Writers of Pro Football Prospectus 2008

Most Recent FO Features


» Defense and Rest Time

Do defenses really wear out over the course of a game? Do defenses benefit from long drives that give them more time to rest on the sideline? Guest columnist Ben Baldwin investigates.

14 Dec 2011

Scramble for the Ball: Strength of Schedule

by Tom Gower and Mike Kurtz

I Propose Best-Of-Three RPS

Tom: With the playoffs approaching, we're now in "figuring out the tiebreaker" season. And one of those tiebreakers that figures heavily, both for the draft and for the playoffs, is strength of schedule. Mike, what do you make of the NFL's use of strength of schedule?

Mike: It seems reasonable. There has to be some way to sort out teams who end the season with the same record, and given the short NFL season, there are going to be a lot of them.

Tom: As do several seemingly non-bothersome things, it kind of annoys me. I don't think it actually tells you what it supposedly tells you.

Mike: Well, it supposedly tells us that the team should be given more credit for its victories, since it dealt with superior competition. What do you think it actually tells us?

Tom: See, I don't think it actually does a good job of telling you that. One of the reasons is that your record is incorporated into your opponents' record. And if you win a division game, that effectively has twice as much impact on your strength of schedule.

Mike: But surely that's not an issue, since you are comparing two teams with the same record -- opponents' records are affected equally.

Tom: Divisional records count twice. So if Team X and Team Y are both 10-6, but Team X is 6-0 in its division and Team Y is 0-6 in its, ceteris paribus Team Y will have a better strength of schedule.

As an example, let's use the 2010 Indianapolis Colts, who went 10-6. The Colts went 4-2 in the AFC South. and Colts opponents had a record of 121-135 last year total. If you remove all games against the Colts, their opponents had an aggregate record of 113-121. Had the Colts gone 6-0 in the AFC South, their opponents’ aggregate record would have been 119-137. Of course, if you remove games against the Colts, I think it also makes sense to remove all of the games that essentially count for double. The Colts played all four teams of the NFC East, who each played six games against the other teams in the NFC East. The results of those games all perfectly offset each other.

Mike: Why are you doing this, though? You need to have some reason to throw out games, like doing so correlates with more accurate tiebreakers.

Tom: From the perspective of the Colts' strength of schedule, it doesn't matter which of the Eagles or the Redskins win their divisional games. I'll fast-forward to the conclusion then: the vast majority of the information in any team's strength of schedule measure comes from two sources: the two intraconference foes you play that your division mates don't, and how your divisional opponents fare against common opponents.

Mike: Wouldn't the latter part of that formulation be extremely helpful? Common opponents is a valid tiebreaker in and of itself.

Tom: It doesn't matter much how good the division you play in the opposite conference is, because a lot of that information comes out in the wash. Sorry, by common opponents, I meant "on your schedule." So it's how the Jaguars, Texans, and Titans do against teams also on the Colts' schedule.

Mike: Right. How is that not helpful? You're able to compare how one team did compared to its division mates.

Tom: If we want to compare the Colts to, say, the Ravens, then that information doesn't strike me as very important.

Mike: True, but we don't just do that. Otherwise Seattle would not have made the playoffs last year.

Tom: Well, it matters for wild card purposes. Think back to 2008, when there was a great hue and cry about the 11-5 Patriots not making the playoffs while the 8-8 Chargers did. Both teams went 7-5 against the AFC, while the Pats went 4-0 against the NFC West and the Chargers went 1-3 against the NFC South. The NFC South was a much better division that year than the NFC West. This is not very disputable.

Mike: Yes, but even for a wild card, the division you are in matters a lot.

Tom: But we're losing the indirect impact of this, because the strength of schedule measure doesn't reward you for playing a much stronger division in the other league. In fact, it hurts your because they're more likely to win games against your divisional foes.

Mike: OK, fair enough. That still only matters if you're in a particularly good or bad division. Most divisions on balance are average.

Tom: It bothers me, as a matter of principle, that playing a strong division in the other conference punishes you doubly: you're less likely to win, and it's likely to hurt your strength of schedule. In fact, you should be viewed as a better team for it.

Mike: How would you suggest we change the system, then?

Tom: It's not that I particularly object to the current measure, since I think it does capture some of the actual variation in the real strength of schedule. You do play the same games as your divisional foes, so the Colts playing the Patriots and Steelers and the Jaguars (who would've tied the Colts at 9-7 if they'd won and the Colts lost Week 17) playing the Bills and Browns essentially is the difference in their schedules. If there was just a widespread acknowledgment of the limits as to the actual information in a team's strength of schedule calculation, I'd be fine with that.

Any tweaks I'd want to see would be too complicated for most people to like. I'd want a team's strength of schedule calculation to formally be the record of the two foes you play that your divisonal mates don't, as well as the intraconference record of the teams in the division in the opposite conference. That, in a nutshell, is the information you're getting already. It probably doesn't put enough emphasis on the quality of your division, but I can more or less live with that.

Mike: I just don't see what, exactly, that gets you compared to the current system. Sure, your data is slightly different, but does that actually give you better results? I don't really see how it would.

Tom: I think it helps capture more of the true differentiation in strength of schedule. I'm not saying it has more than a snowball's chance in heck of getting adopted, but that's the real variation in schedule.

Fantasy Football Update

Tom: Thank goodness for unimpressive first round playoff opponents. I put up my worst score in a few weeks, in a week where a number of teams in the league had good games, but I won anyway. Because, frankly, my opponent was not that great. The trio of Aaron Rodgers, Wes Welker, and LeSean McCoy all had great games, but my normally strong IDPs were not impressive.

Mike: It's really weird that you rely on the strength of your IDP. I imagine most of your league forgets that you're even playing IDP.

Tom: We do start 11 of them. I think it's kind of hard to forget. I should point out I won by 31 points. Over the course of the season, I averaged outscoring my opponent by 65. And I'm currently projected to win my next game by 54.

Mike: Woo, projections.

Tom: Yeah, yeah. I also won the final game of the season with my bad team, thanks in part to my apathetic opponent starting Devery Henderson over Percy Harvin.

Mike: Only a lazy fool would ever do such a thing!

Tom: By winning that game, I finished eighth (of ten) and made the playoffs. I finished seventh in points, so I didn't end up with an undeserved fate.

Mike: Eight-team playoffs in a ten-team league is kind of insane, especially since it sounds like your championship is Week 17.

Tom: Well, I made the consolation bracket. Not the real playoffs.

Mike: That isn't the playoffs! Do not try and hide your shame from our readers!

Tom: I brainfarted, OK?

Mike: Yeah, sure.

Tom: Did you make the playoffs? Did you win?

Mike: I made the playoffs in both of my leagues. This week was the final regular-season game in one, although it's only a four-team playoff. Of course, it's an eight-team league, so four teams is somewhat standard.

Tom: Quite true.

Mike: I had commanded the top spot in the league for most of the year. In fact, since about Week 3. I fell from grace about four weeks ago, losing two in a row, but in the meantime I clawed my way back and finished the season with the top seed. Sadly, the third-place team put up more points than I expected, so I fell just short of the points crown.

Tom: Well, you're in the playoffs, and that's what really counts.

Mike: If you ever need to know anything about fantasy football, however, take this message to heart: The fourth seed is my friend Simon. He is from the Netherlands. He knows nothing about football. He joined because we needed an eighth, had Yahoo! auto-draft for him. He has defeated each of the other playoff teams once, including myself. I think that is a powerful message, really: Do not take this game very seriously.

Tom: After the mediocre job I did of predicting how the Titans-Saints game would go last Sunday, I am not surprised.

Mike: Of course, those who play a lot already know this. Heck, the first season of The League ended up with the clueless doctor winning the championship. But some people, for some strange reason, think that fantasy football success is a perfectly repeatable skill. It simply is not, which leads me to my other league. It features a fairly lame eight-team playoff in a ten-team league. After a nasty midseason tumble, I ended up with the sixth seed. The top seed was last year's champion, the eighth seed hasn't made it past the first round of the playoffs in the six years we've run this league. The eighth seed won, 166.54-160.04.

Tom: I hope Tim Tebow was involved.

Mike: Not quite, but nearly so ... Tony Romo. 38.84 points. And an assist from Shonn Greene, with 26.70.

Tom: So did you win?

Mike: I had the largest margin of victory in the league, 154.58-94.60. Third-highest points on the week, close behind the game I just mentioned. Jacksonville was a really, really good pick-up. 35.00 points, thanks to ... how many DST touchdowns? A ton.

Tom: With two D/ST scores? Oh yes. Apparently "giving up" trumps "ridiculously injured".

Mike: Yep. Of course, I have turned around and put in a waiver claim for Atlanta for this week, who are playing Jacksonville and Jackonsville's awful, awful offense. I have priority, so I'll get them. A defense seems like a silly thing to put a claim in for, but the rest of my team is solid, and it's a great matchup. Despite my success in this league, I am somewhat stuck.

Tom: How so?

Mike: I'm staring at my roster, and due to injuries to other players, I am seeing Steven Jackson and Jonathan Stewart. This is a deep league; we have two WR, two RB, W/T and R/W, plus eight bench spots.

Tom: Josh McDaniels did re-discover that the Rams employed Jackson and could give him carries in the red zone in the fourth quarter of Monday night's game. Unfortunately, the Rams are playing Cincinnati.

Mike: My bench is ... Lance Moore, Mario Manningham, Jermaine Gresham, Robert Meachem and James Jones.

Tom: I'd guess Jones would be one of the beneficiaries of Greg Jennings' injury.

Mike: On one hand, I'm really not hot on Jackson or Stewart, but then again, I'm not really hot on any of my bench players at this point. Either Moore or Meachem could randomly have a huge game, but there is no way to tell. With Stewart, I'm basically hoping for vulture touchdowns, and as you said, Cincinatti's rushing defense is pretty good; 14th in the league by DVOA. I probably don't have any choice but to throw the dice with Jackson anyway, so I guess it's just between Stewart and Moore unless I wanted to get really, really cute and go with Gresham.

Tom: Gresham in my book is indeed too cute by half.

Mike: Although I'm still somewhat undecided between Gresham and Kellen Winslow for my tight end spot.

Tom: I had no idea St. Louis was tops in the league against tight ends.

Mike: Basically, my team is Drew Brees, Julio Jones, Brandon Marshall, Vincent Jackson, Ray Rice and then blaaaaaargh. And yes, St. Louis' tight end defense is really weird since the only other category they're above 20th place in is against No. 1 WRs, which doesn't bode too well for A.J. Green in my other league, but I think he's good enough to overcome that. Anyway, which of Stewart/Moore/Manningham/Jones?

Tom: I lean toward Jones.

Mike: There will be much mulling this week. It would be nice to take the trophy home again.

FO Staff League Update

So, a funny thing happened in the first round of the FO Staff Playoffs: All the top seeds won. All three of them! Madness! Yes, our seedings are all messed up because Tom screwed up the bye week situation. Send all your complaints to him.

Discalculia Plus Ones (No. 1, Will, 8-2) 104 def. Edmonton Eulers (No. 8, Tanier, 4-5) 85

The top seed won in a landslide, which is always a sigh of relief for a standard seeding system. Of course, Tanier started Matt Forte and his mighty zero points. And, uh, Lance Ball? OK? Will didn't have a lot of firepower from any one player, but a great overall effort moved him on to the semis.

That's Great Hustle! (No. 2, Sean, 8-3) 92 def. Intentional Rounding (No. 7, Danny, 5-6) 86

This one was much closer, although the ideal scenario for Danny was only a one-point victory, so it was still a pretty sound defeat. Sean, meanwhile, left a total of 23 points on the board by benching Matt Ryan (11 more points than his starer), Aaron Hernandez (eight more) and Eagles DST (four more). Of course, nobody will ever criticize him for not starting the Eagles DST after the first half of the season.

Known Chumpsky (No. 3, Rivers, 6-5) 126 def. Equipo del Jefe (No. 6, Aaron, 5-6) 108

The new guy defeated the bossman. That's gotta be awkward. Fortunately, FO's equivalent of the water cooler is much more about weird tactical decisions and complaints about officiating than it is about office grudges, so he'll probably be OK. Rivers' team should also be OK, considering three of his players supplied 20 points or more each (Eli Manning with 22, Shonn Greene with 23 and Larry Fitzgerald with 20). His bench? Felix Jones, Andre Johnson (wishful thinking!) and Ryan Mathews. This team is dangerous.

Reverse Jinxes (No. 4, Elias, 6-5) 88 def. Los Pollos Hermanos (No. 5, Rob, 6-5) 88

Yes, you read that correctly. Two teams with the same record had consecutive seeding, and in their playoff matchup they scored the same number of points. No, I have no idea how this tie was resolved. All I know is that the Jinxes won, so yay for Elias?

We Have a Great Music Scene, Honest!

Rather than mock a commercial this week, your Scramble writers decided to provide you with the most unusual musical accompaniment involving their local metropolis they could find. They believe this song wins that honor.

Loser League Update

KICKER: O-lindo lower now. How low can he go? Well, 1 point isn't really that bad a Loser League score for Olindo Mare, but it's still the lowest for the week.

WIDE RECEIVER: Tom is grateful he's in a PPR league, as he earned more than Davone Bess's -1 Loser League score. At 1 point were Andre Caldwell, Harry Douglas, and Seattle Mike Williams.

RUNNING BACK: Tom would also like to note he had Ahmad Bradshaw and his 1 Loser League point on his other fantasy team. At 2 points were Beanie Wells, Donald Brown, Montario Hardesty, and Peyton Hillis.

QUARTERBACK: As bad as Christian Ponder was, he had two touchdowns, and those keep him out of last place. Instead, we're adding insult to injury for Matt Moore, who more than negated his touchdown with two fumbles and an interception for 2 points.



MIKE MARTZ AWARD: Beyond MB3, the Bears had great success playing a fairly aggressive man coverage scheme for roughly the first 55 minutes of Sunday's game before switching to a soft deep zone. Lovie Smith and/or Rod Marinelli, why? Why, why, why?

COLBERT AWARD: Can your Scramble writers really say something positive about Norv Turner? It looked like the game was starting to slip away from the Chargers, as a dominating first half and 16-0 lead had become a tight 16-10 game, but Eric Weddle audibled to a fake punt, the Chargers converted, took the ball down the field for a score, then picked off Ryan Fitzpatrick nine seconds later for a 30-10 lead. And all because Norv at least permitted Weddle's boldness. Go NOOOOOOOOORV!

Scramble Mailbag

Flores: Thanks so much for all your advice! I'm in the playoffs for my PPR league, and as usual, I can't make up my mind about anything. OK, I've made up my mind that I should play Wes Welker for WR and Darren Sproles for RB. I have no idea if Andre Johnson will play this week, so I'm guessing I'm pretty much compelled to play Julio Jones. For my other RB, as dysfunctional as the Rams are, Cincy isn't that good on run-D, so I'm guess S. Jackson should be my other RB. My flex is a disaster though. Santonio Holmes vs PHI, or Mike Tolbert vs. Ravens D both sound like awful choices. I'm praying desperately I get Felix Jones off of waivers for my flex, but if I don't, which of those? And if AJ by some miracle does play, should he play in place of any of the above choices?

For QB (yes, the Eli Manning vs. Ben Roethlisberger debate again), I'm thinking Eli since Ben might be hurt and SF's pass D is better (though only by a bit)? Finally for DEF, SEA vs CHI or CIN vs STL? SEA DEF is better and Chicago may potentially have no scorer besides Devin Hester depending on if Lovie Smith benches Marion Barber after his stupidity in the last game or not, though STL is about as bad at scoring.

There's also a (pretty crazy) IDP league I play in that has the following scoring for QBs: -.5 per incompletion, .5 per completion, 6 pts TDs (passing or rushing), 1 pt/20 yards, bonus points at 300/400 yards, 1 pt bonus for 40 yard completions and TDs, -.5 fumbles and an extra -2 if the fumble is lost. I drafted Philip Rivers and he's finally been doing decently the past few games...but this week he goes up against the Ravens D. Joe Flacco is the best QB available off of waivers (hurray, 20 person league!). Should I play Flacco over Rivers? Thanks again for all the help this season!

Mike: I'm not sure there is a universe in which you play Flacco over Rivers, especially with that weird scoring and in a game where Baltimore is going to be running with wild abandon.

Tom: A universe where we don't experience time unidirectionally and thus know with certainty which week Rivers is lousy and Flacco has a good fantasy game.

Mike: Fair enough.

Tom: Johnson isn't going to play this week, and I wouldn't trust him to play much even if by some improbability he is active. Double down on your Jones prayers, as with DeMarco Murray's injury I think he's a good play going forward.

Mike: Triple them, really, since your alternatives are Holmes and Tolbert. If you can't get Jones, though, I'd go with Tolbert and hope that San Diego pulls off the upset. (I will be hoping along with you.)

Tom: It's a coin-flip for me between Tolbert and Holmes. An ugly, tarnished coin. In your shoes, I'd probably play Tolbert but personally I would probably rather play Holmes just because I hate San Diego's use of Tolbert v. Ryan Mathews SO MUCH.

Mike: It's true. He's literally angry with rage.

Tom: I have no confidence Ben will be healthy. San Francisco has a good pass defense. Play Eli.

Mike: I'm going with Roethlisberger, actually. He'll be healthy enough, and Rashard Mendenhall is going absolutely nowhere on Monday night. The game will be in his hands. Of course, he only has to score, like, twice, but still, in his hands.

Tom: Well, I concur that Mendenhall is not going anywhere.

nick thunderdome: OK, I've got one for you guys. I've byed into the semis of my league basically on the strength of Arian Foster, Cam Newton, the 49ers DST, and carefully watching the waiver wire. If I'm going to win this week it isn't going to be any different.

Choice No. 1: which DST to play? The scoring in my league is such that limiting yards and points is what's most valuable (30 point baseline for a shutout with under 200 yards given up going down to -8 for giving up 36+ points and 450+ yards with the usual scoring for DST TD, sacks, safeties, and interceptions). I've got the 49ers on roster but don't love the match up they've got with the Steelers. Waiver wire options are the Cardinals vs CLE, Dolphins @ BUF and Lions @ OAK. The Saints @ MIN are also available but I wasn't really considering them.

Choice No. 2: Mario Manningham vs WAS or Eric Decker vs NE's pathetic secondary (this is moot if Andre Johnson doesn't play - I'll start both). Demaryius Thomas is also available on the waiver wire as is Donald Driver. .5 points per reception for WR.

Choice #3: My QB for the year has been Cam Newton but I've also got Tim Tebow on the bench. Cam Newton has a tough match up @ HOU while Tebow has a cake matchup vs NE. Scoring is 6 points per TD whether rushing, passing or receiving (hey, could be relevant - Newton has a 30+ yard reception on the year). As in most leagues, rushing yards count. Love the column and thanks in advance for the advice.

Mike: Wow, someone actually said they love the column.

Tom: I know, and I'm pretty sure nick isn't one of my relatives!

Mike: Anyway. Cardinals DST seems like the best option. Oakland has a bad but non-laughable offense, and Buffalo is definitely dangerous.

Tom: Carson Palmer has been making so many mistakes lately I'd be tempted by the Lions. The problem is if he suddenly stops throwing the ball to the other team, you might be sunk.

Mike: I don't think the Lions can generate enough pressure to force those errors out of Palmer.

Tom: I don't have that much confidence in the Raiders' offensive line, which has not been playing well of late. I like the Lions, but see them as a higher-risk proposition. I don't have much respect for Colt McCoy though, and thus would probably go with the Cardinals as an overall better choice.

Mike: At wideout, assuming that Johnson does not play, this is actually a tricky question. Manningham, against equal defenses, is a better option. Sure, he'll drop a touchdown, but he has a decent chance of also catching one. New England's secondary is just so bad, though.

Tom: I would rank Decker over Manningham myself. I also picked up Thomas off waivers for my good team and plan to play him. But that's over Davone Bess, whose name appears elsewhere in this week’s column.

Mike: Low bar.

Tom: And Jacquizz Rodgers, as well, who I did start on Shady's bye week. Yes, I would start Tebow this week over Newton.

Mike: Tebow is averaging about 17 points per week in a standard league with those scoring rules. Newton is averaging in the mid-20s. Plus, Tebow isn't torching secondaries; he's mixing runs and passes, even against Chicago. Denver's defense won't be able to stop New England's offense, and Bill Belichick isn't going to be stupid enough to shut down his defense (or tell Tom Brady and Wes Welker to cool their jets) when they have a lead. I just don't see Tebow outperforming Newton, even though I don't think Newton will be a world-beater.

Tom: I just think and see Dan Orlovsky throwing for a zillion yards, even if it's late in the game against particularly bad scrubs, and think Tebow will be pretty successful. And Newton's particularized extra value depends on rushing touchdowns, and even for good rushing QBs who get red zone carries, I can't trust that.

The final stretch! Send in your questions to scramble-at-footballoutsiders.com or our handy forum thread!

Posted by: Mike Kurtz and Tom Gower on 14 Dec 2011

51 comments, Last at 17 Dec 2011, 1:09pm by Tom Gower


by MJ (not verified) :: Wed, 12/14/2011 - 4:41pm

So I did not manage to get Felix Jones off of waivers. Woe is me. However, apparently both Chicago RBs are available on waivers. Would one of them be a better option than Holmes/Tolbert? I was looking in particular at Kahlil Bell since a) Lovie Smith might bench Barber for his idiocy, and b) with PPR in effect and Bell being used like Matt Forte, he seems to get a decent number of receptions as well.

by MJ (not verified) :: Wed, 12/14/2011 - 5:49pm

Also, Kevin Walter is available as well, as I assume he'll get more attention with Andre Johnson likely out.

by Mr Shush :: Thu, 12/15/2011 - 6:34am

I'd be wary of Walter. Owen Daniels definitely appears to be Yates' preferred target, and given the quality of the Panthers rush defense the Texans could end up with like a 3:1 run:pass ratio.

by Tom Gower :: Thu, 12/15/2011 - 2:39am

I have approximately zero confidence in any of Chicago's offense right now, so I'd stick with Tolbert this week.

by bubqr :: Wed, 12/14/2011 - 5:43pm

Quick note: This group of Loser League team names contains some good ones, better than the first part ones IMO. That's it.

by Stats are for losers (not verified) :: Fri, 12/16/2011 - 2:42pm

"Tebow The Letter" is my favorite of this crop, but "Arrelious Borealious" was close.

by Shattenjager :: Fri, 12/16/2011 - 7:52pm

I rather like "Brandon Flowers for Algernon."

by Perfundle :: Wed, 12/14/2011 - 5:50pm

"One of the reasons is that your record is incorporated into your opponents' record.

"Is that how the NFL calculates SOS? I know many sites' SOS will throw out the team's games themselves when looking at opponents' record.

"It bothers me, as a matter of principle, that playing a strong division in the other conference punishes you doubly: you're less likely to win, and it's likely to hurt your strength of schedule."

Huh, I've never thought about that fact. However, I tried out an extreme version of this scenario, and it doesn't seem to work out like this.

Let's say that all the teams have a 50% chance of beating every other team (and beats exactly 50% of those teams), except for the AFC North and South, who beats and loses to every team outside the division, respectively, and wins half of their divisional games. Let the NFC West and the NFC South play those two divisions, as they currently do.

Then each team in the following divisions would have the record:
AFC N: 13-3
AFC S: 3-13
NFC W: 6-10
NFC S: 10-6
NFC N and E: 8-8

If you tally up SOS for the NFC W and S, by your logic you would expect the NFC W to be punished for playing such a strong division and the NFC S to be rewarded. But the NFC W teams have either 138-118 or 144-112 as their SOS, depending on whether they played the NFC S or not, and the NFC S have opposite SOS's. It seems that if the AFC N is truly strong, and not just a function of the NFC W being weak, then the wins the AFC N will get from their other games more than outweighs the losses the NFC W incurs from playing them.

by tuluse :: Wed, 12/14/2011 - 6:34pm

I was going to point out that if your divisional opponents are losing games due to a hard SoS, it means they have less wins so it should be easier to win the division.

by Dr. NFL Tiebreaker Expert (not verified) :: Wed, 12/14/2011 - 6:09pm

The discussion about strength of schedule seems largely specious, at least as far as playoff teams are concerned. In the modern incarnation of playoff tiebreakers, the strength of schedule tiebreaker has never come into play. Why? Strength of victory is the tiebreaker before strength of schedule, and on the somewhat uncommon occasions in which strength of victory has come into play, there have never been identical winning percentages for the two (or more) sets of teams defeated by the tied teams. A moment's thought will suggest why: for, say, a 9-7 team, the opponents defeated by that team will have played 135 games other than those defeats (with some double-counting for division opponents defeated), and it's pretty unlikely for two randomly selected numbers from 0 to 135 to be identical, even though the random distribution follows a bell curve.

by Dr. NFL Tiebreaker Expert (not verified) :: Wed, 12/14/2011 - 6:17pm

Of course, strength of schedule does matter for draft order, and I've been closely following the strength-of-schedule possibilities for the Colts, Rams, and Vikings. The Colts won't quite clinch the first overall draft pick this week (unless they lose and the Rams and Vikings both win), but even if the Colts end up in a tiebreaker, the teams that they'll end up having played this season have only won 114 games so far, while the Vikings are at 118 and the Rams are at 121. (Remember that weaker strength of schedule means a higher pick in the draft.) Plus, when you compare the Colts and the Vikings, you can throw out the NFC South (each NFC South team has both the Colts and the Vikings on their schedule); when you compare the Colts and the Rams, you can throw out the AFC North (similarly). Plus, teams on your schedule play each other a lot, and you're guaranteed the same effect on your strength of schedule regardless of the result of that game; that cuts down on the possibility for strengths of schedule to change much.

by Jerry :: Wed, 12/14/2011 - 10:23pm

"The strength of schedule tiebreaker has never come into play" is the important point. The list of tiebreakers is long so that the NFL can avoid flipping a coin; as long as they sound reasonable, like strength of schedule, they'll stay.

by dmstorm22 :: Wed, 12/14/2011 - 11:07pm

When was the last time strength of victory came into play? ESPN's playoff machine tells me that it will if the Texans, Pats and Ravens all win out, because each would have a 10-2 conference record. I'm not sure how common opponents would work for three teams. There is one common opponent (Pittsburgh), and two of the three beat Pittsburgh, but apparently, ESPN says it would go to strength of victory.

by Jerry :: Wed, 12/14/2011 - 11:15pm

From the official list of tiebreakers, common games requires a minimum of 4.

I see that net points are also at the bottom of the list.

by dmstorm22 :: Wed, 12/14/2011 - 11:26pm

Makes sense.

I guess it is pretty rare for three teams to each have the same record and conference record. The playoff simulator told me that if that does happen, Baltimore gets the #1 seed, and NE the two. My gut feeling is one of those teams will drop a game. I thought it would have been Houston, but unless they lose home to Carolina or Tennessee, it probably won't happen. Might be this Sunday Night with Baltimore @ San Diego.

by PatsFan :: Thu, 12/15/2011 - 12:58am

Really? I thought SoS was what knocked out the 2008 Pats (or was it SoV?).

by dmstorm22 :: Thu, 12/15/2011 - 1:11am

Conference record. Miami went 8-4, New England went 7-5.

by Dr. December to Remember (not verified) :: Wed, 12/14/2011 - 6:21pm

I'd like to make a commercial-mocking request, since you might need an idea: any and all of the Lexus "December to Remember" series (one is linked in my alias). Sure, the idea of playing music that lets someone know they're getting a Lexus is cute... except that the Lexus jingle in question is the least memorable piece of music ever. Seriously, elevator music is catchier.

by Dr. Yes I Just Made Four Posts in a Row (not verified) :: Wed, 12/14/2011 - 6:32pm

[music plays]

[expectant waiting]

He: "What is that new ringtone? Did you put that on my phone?"

She: "Yes, I did..." [smiles] "Recognize it?"

He: [confused look] "No. Should I?"

She: [hurt look]

He: "Uh, I mean of course I do! Isn't it something from that Yanni/John Tesh concert we went to on our first date?"

She: [leaves room crying]

He: [calling after] "Wait! I meant the instrumental Roger Miller tribute!"

by BaronFoobarstein :: Wed, 12/14/2011 - 6:35pm

Also a far more honest reaction would be, "Really, honey, a car? You thought it would be a fun surprise to buy me a car as a gift? You don't think maybe such a large purchase decision should be made jointly?"

by bernie (not verified) :: Wed, 12/14/2011 - 7:29pm

If someone is buying you a Lexus as a Christmas gift, I'm guessing your financial position is one in which such things are a trifling matter.

by John (not verified) :: Wed, 12/14/2011 - 9:54pm

"I hope you don't mind that I signed your name on the loan paperwork, honey."

by Kevin from Philly :: Thu, 12/15/2011 - 10:47am

If that's the case, why wouldn't they buy something better than a Lexus? If any of my relatives are reading this, I'd prefer a Ferrari. Hell, I'd settle for a Jag over a tricked out Toyota.

by MVPFF (not verified) :: Wed, 12/14/2011 - 6:26pm

"Cincy isn't that good on run-D"

Er...Cincy is 10th in RB fantasy points allowed and held Arian Foster to his 2nd worst game of the year last week.

Not that SJax still isn't that guys best play, but it's not a good matchup at all for him. Especially considering STL has no passing game to keep the DEF honest.

by Wikitorix (not verified) :: Wed, 12/14/2011 - 6:50pm

"There has to be some way to sort out teams who end the season with the same record,"

Major League Baseball has it figured out - play another game. Baseball has the best method for settling a tied game, too - they don't do anything different in extra innings than they do in the first nine.

I recognize that NFL teams can't play extra games to determine playoff spots, but I think they should just play a full extra quarter to settle ties in regulation. There certainly shouldn't be different rules in effect for the regular season vs. the playoffs.

by BaronFoobarstein :: Wed, 12/14/2011 - 7:18pm

In a full extra quarter who receives the kickoff? I'd do two five minute periods where each teams starts with a kickoff. If it's still tied, then it's a tie. If it's a playoff game repeat until it's not a tie.

by Jerry :: Wed, 12/14/2011 - 10:34pm

Strategy does change in extra innings. Teams don't move the outfield in, or insert a fifth infielder, unless the winning run is on third.

And, FWIW, the Divisional round before the Conference Championships is called that because it officially includes (historically) the games where teams tied for the division lead played another game. Of course, that was before multiple rounds of playoffs designed for television.

by Wikitorix (not verified) :: Thu, 12/15/2011 - 1:03am

Strategy may change, but the rules don't. And the extra one-game playoff is part of the regular season.

by Jerry :: Thu, 12/15/2011 - 3:29am

Any rule for when to end overtime is going to be arbitrary. No less arbitrary than declaring one- or three-game playoffs part of baseball's regular season.

by Danish Denver-Fan :: Wed, 12/14/2011 - 7:01pm

Have a crack at this one:

First round of the playoffs. I need two RBs two WRs and a RB/WR-flex:

WR: Welker @DEN, Wallace @SF, Austin @TB and L. Robinson @TB.
Welker and Wallace probably, but I'm torn regarding the Cowboys.

RB: C. Johnson @IND, McCoy NYJ and Felix Jones @TB.
Any combination is open, I guess. McCoy has a horrible matchup in a game of little importance to his team. Johnson and Jones have great matchups but both have question marks.

TE: Antonio Gates BAL og Finley @KC.
This dilemma has been a nightmare all year. Gates is the better player, but again the opponent has me second guessing myself.

by Tom Gower :: Thu, 12/15/2011 - 2:48am

Interesting questions. I'd go with Welker for sure, then Wallace, then Robinson, then Austin would be my pecking order. At RB, even after last week's dud by CJ, I'd still start him against the Colts. The Jets aren't quite as strong a run defense in conventional stats as they are in DVOA, so I'd start McCoy with a certain level of confidence. The question then is Robinson or Felix at flex, and I'm confident enough Felix will get work to go with him.

Finley is apparently expected to get more work with Jennings out, and KC is lousy against tight ends. I think Gates is a good fantasy play, because he's healthier now and Rivers' most consistent target, but see Finley as a better one.

by Kevin from Philly :: Thu, 12/15/2011 - 10:44am

You're forgetting the Andy factor - given that the opposing D is weaker against the run than the pass, Andy will almost always opt for a 5-1 pass/run ratio.

by Led :: Thu, 12/15/2011 - 10:49am

So you'd go with L. Robinson over Austin? Despite Robinson's big day against the Giants, Austin played twice as many snaps. I wonder if Robinson will continue to put up big numbers if he's not seeing the field. But then again Austin had been underwhelming for me all year before his injury. I'm agonizing over this one.

by Mr Shush :: Thu, 12/15/2011 - 12:24pm

Speaking of McCoy, McCoy or Helu?

by Jonadan :: Wed, 12/14/2011 - 7:37pm

When you're looking at tiebreaking, there are the obvious ones:

1. Record (duh).
2. Head-to-head (the soccer fan in me argues that a 2-game split could still be decided right here by point differential).
3. Common opponents

2 and 3 present difficulties with the NFL's current schedule, in that it's possible that two teams could go a season without managing more than one game that fits into either category: e.g. consider the Cardinals and Giants hypothetically having the same record in a year the NFC West plays the NFC South. If the Cardinals played the Giants they'll have no common opponents (though it doesn't matter unless we somehow had a tie). If the Cardinals didn't play the Giants, there's exactly one opponent in common: maybe the Redskins.

Then you throw the draft in the middle of this, and the Titans also have the same record and the AFC South played the NFC West, so the Giants now have one common opponent with the Texans, while the Cards acquired 6 common opponents.

Anyway the short version is that the things which make sense logically don't make much sense in a league where you don't even play your whole conference (9 of the other 15 teams). As a game schedule planned for compelling interest, the NFL has it just about perfect right now: division matchups emphasized, conferences staying at the front, but enough variety to keep things interesting. For actually deciding who the better team is, it kind of sucks.

I really think this ends you up with the best solution simply being:

1. Overall record
2. Conference record (never mind which conference)
3. Division record (again, whatever division)
4. Head-to-head if applicable

My reasoning here is that with the limited season the NFL plays, we want to examine in every case the greatest number of games ("data points") that are comparable: reduce the sample size until we get something meaningful, even if it's actually less precise. Of course, this is liable to problems with a good team in a really bad division getting a cheap advantage over a good team in a good division (49ers have it easier than the Jets, for instance, at least when the gold pants aren't losing to the Cardinals...). But at some point, any system is results-based, and is blind to the fact that Chicago is suddenly starting Hanie, or the Eagles defense shouldn't have been allowed on the field this year.

Anyway, field test. NFL last season on these four criteria (ignoring the postseason) would be ranked:

1. *Patriots (14-2)
2. *Falcons (13-3)
3. *Steelers (12-4, 9-3, 5-1)
4. *Ravens (12-4, 9-3, 4-2)
5. *Jets (11-5, 9-3, 4-2) AND *Saints (11-5, 9-3, 4-2) <-- no head-to-head game
7. *Bears (11-5, 8-4)
8. *Colts (10-6, 8-4, 4-2) AND *Packers (10-6, 8-4, 4-2) <-- no head-to-head game
10. Buccaneers (10-6, 8-4, 3-3) AND Giants (10-6, 8-4, 3-3) <-- nhthg
12. *Eagles (10-6, 7-5)
13. *Chiefs (10-6, 6-6)
14. Chargers (9-7)
15. Jaguars (8-8, 7-5)
16. Raiders (8-8, 6-6)
17. *Seahawks (7-9, 6-6)
18. Rams (7-9, 5-7, 3-3)
19. Dolphins (7-9, 5-7, 2-4)
20. Texans (6-10, 5-7, 3-3)
21. Lions (6-10, 5-7, 2-4)
22. Vikings (6-10, 5-7, 1-5)
23. 49ers (6-10, 4-8, 4-2)
24. Cowboys (6-10, 4-8, 3-3)
25. Redskins (6-10, 4-8, 2-4)
26. Titans (6-10, 3-9)
27. Browns (5-11, 3-9, 1-5) AND Cardinals (5-11, 3-9, 1-5) <-- nhthg
29. Bengals (4-12, 3-9, 2-4)
30. Bills (4-12, 3-9, 1-5) AND Broncos (4-12, 3-9, 1-5) <-- nhthg
32. Panthers (2-14)

* indicates actual playoff team
Quick notes: some changes with this seeding plan. Giants win the NFC East. Steelers have the 2-seed instead of the Ravens.

Verdict: while the seeding looks reasonable overall, we still need another tiebreaker. Does common opponents do the trick here (guess: no)?

5. Jets (3-1)
6. Saints (2-2)
Browns (Jets 1-0, Saints 0-1); Ravens (Jets 0-1, Saints 0-1); Steelers (Jets 1-0, Saints 1-0) Vikings (Jets 1-0, Saints 1-0)

8. Packers (3-2)
9. Colts (2-3)
Cowboys (Colts 0-1, Packers 1-0); Eagles (Colts 0-1, Packers 1-0); Giants (Colts 1-0, Packers 1-0); Patriots (Colts 0-1, Packers 0-1); Redskins (Colts 1-0, Packers 0-1)

10. Giants (5-0)
11. Buccaneers (4-1)
Lions (Buccaneers 0-1, Giants 1-0); Panthers (Buccaneers 2-0, Giants 1-0); Redskins (Buccaneers 1-0, Giants 2-0) Seahawks (Buccaneers 1-0, Giants 1-0)

27. Browns (2-3)
28. Cardinals (1-4)
Buccaneers (Browns 0-1, Cardinals 0-1) Chiefs (Browns 0-1, Cardinals 0-1); Falcons (Browns 0-1, Cardinals 0-1); Panthers (Browns 1-0, Cardinals 0-1); Saints (Browns 1-0, Cardinals 1-0)

30. Broncos (1-3)
?31. Bills (0-3)
Chiefs (Bills 0-1, Broncos 1-1); Jaguars (Bills 0-1, Broncos 0-1); Ravens (Bills 0-1, Broncos 0-1)

Sorta? Well Broncos do have a technically better record in that they won a game, but they also got an extra game. Still, we can call that a 5 out of 5 (or at least a 4.5 out of 5) in a "randomly" chosen year (not really random) for a five-step process which now looks like:

1. Overall record
2. Conference record
3. Division record
4. Head to head
5. Common opponents

Or you can switch 2 and 3 so the division record remains more important. Whatever. After this, though, there's no plausible direct comparison, which I imagine you need in some years. So what's left? I guess things like net points, and strength of schedule (however you calculate that). So I didn't really accomplish anything here except to demonstrate that when is something like SOS actually needed? It can't be that often... right?

"When you absolutely don't know what to do any more, then it's time to panic." - Johann van der Wiel

by Travis :: Wed, 12/14/2011 - 9:01pm

2 and 3 present difficulties with the NFL's current schedule, in that it's possible that two teams could go a season without managing more than one game that fits into either category: e.g. consider the Cardinals and Giants hypothetically having the same record in a year the NFC West plays the NFC South. If the Cardinals played the Giants they'll have no common opponents (though it doesn't matter unless we somehow had a tie). If the Cardinals didn't play the Giants, there's exactly one opponent in common: maybe the Redskins.

Pairs of teams in the same conference always have at least four common opponents under the current scheduling formula, except for the teams that play head-to-head based on last year's record.

The Giants, for example, have 9 common opponents (12 games, counting division opponents twice) with other teams in the NFC East; 8 (11 games) with the Rams (who, like the Giants, finished 2nd in their division last year); 6 (9 games) with the other NFC West teams; 2 with the Packers and Saints; and 4 (5 games) with every other NFC North and South team.

by Jerry :: Wed, 12/14/2011 - 10:28pm

Point differential of one sort or another used to be on the list, but the league really wants to avoid the situation where a team just needs to lose by less than N points. If that team (call it Team A) is playing one that needs to win (Team B), both teams could kneel out the game once B takes a lead that's less than N.

by Jeff Fogle :: Thu, 12/15/2011 - 12:30am

Here's a link to the NFL Tie Breakers:


I have to say...the sentence "And one of those tiebreakers that figures heavily, both for the draft and for the playoffs, is strength of schedule," has to be one of the most inexplicable things I've ever read at this site. Figures "heavily?" It's barely ever in the discussion and hasn't been used as the good doctor pointed out in an earlier comment. Quality control just has to step in on stuff like this. It's in the headline and lead paragraph. Head to head, divisional records for divisional ties, common opponents, conference record...don't even recall general discussions about the role SOS plays in tie-breakers. An obscure element can't "figure heavily."

by dmstorm22 :: Thu, 12/15/2011 - 12:50am

Isn't it the only tiebreaker used in draft order (for non-playoff teams with the same record)?

by Tom Gower :: Thu, 12/15/2011 - 2:52am

As other people have noted, it is indeed a pretty unimportant playoffs tiebreaker. As you note, though, it matters a lot for draft purposes. I should have edited that part to say something intelligent instead of what I actually typed in our conversation.

by jmoh (not verified) :: Thu, 12/15/2011 - 5:22am

Hoping for some input here:

PPR league, 1 point per reception, all other scoring standard. Need a win this week to finish in the money. I have to start 2 of these 4:

Dez Bryant @ TB
Ryan Grant @ KC
Donald Driver @ KC
Pierre Thomas @ MIN

A lot depends on injury status, but if Starks doesn't play I'm leaning toward Bryant and Grant, though Driver and Thomas both look like they might be good plays with Jennings and maybe Ingram out. These decisions make me hate fantasy football so much... I just know I will leave points on the bench somehow.

by Chris UK :: Thu, 12/15/2011 - 9:03am

I would go Bryant and Grant if I was you, but that is probably why I got the thrid most point in my league and went 5-8.

by Tom Gower :: Thu, 12/15/2011 - 7:16pm

The Bucs are apparently planning on playing Aqib Talib against Bryant, but I don't think that's close to enough reason not to play him. If Starks is indeed out, I'd go with Grant as well. I don't trust Saints players beyond Graham when healthy and Sproles, so I'd look to avoid Thomas and if Starks is healthy would go with Driver.

by jmoh (not verified) :: Fri, 12/16/2011 - 1:51am

That's pretty much what I was thinking... great minds think alike.

Thanks for the replies!

by Chris UK :: Thu, 12/15/2011 - 8:59am

Also hoping for some input, I usually hate doing this, but:

ADP vs NO, Hillis @ ARI, Sproles @ MIN, Brandon Marshall @ BUF, AJ Green @ STL, Marques Colston @ MIN, Santonio Holmes @ PHI and Daniel Thomas @ BUF for two RB two WR one flex?

by Tom Gower :: Thu, 12/15/2011 - 7:24pm

It sounds like ADP's planning on playing, so play him. Sproles as well.

Matt Moore practiced today; he's up-and-down, and Marshall had a stinko game in the previous Bills matchup, but he's been valuable in the other games. Green was only fair last week, but the Rams have ridiculous injury issues at corner; I'd start him as well.

I'm not a huge fan of any of Colston, Holmes, or Thomas. Thomas has a good matchup, but I don't trust him to get volume. Holmes doesn't get volume reliably enough, so you're dependent on TDs for value, which I always hate. Colston's up and down as well; he had a good week last week because he caught the two deep passes the Saints hit. I wouldn't expect him to put up 20 points again this week, but I'd go with him as your best option.

by ABW (not verified) :: Thu, 12/15/2011 - 10:57pm

2 QB league, I'm planning on starting Rivers in one slot, do I start TJ Yates over Josh Freeman in the other? 6 pts per passing TD, 1 pt per 25 yds passing.

I've even considered starting both Freeman and Yates and benching Rivers vs. the Baltimore D, but that's crazy, right?

by Mr Shush :: Fri, 12/16/2011 - 6:31am

Freeman, all the way. The last two weeks, the Texans have been playing good teams with good run defenses, and have had to throw the ball. The Panthers have a terrible run defense, the Texans are at home, and if they can possibly give Tate and Foster 25 carries each they will. Look at Schaub's attempt counts against teams like Cleveland and Tampa, and then factor in that Carolina has a worse run defense than either, and Yates is not Schaub.

by Tom Gower :: Fri, 12/16/2011 - 9:22pm

Concur here. I don't expect Yates to get volume or TDs (the Panthers are 32nd in red zone rush D DVOA). If you really don't want to start Rivers, I'd look at TEN QB, either Hasselbeck (likely based on Friday's presser) or Locker, but I personally would start Rivers.

by Ivarsson.se :: Sat, 12/17/2011 - 7:42am

Okay, so got a QB question here. Do I oversell BAL D if I consider starting Rex Grossman (@NYG) or Mike Vick (vs NYJ) over Philip Rivers (vs BAL)?

I'm leaning towards Grossman or Rivers, Vick reinjury risk and rust vs a tough defense. It could be said that at least a mediocre showing is more important than the shot at a really good one, as the rest of my team *should* outscore his... (RB: R.Rice/BAL, Helu/WAS, WR: Welker, S.Smith/CAR)

by Tom Gower :: Sat, 12/17/2011 - 1:09pm

It's BAL D, SD OL, and SD's trouble winning against coverage more than the straight defensive matchup. I don't think sitting him against any decent option is anywhere close to obviously wrong. Even though he's been a surprisingly consistent fantasy guy lately, I fundamentally don't and probably won't ever trust Rex Grossman, so even despite the unfavorable matchup would start Vick over him.