unbalanced scheduling. Won't happen of course...

But I agree that the whole point is to make more week 17 games significant. We've seen recent examples of WC teams taking the away game path to the SB title so homefield is just not that critical.

As for helmet radios, my first choice would be to take them out of the qb helmets. Since that's not going to happen, I'd give them to the defense too, probably three players (offenses typically have 3 qbs w/radios, I think). Allow the defense to have only one radio-head on the field at a time (a radio player could have two helmets and only use the radio one as necessary so he could still play w/o the radio). Make it a challegeable offense (like 12-men on the field). Put the radio "dot" where it can be easily seen by TV. Putting a radio in an unmarked helmet costs you a 1st round pick and fines of 20% of salary levied on all culpable coaches/players.

Does anyone know how the qb helmets work when the offense puts a slash-type guy on the field?

Oh yeah - I'd keep forceout rule as is but if we want to simplify calling the game, make it one foot down and no forceout exception. This gives the defense and the offense a rule benefit and makes both calls easier.

Does anyone know how the qb helmets work when the offense puts a slash-type guy on the field?

My recollections are that when Kordell Stewart was Slash, he had a helmet receiver, but now, Seneca Wallace has two helmets.

DIVISION A
TEAM1: 13-3
TEAM2: 10-6
TEAM3: 3-13
TEAM4: 3-13

DIVISION B
TEAM5: 9-7
TEAM6: 8-8
TEAM7: 8-8
TEAM8: 7-9

Who's the better team, 2 or 5? Is there enough info to tell?

The proposed playoff revamping is like taking ibuprofen for herpes... yeah, it might make the pain go away, but only for a little bit.

You're trading the chance at better week 16/17 games for what might very well be stinker playoff games. The 8/8 or 9-7 division winners are likely to get exposed and thoroughly torched in the first round if they have to play a road game against a good team. Since even the 1st round of the playoffs has a MUCH lower game selection pool then weeks 16/17, I think they'd actually be *decreasing* quality of the product with this proposal.

Oh, and let the damned dirty hippies and rastafarians have their long hair. At least until more than two people on the field (for the same team) have it obscuring their name. Then they have to roshambo (South Park style) for the right to have long hair on the team.

#50 - Nothing like enough information to tell.

Let's look at the effects these changes would have had on each season since 2002.

2007: Jags would have hosted Steelers, and Giants would have hosted Bucs. Per DVOA, that's one game moving the right way and one the wrong, with an aggregate 12.2% swing in the wrong direction.

2006: No change (I think - can't be bothered to work out tiebreakers between SEA and DAL and in any case think "won division title" should be the first tie-breaker under the new system).

2005: Patriots would have been at Jaguars (13.6% the right way). No other changes.

2004: No changes.

2003: Titans would have hosted Ravens (4.8% the right way). No other changes.

2002: Colts would have hosted Jets (14.9% the wrong way). No other changes.

Aggregate swing over 6 years: 8.7% the wrong way.

Games affected: 5 of 24

Games affected the right way: 3 of 5

Games affected the wrong way: 2 of 5

In other words, it looks very much like a wash in terms of fairness, benefiting some excellent teams that played in divisions with another excellent team (2003 TEN, 2007 JAX) and penalising some good teams that played in divisions that were tough across the board (2002 NYJ). Overall, I'd marginally favour it on those grounds, as the former are probably more realistic superbowl shots.

As mentioned above, however, the really important point is the benefit to competitiveness in weeks 16 and 17, which clearly far outweighs the impact on wild card weekend.

Incidentally, masoch, your hypothetical actually looks an awful lot like 2002's Jets-Colts game, which DVOA says was played at the right venue under the current system.

Apologies for double post.

re: 33 (richie)

the problem with weighting any given field like you suggest (Delphins play #31, 30, 29, 28 ... 25 etc) is that there are too many teams 'seeking' games with the teams in the middle of the sorting. For example, by that criteria, the 16-0 Patriots are expected to play the 8-th best team... but the 8th best team expects to play #s 5-12. there are solutions that get ferociously complicated that i've worked out for six-team fields but with 32 it would likely look something like this:

the top team plays # 2, 3, 4, 6, 8, 9, 10, 13, 14, 15, blah blah, 22.

so you sort of have to create a big matrix that still gives each team 16 games. its not a great system.

calbuzz,

not enough info. Even assuming each's team's record is, to first order, an estimate of how good it was, and all you're looking for is to characterize strength of schedule as a second order effect (which I think may be a mistake...further iteration is probably necessary), the info you give only gives us information about opponents in six of each team's games--we know nothing about the other ten opponents. For instance, your Team 5 may look like the best team in a mediocre division, but that whole division might have had to play eight games against two very good other divisions (like the NFC East this year), while your Team 2's division might have been matched up with the equivalent of this year's NFC West.

Actually, I've always been against using record for playoff seeding, because one or two games are way too much in the noise given difference in strength of schedule.

How about this idea--it's something I've kicked around in my head for a while:

Use a point system. Winning a game earns you positive points, while losing a game earns you negative points. When you beat a team, you get one point for every other win they get that season, while when you lose to a team, you lose one point for every game they lose that season. So beating a team that ultimately ends up 10-6 gives you 10 points, while losing to a team that ultimately ends up 13-3 only costs you 3 points.

Then just seed teams according to who ends up with the most points.

I've done this calculation for a number of years since 2000, and it MOSTLY ends up ranking teams the way W/L records would rank them. When two teams are tied, it almost always indicates that the team that common sense claims is better should be ranked higher. In a few cases where strength of schedule differences are extreme, it sometimes puts a 9-7 team above a 10-6 team or something like that. It's more accurate than W/L rankings as compared to fully iterative methods for ranking teams (such as Maximum Likelihood Estimation), which isn't surprising, because it IS similar to the first iteration of such techniques. Yet it is simple enough for non-mathematicians/statisticians to easily understand and calculate by hand.

It gives more meaning to late season games, reduces the need to go to tiebreakers, and gives fans a stake in games that involve teams other than their favorite team, since such games can have reprecussions on their own team's standings.

RE #56:

I don't get it.