Foist!

I'm not sure what people expect out of Bill Cowher. The Steelers were hardly a defensive juggernaut with him at the helm. They were good, but so was the talent at key positions.

Yeah - Cowher is way overrated. Two SB appearances in 15 years, 1 in which they were slapped around, and the other in which they slipped into the playoffs under the radar and rode momentum to the SB victory. Not to mention that the NFC Champ Seahawks were probably one of the weakest Super Bowl teams since the '00 Giants, and the officiating was atrocious.

Cowher is able to keep his teams consistently competitive, but consistently challenging for the conference championship, no.

My unofficial calculation is that PIT's average defensive DVOA rank from 1996-2006 was 8.5, which is the 3rd best:

1. BAL 7.1
2. TB 7.6
3. PIT 8.5
4. MIA 9.0
5. DEN 11.1

The gap between #4 and #5 is by far the largest between any two ranks in the entire list so I'm comfortable that PIT had one of the top 4 defenses - on average - in the league during the DVOA era.

Were they ever a "juggernaut"? They never ranked 1st in Def DVOA in a season. Their best def DVOA was -17.60% in 2004 (3rd) - the 27th best def DVOA, to the extent that DVOA is comparable across years.

The worst average def DVOA ranks:
32. ARI 25.0
31. HOU 24.2
30. CIN 23.9
29. IND 23.4
28. MIN 23.0

Bill Cowher has a .619 winning percentage, made the playoffs 10 times in 15 years, including 2 AFC Championships and 1 SB. While they are not going to rename the Lobardi trophy or anything, the guy is a very good coach with a proven tack record. Who else that's available now is better?

#4, hear, hear. It absolutely astounds me that people discount the accomplishment of a .619 winning percentage over 15 years in the salary cap era. Because an extremely small number of playoff games were lost? Unbelievable. Cowher, if he takes a job next year, is going to command huge compensation, and for very good reason.

I ran the average total DVOA ranks too (1996-2006).

1. DEN 7.9
2. KC 9.5
3. NE 10.0
4t. GB 10.2
4t. PIT 10.2

Denver's top ranking was by a wide margin. Enjoy Shanahan backers. As a Chiefs fan, I take no solace in their #2 standing which includes not only no Super Bowl wins, no appearances, but not even an AFC championship game in that entire span despite fielding, year in, year out, a good team.

Bottom five (no surprises):
32. ARI 27.9
31. CLE 25.0
30. HOU 24.7
29. DET 22.1
28. NO 21.7

The gap between ARI and CLE is the biggest between any two teams. The gap between DET and HOU is the 2nd biggest, that between DEN and KC is the 4th.

Who else that’s available now is better?

Schottenheimer! Oh, wait....

Is John Fox really on the hot seat? The guy took over an awful team and has taken it to two NFC title games & a Super Bowl in the last five years? Everyone else on this list I get. I mean, how Couchlin even has a job this year is beyond me but Fox? I just don't see it.

Would he even be here if Bill Cowher didn't happen to live in Carolina?

No way John Fox gets fired after this season unless there is a complete player mutiny in Carolina or something.

Also, if the Browns fire Crennel (and they may well do so), it's possible people will look back on that in 15 years as a move akin to dumping Belichick in 1995. Crennel has proven himself coaching in other locations, and I don't know if there is a coach out there who could have done much better than 10-22 with the talent the Browns have had over the past two years. Now his job is on the line, but he's also expected to break in a rookie QB.

If he gets fired, I can't believe that it's anything more than scapegoating.

#9/re:Crennel

Not to mention, if Romeo does begin breaking in Quinn this year axing his head coach and offensive coordinator could seriously and negatively impact his future.

For what it's worth, Bill Belichick was on the chopping block going into the 2001 season.

Phil (#11 )--

Was he? Belichick had only been coaching the Patriots for a year.

True, he wasn't exactly popular with the Boston press, but even unpopular coaches (like, say, Belichick in Cleveland) usually get three seasons.

Ah, the offseason.
24 days 8 hours until the 1st preseason game.
55 days 8.5 hours until the regular season.
3 weeks and 3 days (repeat to self)

Can anyone think of five candidates for head coaching jobs? Are they likely to be any better than these guys? Weren't there rumours that Holmgren and Reid might be about to call it a day after the season?

Starshatterer
Obviously I don't base that statement on any "inside" information. It was based on Boston media reports, which admittedly, as you say did not like BB (some even to this day as we all know).

But the rumblings did get quite loud when they were sitting at 0-2 to start the season out and playing AT NYJ.

Had they finished the year out with Drew and a .500 record, than yeah you're right, Kraft most definately would have given him another year.

Saying someone is on the chopping block doesn't neccisarily mean that they're almost certain to be canned after that year. Tom Coughlin immediately comes to mind. He's been on the hot seat for about 2-3 years now. Same could be said with Fox.

It's hard to know who GMs are looking at for head coaching jobs. Much less who will be successful. For instance, did you ever hear of Lovie Smith or Sean Payton before their head coaching jobs?

However, if I had to guess at some names, there are the obvious Cowher and Schottenheimer, then the Ryan brothers, and maybe people are still looking at Rivera as a head coach candidate. One interesting thing is that most of these coaches are from the defensive side of the ball. I don't know who a team would go after if they want an offensive coach.

The Ryan in Oakland ain't gettin' a hc job until he gets a haircut and shaves. That may not be entirely rational or fair, but guys who look like him, outside of the movie or music business, don't get to be put in charge of 100 million dollar payrolls.

Re #14: Can anyone think of five candidates for head coaching jobs? Are they likely to be any better than these guys? Weren’t there rumours that Holmgren and Reid might be about to call it a day after the season?

Schotty and Cowher are no-brainers. After that, you've got your stud coordinators (Ron Rivera, Jim Bates, Al Saunders), your second-chancers (Greg Williams), your hot college candidates (Pete Carroll), and whatever 13-year-old Davis hires to replace Kiffin at the end of the year (speaking of which, why isn't Lane Kiffin's name on the hot seat list?)

As for whether they're likely any better... Cowher and Schotty are definitely better than any names on that list (with the possible exception of Fox and Fisher). Rivera and Bates also look like the real deal to me (Bates was great as an interim coach in Miami, and they were fools to grab Saban instead of promoting Bates). I'm not sold on Saunders or Williams, and I am always against college coaches coming to the pro game.

"I’m not sold on Saunders or Williams, and I am always against college coaches coming to the pro game."

That's just crazy talk. The 'ole Ball Coach' won games in DC while only putting in 6 hour work days.

Re #16: It’s hard to know who GMs are looking at for head coaching jobs. Much less who will be successful. For instance, did you ever hear of Lovie Smith or Sean Payton before their head coaching jobs?

Actually, Lovie Smith was a big name around here at Footballguys. He was the guy who turned St. Louis from an offensive juggernaut to a defensive juggernaut. St. Louis ranked 26th in Defensive DVOA before Lovie showed up, jumped to 5th in his first season, was 12th in his second year, was 5th in his third year, and then fall all the way back to 28th when he left. I remember there was a lot of discussion about how the general perception of St. Louis at the time was out of line with reality, as well as about the incredible rate at which Lovie's defense got turnovers (St. Louis had a ridiculous 46 takeaways in 2003).

Jim Schwartz, Pete Carrol, Kirk Ferentz and Mike Singletary are usually worth a mention for head coaching jobs. If I was an owner I'd also take a look at Dean Pees and Dante Scarnecchia from the Pats. If Bates does a decent job in Denver then he should get a job though he should have been hired in Green Bay and Miami which suggests that he might not interview very well, something that may also be true for Rivera.

Which coach would you rather have?

A) Coach A has a .530 career winning percentage. His teams have made the postseason in 31% of his seasons coaching (including partial seasons). His teams have won close games in the postseason, (.800 record in games decided by less than 7 points), but have not failed well in games decided by 7+ points (.250).
B) Coach B has a .613 career winning percentage. His teams have made the postseason in 62% of his seasons coaching (including partial seasons). His teams have lost close games in the postseason (.272 in games decided by

Wouldn't Pete Carroll be considered a "second-chance" coaching candidate?

I would think, barring a total collapse of his defense this season, Mike Smith might get some looks for a headcoaching gig. He's the DC in Jacksonville under Jack Del Rio and you have to give them a lot of credit for what they've done over the last 4 years. Realistically, Jacksonville had never really had more than an "adequate" defense. Last year was probably Smith's greatest accomplishment, having lost MLB Mike Peterson, DT Marcus Stroud and SS Donovin Darius three guys right in the heart of the defense, all pro-bowl calibe guys, for much of the season and still pulling off a great defensive performance. Not to mention Reggie Hayward.

Cowher is able to keep his teams consistently competitive, but consistently challenging for the conference championship, no.

It's really about keeping teams consistently competitive that makes a good coach. Do that long enough, and eventually, like Cowher, Dungy, Shanahan, etc., you'll probably win a SB (I say "probably" because you always have your star-crossed Schottenheimer's out there). By the time you get to the playoffs, the competition level is so high, both from the other teams' players and their coaches, that it's really about luck.

I'm a die hard Patriots fan, but I'll be the first to admit that their multiple SB runs couldn't have happened without luck. Only once in the Belichick era were the Patriots clearly the dominant team in the league (2004). In 2003 and 2006 they were among the dominant teams, and in 2001, 2002, and 2005 they were very good but not great teams. It was the fact that they had six years running of being a good to great team that put them in the position to win at least 1 superbowl, but it was luck that they won three. That luck could have easily gone to the Colts or the Steelers.

But from the owner's standpoint, as long as your team is competitive every year, you're going to sell tickets and merchandise and make money. Hence that, and not SB rings, ought to be your primary criterion for a coach.

For what it’s worth, Bill Belichick was on the chopping block going into the 2001 season.

I don't think Belichick was on the chopping block in 2001. The Pats had given up first round draft picks to a division rival to get him, which probably made Kraft more willing to stick with him for at least another year. Besides, although the Boston sports media didn't like him, it was widely accepted among Patriots fans (at least the ones I interacted with), that Pete Carroll had pretty much destroyed the good team that Parcells had built in the mid-90's (with some help from Parcells himself in New York), and hence the poor 2000 finish was not in the least surprising. I would hope that the Krafts realized the same.

"some may pull an Eric Mangini and make something out of nothing"

I think Mangini is an outstanding coach who will be even better, but it's a little disingenuous to say he made something out of nothing. The Jets had a lot of young, hidden talent that had a lot to do with their success in 2006 (Cotchery, Washington,Rhodes, Mangold, etc.)

Post #23
"Wouldn’t Pete Carroll be considered a “second-chance� coaching candidate?"

I don't see it that way. Pete Carroll is being lusted over because of his College success not his NFL experience(s), as oppose to someone like that fat-ass at Notre Dame, where he has a pretty good track record as a OC (but a bad Gastrointestinal Track).

MJK
Were you intending to lump my post in with the other one?

Re 22: I think NewsToTom's boss snuck up behind him. If anybody here can help him, he'll be looking for a new job about now.

I think he got caught by the dreaded "less than" symbol causing a premature end to his post. I'm certain he was about to reference some stat about games being decided by less than 7 points.

Re: Gruden

"[T]he offense that worked so well in 2002" (as Judge puts it) finished 21st in offense DVOA. In fact, in terms of offense DVOA, the Buccaneers have finished in the bottom half of the league and in the negative in all five seasons that the alleged offensive genius Gruden has been the head coach, culminating in last year's pathetic performance (30th ranked, -17.9 DVOA). The last time that the Buccaneers finished the season in the top half of the league in offense DVOA was 2001, and the last time that the Buccaneers finished the season with a positive offense DVOA was 2000. Note that those are the two seasons that immediately preceded Gruden's tenure.

I'd also point out that Jacksonville has only picked up remotely high-profile free-agents on defense (Mike Peterson, Reggie Hayward) -- guys who weren't really on the radar as anchor-type guys. And until this year, Jacksonville had gone Offense in every first-round of the draft (Leftwich, Reggie Williams, Matt Jones, Marcedes Lewis)

Marty Mornigwheg's (or something like that) play calling helped save the Eagles last year and if they perform well again (Super Bowl? Pretty please?) I imagine he'll have some sizzle.

Re #28
Coach A is Jeff Fisher. Coach B, as the dismal playoff record might indicate, is Marty Schottenheimer. I just think it's amazing how much one good playoff run and a dozen years in the same place can color our perceptions of how good a coach is. Mind you, I'm not trying to argue Fisher isn't a good coach, just that his W-L record doesn't indicate why he's thought so highly of.

Marty Morninwheg? Are you serious? Maybe he's a good OC, but he was horrible as a head coach for the Lions. I would be shocked if any owner hired him again as a head coach, if for no other reason than to not have to answer questions like, "How could you hire a guy who in his previous head coaching job won a coin toss in overtime and chose to kick off"?

On the other hand, Bill Belichick didn't set the world on fire in his first head coaching job, either, and he seems to be doing OK now.

16: As kibbles said, Lovie Smith wasn't exactly an unknown when the Bears hired him. He was one of the hottest coordinators around, and several teams were interested in him.

Marty Mornhinweg's play calling also includes taking the wind instead of the ball in overtime, back when he was coaching the Lions. He's gonna have to fast-talk his way out of that one during interviews.

Seriously, who is the last Lions HC that has had any success as a head coach after leaving Detroit?

Re #35
Aside from Jauron's interim stint in 2005, the previous Lions coach to get a job with another team was George Wilson (1957-64), who coached the Dolphins from 1966-69 to a 15-39-2 record. Raymond Parker (1951-56) was immediately hired by the Steelers and put up a 51-47-6 record from 1957-64. Alvin McMillin (1948-50) has the best WinPct; he was 2-0 with the Eagles in 1951, then found out he had stomach cancer.

It says ... something about a franchise when nobody who's coached you on a full-time basis in the last 40 years has been hired by another team.

You could say something similar about the Steelers. Of course, the reason why no one has hired one of their ex-coaches in a really long time is that they only had 2 head coaches between 1969 and 2006. And 1 of those 2 ex-coaches (Cowher) proably will end up coaching somewhere in 2008.

NewsToTom,

I don't think most people would say that Jeff Fisher is obviously a better coach than Marty. Fact is, Marty has had two separate careers: He's a HOF coach in the regular season, but a lousy coach in the playoffs.

Fisher is a good coach, but he's not one of the best in the league. Marty is one of the best, but only until January. Then he turns into a pumpkin.

The anti Marty? John Fox. 44-36 in the regular season (.550), 5-2 in the playoffs (.714). It's a small sample, but remember he was also DC for the Giants team that overachieved and went to the Super Bowl against the Ravens.

Gibbs will not be coaching the Redskins next year.

I don't understand why everybody always jumps to Romeos defense in cleveland. If Mike Mccarthy had the same 10-22 record everybody would want him axed.

John Fox should not be on the hot seat list, and neither should Jack Del Rio. Those guys could end up being the victums of their own early successes.

Parcells, Martyball, Vermiel, Cower, are the top coaching candidates out there. You might think that 2 or 3 of the 4 are off limits, but remember that Dan Snyder brought in Gibbs out of retirement, and nobody thought he would get Martyball. Redskins fans are already hoping the Danny opens his checkbook for Cower next year.

Mike Shannihan was on the hot seat the year before Denver went to the AFC championship for the same reasons people want Gruden on the list.

If Kiffens Raider Joes stink up the joint like Art Shell last year, I wouldn't be suprised to see one and done again in Raider nation.

I think the thing to note about Jeff Fischer is that he is just finishing his second round of coming through salary cap hell. When he came to the Oilers, they had just gotten rid of the core of their Warren Moon teams. He was 32-38 in his first 4+ seasons, and then went to the Super Bowl in his 5th. He won 10+ in 4 of the next 5 seasons and then started shaving salaries. They had 2 bad seasons, and then back up to .500 in 2006. I question whether they were really that good in 2006, and wouldn't be surprised to see another 7 or 8 win season in 2007, and then more improvement in 2008.

Jeff Fisher can coach my team.

It's pretty clear Bill Belichick is becoming the patron saint of failed coaches. Every coach who performed poorly at his first job can point to BB and say 'look what can happen if you give a coach a second chance!'

I would think most guys here at FO would understand Schottenheimer's playoff record means squat.

First, would someone explain to me what is different about playoff football compared to the regular season? The rules are the same. There's the intensity, but it's already insanely high during the regular season. The "intensity" is mostly in the media hype.

The only difference is the strength of the opponent. We should expect good coaches to have worse playoff records than regular season records.

Schottenheimer is 5-13 in the playoffs. That's really only 4 games below .500. We would expect, in general, a good coach to have a 50/50 shot at winning any given playoff game. Doing the math, 1 in 10 coaches with a 50/50 shot at each game would be 5-13 after 18 games.

He is well within the bounds of pure luck. We can't say he is a bad playoff coach on the basis of his record alone.

I didn't say that I thought Morningwheg was necessarily a good head coaching candidate, only that he would get a look. Maybe he's learned something working for a successful franchise.

There is a lot more pressure during a playoff game. Maybe Schottenheimer isn't good at psychological part of coaching. It looked like to me that Rivers didn't settle down until late in the game. Plus, since he has his reputation, players feel even more pressure to come through for their coach. Also, Schottenheimer isn't a great strategist and usually relies on out executing the other team. Maybe against better teams the strategy becomes more important?

Morningwheg's problem isn't that he hadn't worked for a successful franchise before becoming a head coach. He was an assistant coach for the Packers when they won the Super Bowl. He also was an assistant coach for some successful 49ers teams in the late 1990s. Presumably, he learned something from those experiences. However, it didn't look like it based on his tenure with the Lions.

The vast majority of assistant coaches who become head coaches were from successful franchises. The success of their teams obviously had a lot to do with their attractiveness as head coaching candidates. Who wants to hire an assistant coach from the Lions, the Raiders, the Texans, the Cardinals, etc., as head coach?

But just because they worked for successful franchises as assistant coaches doesn't mean they learned anything that will make them good head coaches. For example, look at Norv Turner and Dave Wannstedt. Turner was offensive coordinator for two Cowboys teams that won the Super Bowl, and Wannstedt was defensive coordinator for the first of those teams. Their success with the Cowboys as assistants hardly has translated into success for them as head coaches.

42, 44

I think the biggest knock on Marty is that his teams are built for the regular season but not the playoffs. How is that possible? Maybe they are undertalented and overachieving, but get exposed against quality opponents. Maybe he is too conservative and manages them into the ground instead of taking a risk when necessary. (Parcells in particular is known for mixing in a trick play at the right moment.) Maybe his teams are designed to get an early lead and sit on it, but are helpless when they fall behind against somebody decent. I think the last explanation is the most likely, but it may be elements of all 3.

Brian,

Also, you missed a key aspect of playoff records: You can't play 18 games in a row. When you lose, you're done. Marty isn't 4 games under .500 because if he had won 4 games instead of losing, he'd still have a record of 9-13, assuming he didn't win a Super Bowl. Let's say over the course of those 4 additional wins he also won 1 SB. In that case he'd have a record of 9-12. Don't you see? He's not barely under .500 due to some fluke. He's been in the playoffs 13 times (hence 13 losses) and in those 13 opportunities in the playoffs he's managed only 5 wins! To be at .500 he'd have to have won 13 games over 13 playoff appearances (assuming no SB win). So he's actually 8 games under .500, not 4. 8 games under .500 over the course of 18 games is quite poor, especially when you consider that 1) Marty has a fine regular season record of .613, and 2) How many coaches in history have coached 18+ playoff games and done as poorly as Marty? Off the top of my head, I'm pretty sure the answer is zero.

44- but if Marty is bad at game planning against playoff teams, then we should look at his record against playoff teams in the regular season, I doubt he has a 27% winning pct. against playoff teams in the regular season. The thing that also sticks in peoples minds were the upsets, and "the drive" that gets shown over and over on NFLN/ESPN and all the commentary about his record.

40- Jeff Fisher is a top notch coach. He is a good teacher, smart, good people skills, knows when to try trick plays etc, and manages the game well. He ran circles around Dungy in the second Colts/Titans game of this year with that whole punt/field goal fiasco. Dungy looked Mike Martz foolish after that game.

#46:

Mora The Elder is winless in the playoffs but I think he only coached in 8 or 9 games. Otherwise I'm with you, I can't off the top of my head think of anyone who's coached in that many postseason games with that poor a record.

You can quote numbers at me all you want but Marty has lost too many times in too many ways (especially when his team is the heavy favorite, has more talent, etc.) for me to believe he's anything other than a playoff choker. He needed Joe Montana to drag him kicking and screaming to the AFC title game in '93--the only time in TWENTY TWO YEARS he's ever won as many as 2 playoff games in a season. And he hasn't won a playoff game since that season either. No playoff wins in the past 14 years is something a bit more than just bad luck.

If I was a GM, I would only hire Chokenheimer( as I like to call him) if my team was an absolute bottom feeder. Then after he got the team turned around I'd fire him and bring in someone else --almost any halfway competant coach would do--to actually win me some playoff games. That's about all Marty's good for. And to give him credit he is very good at that part of it.

Looking in my NFL Record & Fact book, here are some other lousy playoff coaches.

Mora The Elder is 0-6--the only coach in NFL history who hasn't won at least one playoff game I believe. Sid Gillman is 1-5, George Allen is 2-7 (both wins in '72 of course), Don Coryell is 3-6, Dennis "Crown Their Asses" Green is 4-8, and Steve Owen is 2-8. Chuck Knox, who was the Marty of the 70's with the Rams, has coached the same number of games as Chokenheimer but is 7-11.

Before the title run, Dungy was 5-9 and before his title run Cowher was 8-9, not that bad actually. But people choose to defend Marty's playoff record and bash his, unbelievable.

To put Marty's playoff foibles in perspective, Sean Payton won 2 playoff games last year in his first playoffs ever as a head coach after being thought of as an incompetent nobody not very long ago. If he wins one more playoff game next year (certainly not outside the realm of possibility) he'll be over halfway to Marty's playoff win total over 22 years in only his second season.

BTW I think Cam Cameron may have set a record for chopping block expediency after the Ginn pick. I don't know any other coach who was hired in March and on the chopping block before April was done.

Playoff record is a bad stat for coaches IMO. For starters, there are SO many players/other coaches on their team who can stuff up eg. Tony Romo. It is such a small sample size that a bad series of coincidences or a couple of bad matchups for your team could mean that you have an undeserved reputation as a choker. Please note: I am not saying that this is what has happened to Marty - I'm just talking generally, since I don't give a rat's about Marty.

I mean my basketball team has lost its last 4 finals, are we chokers? Is our coach crap? No, at least we made it there. Here's what happened: one year a guy hit a fluke shot on the buzzer, another year our gun point guard didn't show and we got pressed to death, another year our opponents were just much better and I can't really remember the other one.

It also works in reverse. Don't the Patriots have an insane record in close games? When did that suddenly become clutch play and genius coaching, rather than luck. Surely the small sample size and huge amount of variables can hardly confirm that Bill B is a coaching genius. Sorry, that's sounding a bit anti-Pats, of course it could apply to any team/coach, it's just New England would be the most obvious example at present.

RE 2:

I am not going to bother addressing your "Cowher is overrated" garbage since plenty of others have already done so.

Nor am I going to address your criticisms of the 2005 Superbowl run, which was arguable the greatest playoff run in NFL history.

But to say the the Steelers were "slapped around" by Dallas in SBXXX is just wrong. Pittsburgh was down by 3 with about 4 minutes left in the game, they had the opportunity to get the go-ahead TD (or tying FG) when Mr. O'Donnell threw his 3rd INT that cost them the game. The fact that they were still in the game after throwing 2 picks is a testiment to the fact that they were NOT getting smacked around.

It was a 3 point game with 4 minutes left, that is not getting smacked around. Examples of getting smacked around in the SB include...

Bears 46 Pats 10
Ravens 34 Giants 7
Niners 55 Broncos 10

Re 2 (again): Cowher is able to keep his teams consistently competitive, but consistently challenging for the conference championship, no.

I actually do need to address this. Was that statement a joke? Cowher went to 6 AFC chamionship games in 15 years.

Re 1: I’m not sure what people expect out of Bill Cowher. The Steelers were hardly a defensive juggernaut with him at the helm. They were good, but so was the talent at key positions.

Cowher had a top 10 defense for 14 straight years. Top 5 defense 8 times. Your argument is that he had talent at key positions so he cant get credit for the good D. Who do you think brought in that talent?

Re: #46
I think your response to Brian perfectly articlulates why Martyball is a bad playoff coach.

Well done.

Re: 44

Please don't bring up that game as evidence of Marty's playoff ineptitude. I've said it before - the only reason the Chargers lost that game is because of Tom Brady's Satan worship.

Re: 50

Right on. Wayne needs to pull a Dan Snyder and get Don Shula to come out of retirement. Not that I think he would be a savior at this point, but it would at least give me some hope in the offseason. I just hope he wouldn't have to play a team coached by Jay-Z because apparently Jay-Z knows more about football than one of the greatest coaches of all time. Did that commercial piss anyone else off?

"Nor am I going to address your criticisms of the 2005 Superbowl run, which was arguable the greatest playoff run in NFL history."

What makes you say this? Aside from you being a Steelers fan?

46:
I think you misunderstand the term "x games under .500." Although it's not a common football expression, it's used in baseball everyday. If you "flip" 4 of the games in Marty's favor, he'd be .500. It's just a semantic point.

Your point is taken, but mine also stands. First,it doesn't matter if the games are in a row. Besides, his playoff games (which we're discussing) were in a row. There were no other intervening playoff games.

Second, my original point was that 1 out of 10 "good" playoff coaches would arrive at 5-13 after 18 games. A "good" playoff coach being defined as one who would expect to have a 50/50 shot against any given playoff opponent.

Therefore we really can't say Marty is a bad playoff coach based on his record.

Here's another way to think of it: How do we know that Marty wouldn't go 13-5 in his next 18 playoff games (assuming he could coach that long)?

You may be right that his conservative philosophy tends to lose to more aggressive gameplans. But to back up your point you'd have to base it on hard observations beyond his record.

I could probably dispell the myth that Marty-ball is good in the regular season but bad in the post-season by looking at his regular season games against playoff caliber apponents. I'd bet there's no significant difference.

"Please don’t bring up that game as evidence of Marty’s playoff ineptitude. I’ve said it before - the only reason the Chargers lost that game is because of Tom Brady’s Satan worship."

That game is a perfect example of why Marty's teams always lose in the playoffs.

The chargers were clearly the better team in that game, yet after 50 minutes of outplaying an inferior team, they were up only a handful of points. Martys teams go ultraconservative when they get about a 7 point lead. Thats fine in the regular season, when you play a bad team, you'll still pull away. But when you're playing good teams, if they know what you're trying to do every play, theyre gonna stop you.

Marty constantly takes superior teams, and lets bad teams hang with them. The Ravens/Chargers game is another good example of this. The chargers dominated that game for 50 minutes and gave up one drive at the end and lost.

If you dominate a game for 50 minutes, you should have a big enough lead that you can afford to give up a drive.

#39
I do not think SHanahan has been on the hot seat at any time from Bowlen, in fact he begged him to stay in 2001/2002 when Shanahan was being courted for the Florida job. Your perception probably comes from a few local idiot sportswriters who seem to use their forum to channel their personal feelings. Short of saying or doing something really stupid SHanahan has been UN-FIREABLE since Jan 28, 1998.

Fox, Gruden and probably Del Rio do not belong on this list. Fisher and Gibbs are unique circumstances from your traditional coaching transition. Fisher is a short timer in Tennesee because he probably wants to be. Gibbs would be a tough guy to fire, given everything he has done for the franchise (even though much time has passed).

Schottenheimer would be the perfect fit for the Browns. They are bad and he is very good at turning around bad teams. He delivered them their best run of success in the late 80s and would make them respecable again.

Guys like Elway and Brady always get the best of Marty in big games, and in many cases he has been un-lucky. I will say that often his team's over-achieved to get to the high seed in the playoffs. The '97 Chiefs team was in opinion his best team and just as good if not better than Denver and Green Bay. His decision to start Grbac over Gannon in the Denver-KC divisional game along with a stupid fake field goal attempt eliminated by far his best team.

In order to proclaim that you "know" Schottenheimer is a bad playoff coach, you pretty much have to discount the signifigance of sample size when evaluating a data set. It is a bit of a puzzle at to why this a desirable thing to do.

For coaches with even a smaller sample size of playoff games to examine, compared to Schottenheimer, using playoff records to evaluate coaching performance is so silly it is hardly worth discussing.

58:
I like your thinking. Lets take it further.

Here's how Marty's playoff losses occurred:

'85 21-24 vs. MIA
'86 20-23 vs. DEN
'87 33-38 vs. DEN
'88 23-24 vs. HOU
'90 16-17 vs. MIA
'91 14-37 vs. BUF
'92 0-17 vs. SD
'93 13-30 vs. BUF
'94 17-27 vs. MIA
'95 7-10 vs. IND
'97 10-14 vs. DEN
'04 17-20 vs. NYJ
'06 21-24 vs. NE

That's 9 out of 13 losses within a touchdown. We'd have to know the average playoff point spread for the period for comparison, but I think it's safe to say his conservative play has an effect. Like you said, it probably lets good teams have an opportunity to come back in the final minutes.

On the other hand, lots of close games also indicate that luck is a strong factor. (When the game is close, one blown call by the ref or a favorable "spot" has a relatively important effect on the outcome.)

Rich, Ted Williams had a .200 slugging percentage in 30 post season plate appearances. I suppose one should then say, "Ted Williams constantly took his superior talent into post-season play and performed poorly against inferior talent. He was a bad post-season hitter."?

Any game with a 4 point margin of victory or less really is essentially a tie, where any funny bounce or odd penalty called can provide the margin. Out of 18 playoff games, Schottenheimer's teams have lost five outside of that margin. His team won two games by more than that margin. The other 11 are extremely close games. Judging a guy on 7 decisive playoff outcomes, as opposed to over 300 regular season games, seems really strange.

If somebody wants to do the work, and get Marty's regular season record against teams that won at least nine games, or late season games against winning teams which were fighting to get in the playoffs, and compare Marty's record in this regard with some of his acclaimed peers, and that comparison shows Marty to markedly inferior, I might give the "Marty is a bad playoff coach" notion more credence. Until then, it strikes me as wildy speculative.

Brian,

I know what games under .500 are. However, the concept doesn't translate to football playoffs because flipping a loss into a win means that you advance a round and face a new game (which Marty would presumably lose). So if he had won 4 more games instead of losing, he would also have played 4 extra games, which implies 4 more losses in the next round, still adding up to 13 losses.

You have to look at each playoff appearance as a separate event. It's an opportunity to accrue wins while almost certainly ending in a loss. In those 13 appearances, Marty has ended in a loss each time (no SB wins). He has won 2 games once, 1 game 3 times, and 0 games 9 times. Just to be anywhere close to .500, he would have had to win more than twice as many games as he actually did.

As to close games and luck: Yes, luck is a big factor in close games. However, Marty's 18 games are longer than a regular season. There's luck, and then there's consistently bad luck which implies that you're doing something wrong (e.g. being too conservative).

A plate appearance is a lot different than a football game Will.

A plate appearance is a single occurance, with at most, 10 or so pitches.

A football game is the result of 100 or more individual plays.

"On the other hand, lots of close games also indicate that luck is a strong factor. (When the game is close, one blown call by the ref or a favorable “spot� has a relatively important effect on the outcome.)"

Right, the point is, Marty consistently puts his teams in a position where a single bad call/unlucky bounce loses the game.

Oh, I almost forgot about the "Marty vs. good teams in the regular season."

Of course Marty did worse against good teams and padded his wins against bad teams! Everybody does that. Find me a coach who does equally well against good and bad teams (besides Jack Del Rio). It doesn't prove anything unless he was really bad against good teams, worse than most other coaches. (I still think he's a bad playoff coach, even though here I seem to be defending him.)

Will,

Why is 4 points the magic number? Why not 3?

All things Marty, including the significance or lack thereof of his playoff record, were debated at marathon length when he was fired last February. Link to the XP in my name. Great thread.

Rich, do you really think that the coaching within the game leaves a discernable fingerprint on 100 discrete plays? Yes, the plays are mostly called by the coaches, but there is not a limitless universe of plays to be called. Purely random playcalling would have considerable overlap with plays conciously decided upon.

Again, you have an interesting hypotheses. You are, however, a very long ways from giving that hypotheses a solid empirical base. There's just no way to get around a tiny sample size.

Mcnabb, I can't remember where I saw the statistical basis for 4 points being the breakpoint, as opposed to three. I'll see if I can find it.

Whatever somebody thinks about the "Marty is bad playoff coach" theory, which I would label very speculative, swapping him out for Norv Turner is a really, really, bad decision.

Does anyone else think that since Marty has a reputation for being a bad playoff coach, that it puts even more pressure on his players, and then it becomes a self-fulfilling prophecy. His players are trying too hard, pressing, so they perform worse, then he loses games he should have won.

#57: "Here’s another way to think of it: How do we know that Marty wouldn’t go 13-5 in his next 18 playoff games (assuming he could coach that long)?"

Simple. I've seen him coach. It'll never happen. He plays it too conservative, letting lesser teams stay in the game. And while other coaches ramp up their efforts in the post season, his teams often seem underprepared by comparison.

#71: "Does anyone else think that since Marty has a reputation for being a bad playoff coach, that it puts even more pressure on his players, and then it becomes a self-fulfilling prophecy. His players are trying too hard, pressing, so they perform worse, then he loses games he should have won."

Short answer: No.

Long answer: I don't believe for a second that the pressure of trying to get your coach a win in the playoffs could possibly compare to the pressure of, you know, getting yourself a win in the playoffs, and possibly a Super Bowl ring. No way. Everyone faces lots of pressure in the playoffs. One could even argue that playing for a coach with a good playoff record might cause even more pressure, because the expectations are so much higher. Honestly, I think the whole "playoffs are more pressure filled situations" explanation is just a solution looking for a problem. There's no reason to bother explaining something that's probably just bad luck anyway.

Here are a few things to consider about Marty's playoff record that I think are being overlooked:

He got the Cleveland Browns to the AFC Championship Game, twice in a row! He got the Browns within a field goal of the Super Bowl in 1986. No other coach has gotten them within two touchdowns of a Super Bowl. In only four full years as head coach in Cleveland, he won two playoff games. After he left, the Browns only got to the postseason three times in 15 seasons and won only two playoff games in those three postseason appearances. That includes a full five year span in which Bill Belichick, everyone's favorite playoff guru, coached the team, leading them to only one postseason appearance, in which they lost in the divisional round.

In fact, in the entire Super Bowl era, the Browns have been 4-10 in the playoffs in 34 seasons without Marty, and 2-4 in 4 seasons with Marty. So, on a team that only got to the playoffs 10 times in over 30 years without him, a team whose winning percentage in the playoffs in those years was below 29%, he got to the playoffs four straight years and won 33% of his playoff games.

Now, I'm not saying that this is proof that Marty's good in the playoffs, or even proof that he isn't bad in the playoffs, but I think we need to consider his record in context with the teams he's compiled it with.

A lot of the teams Marty's coached for were pretty awful in the postseason without him. After a quick check, of the three teams he's spent more than one year coaching, the one with the best postseason record without Marty is Kansas City, which has a 5-6 record in the postseason in 37 seasons without Marty. Not exactly the stuff of legends. Considering the teams he's coached, I don't think it makes sense to put all of the blame on Marty.

re 58: my recollection of the game is that there were an unusually high number of drops by the Chargers' WRs (especially Eric Parker) which kept them from pressing home any early advantage.

71: I definitely agree with that. I would even include Marty himself as someone that feels the extra pressure and thus performs worse. Just look at some of the ridiculous decisions he made against the Patriots last year. He tried so hard to overcome his reputation as too conservative that he swung way too far the other way and made some incredibly aggressive and foolish decisions (like going for it on that fourth and long early in the game).

Bum Phillips once famously praised Don Shula by saying, "He can take his'n and beat yours'n and take yours'n and beat his'n." Replace the word "beat" both times in that quote with "lose to" and that would describe Marty as a playoff coach.

Seeing some of these comments suggesting that Marty isn't really a bad playoff coach and that he could go 13-5 in his next 18 playoff games if he coached that long make me laugh. They remind me of comments by Roy Williams last year that it was stupid how close the Lions came to scoring 40 points in a game in which they scored something like 6 points.

It seems to me that Marty's teams inevitably find a way to lose close playoff games. Because it has happened so often, it seems fair to say that he is a bad playoff coach.

I always find the explanation of Marty's playoff record as "he's too conservative" a bit unsatisfactory. Firstly, what does this mean exactly? Too reliant on the run? Or something more comlicated than that? Secondly, considering that Dan Reeves was considered conservative too, and managed to go 11-9 in the postseason including 4 Superbowl berths would suggest that conservative might not always be a bad thing.

Here's something interesting about Marty's 13 playoff losses:

3 losses to John Elway
3 losses to Dan Marino
2 losses to Jim Kelly
1 loss to Warren Moon
1 loss to Tom Brady

then,
1 loss each to Stan Humphries, Jim Harbaugh and Chad Pennington.

Interesting that 10 of his losses are to HOF QB's (assuming Tom Brady is a HOFer).

I think Schottenheimer deserves his own thread. He obviously draws out some strong opinions.

If I have time tomorrow or Sat, I'll dig into the records and do the math. I'll post what I do on my site.

My personal style is that Marty is too conservative. He's limiting his teams' potential by playing a style that worked well in the 70s when he learned the game. I just don't think his record alone proves anything.

He and Gibbs share the same philosophy. I think the only difference between them is that Gibbs was blessed with better talent and was a bit luckier.

Is the Marty Schottenheimer thread finally back? I miss the good times. A couple of points:

As mentioned in 68 there is an older thread that delves into this stuff in pretty good detail.

74: Bud Carson made the AFC Championship Game the year after Marty left the Browns. I believe Gunther led the Chiefs to the playoffs the year after Marty left KC as well.

70: Will, we've each covered this ground numerous times, but the swap wasn't Marty for Norv. The choice was Marty and a new GM or Smith (the current GM) and Norv. You can make the case that that combination was a downgrade but I think it's important to keep in mind Smith's value.

I don't think anyone doubts that Marty has been a victim of bad fortune in the playoffs. (Although people seem to forget that he's won a few games in the playoffs he probably should have lost). That being said, Marty the playoff coach isn't the same as Marty the regular season coach. Marty the playoff coach does strange things (like the 4th down decision to go for it in the 1st quarter against the Pats) and stupid things (like personally getting a 15 yard penalty versus the Jets at a key point in the game a few years ago). These aren't things Marty the regular season coach does.

Seeing a set of 18 described as being consistent with the descriptions "inevitably" or "so often" makes me laugh. Same with the mind-reading that is so often attempted when Marty's playoff coaching is analyzed. A lot of these folks should be on ESPN, playing 5 card hold'em, and winning millions, based on what is ostensibly accomplished in the Schottenheimer analysis industry.

Doubleb, you are still drawing your conclusions about the differences between playoff Marty and regular season Marty off a very small smaple size. I wish just once someone would explain why sample size becomes a trivial matter when discussing Marty Schottenheimer.

Regarding Smith and Schottenheimer, or Smith and Turner, this is where ownership is to be faulted. The owner has a responsibility to either get Smith and Schottenheimer to have a workable relationship, or to break them up at a point where Norv Turner is not the best available alternative. If Smith's position is that he would rather have Turner than anybody else coaching this team, it's time for Smith to go as well. A gm with such poor coaching evaluation skills is a disaster waiting to happen.

Will: If the Patriots and Chargers had switched head coaches before their playoff game last year, do you think the Chargers would have beaten the Patriots? If Bill Belichick coached the Chargers, I think they would have won that game handily.

As it was, the Chargers dominated the game and should have won easily, but they didn't. As I watched the game, I fully expected the Chargers would blow it, notwithstanding their apparent superiority. Why? Because of Marty. That is what I meant by using the word "inevitable." To me, it was inevitable that Marty and the Chargers would find a way to lose a game that they should have won easily.

Based on the fact that you obviously have been a long-time follower of the NFL and thus have a lot of perspective, I'm quite surprised that you are so willing to defend Schottenheimer as a playoff coach.

Much like John Elway and Peyton Manning, Marty is only good for the regular season and is completely incapable of winning the Big One.

Oh, wait...

"If Smith’s position is that he would rather have Turner than anybody else coaching this team, it’s time for Smith to go as well. A gm with such poor coaching evaluation skills is a disaster waiting to happen.

let me first preface this with the fact that I know nothing about Norv Turner.

It seems to me that Marty's greatest strength is that hes great at putting together a team, and bad at in-game coaching.

It seems Smith brings the same ability (just looking at the way hes consistently raped people in draft day trades). If Turner is even a half decent game-day coach, hes an upgrade over Shottenheimer, because they dont need someone to build a team at this point. Its built.

"
Interesting that 10 of his losses are to HOF QB’s (assuming Tom Brady is a HOFer). "

How many of those guys HOF chances are seriously negatively impacted if they dont win any of those playoff games?

Also, are there any better than average QBs from the 80s and 90s who won a playoff game or two who aren't HOF'ers?

Yes, Marko, you assert that you can read Marty's mind, and you think a sample size of 18 gives you sufficient data to judge trends or outcomes as being inevitable, and when you hear someone say that a second sample of 18 might precisely mirror the first sample of 18, you laugh at the remarks. I get it.

"Yes, Marko, you assert that you can read Marty’s mind"

Huh? What are you talking about? I never said I could read his mind.

Rich, I would take issue with the notion that a team is ever finished being "built". Even a supremely put together team can be made to fail, absent good leadership. The mid to late '90s Cowboys, which had a greater talent edge over the rest of the league than the Chargers do, didn't win as much as they probably should have, because of poor coaching selections by Jerry Jones.

Schottenheimer has coached 340 games in the NFL, playoff and regular season, and has won 60% of them. Turner has coached 142 games in the NFL, and has won 42% of them. Yes, Turner had some terrible talent on his roster in Oakland, but there is zero, none, zip, nada, reason to think that Turner is as good a coach as Schottenheimer, and a lot of evidence to suggest that he is worse. If I'm an owner of a NFL team, and my highly paid gm can't get along with my highly paid coach, and the gm insists on making a change, fine. He damned well better not, however, come to me the next day, and tell me that he's going to hire a guy who has a .420 winning percentage over 140 games.

82: There is no sample size that exists I can come up with that would satisfy you. Absolutely none. I could go through every tape of every playoff game he's coached and not find enough examples. That doesn't mean he doesn't have a negative influence on his team in the playoffs.

#85, I was thinking about that after I posted the list of QBs. It's a bit of a self-fulfilling prophecy. But if you look at the examples, I don't think it applies. Brady is probably already in the HOF before he played against the Chargers.

Warren Moon did not make the HOF based on his playoff exploits.

John Elway became famous by beating the Browns and then playing in the Super Bowl each year after he beat them. If he had lost those 3 games, would he have had a long enough leash to still be the Denver QB in 97 and 98 when they won the Super Bowl? Not sure. Would Elway still be a HOFer if his career was exactly the same as it was, except he had lost to the Browns each time he played them? Probably.

Marino is still probably a HOFer if he loses those games to Schottenheimer. Like Moon, his HOF worthiness was not based much on playoff success.

Jim Kelly might be the one exception. Had he lost 2 times to Schottenheimer, then he only plays in 2 Super Bowls. I think he was a borderline HOFer as it was, and 4-straight AFC championships probably made the difference.

Marko-

"I definitely agree with that. I would even include Marty himself as someone that feels the extra pressure and thus performs worse. Just look at some of the ridiculous decisions he made against the Patriots last year. He tried so hard to overcome his reputation as too conservative that he swung way too far the other way and made some incredibly aggressive and foolish decisions (like going for it on that fourth and long early in the game)."

-That's a pretty good description of mind-reading. If you really have such magnificent, detailed, insight as to what is occurring in someone's brain, based off what you see on television, you are to be congratulated.

No there probably isn't, doubleb. Why? Because nobody has ever coached in enough playoff games to establish that they are different coach in the regular season than they are in the playoffs. Which is why using tiny samples consisting of playoff games to attempt to identify trends or discrete phenomena, for any coach, is such an empirically dubious exercise.

Why do you think the concept of needing large sample sizes to have confidence in the persistence of a trend or validity of an identified phenomena is a matter of satisfying me? I mean, these are well-accepted concepts, across nearly countless fields of study. Except in MartySchottenheimerology, apparently.

Will, yeah, the sample size is 18, but if you can say that 18 is not significant in this case, can we really ever judge ANY COACH ON ANYTHING?

An average playoff coach should have won 12 games or so, in 12 playoff attempts.
You go 50% in the first round. 6-6. You go 50% in the 2nd round, 9-9. 50% in the 3rd round, and you're somewhere between 10-12 and 11-10.

Marty is 5-13 in 13 attempts.
Hes 4-9 in the first round, and 1-4 in the 2nd round(of playing, not of the playoffs).

Hes 0-4 in the divisional round after a BYE. All other home teams in the divisional round are 36-4.

No, theres not a huge sample size, but the evidence so far points more to the fact that hes a bad playoff coach, than a good one.

"I’m an owner of a NFL team, and my highly paid gm can’t get along with my highly paid coach, and the gm insists on making a change, fine. He damned well better not, however, come to me the next day, and tell me that he’s going to hire a guy who has a .420 winning percentage over 140 games."

What if Turner's entire issue is drafting/player evaluation, and Smith doesnt intend to give him that power?

No, Rich, you can't. Playoff records provide fodder for fairly wild speculation, and that's about it. I know it frustrates the heck out of a lot of people, but you just can't identify much with any confidence based on 18 games. People convince themselves otherwise with great frequency, and I have the temptation myself, but 18 games just isn't enough to assert anything with confidence. Sure, if somebody, was 0-18, and had consistently lost to much worse teams, by significant margins,
it would be much more highly suggestive than the what we are seeing with Marty. There isn't anything about Marty's record to really give us anything to work with, in terms of statistical confidence, and that's fine; it's fun to speculate. I just wish people would refrain from asserting that 5-13 "proves" something.

Well, Rich, even if Turner's issue is with drafting/player evaluation, he still presumably has to make the final roster cuts, which can be pretty important, and he still hasn't, over 140 games shown any reason to think he is even close to above-average. Insist on getting rid of Marty? Fine, I guess, but what reason is there to think that Turner is an improvement?

Rich, I just flipped a coin 10 times consecutively (I really did), and tails only came up 30% of the time, just like Marty has only won 28% of his playoff games.

There is one thing missing from this debate, and that is someone actually saying Marty is a good coach.

Lots of people are saying he's bad in the playoffs and pointing to his record.

Will and I are saying that we can't use his record over an 18 game sample to make any conclusions. Marty might be a clown in the playoffs, but his record isn't different enough from what we'd expect by chance to prove anything.

Where are the people saying he's a great coach?

To explain better, if I haven't made myself clear, that 30% tails percentage would only be suggestive of something notable about the coin if it were flipped many more times, and the % persisted.

He's a great coach!

Click on my name for an interesting discussion with people who work with statistics for a living, on the topic of Schottenheimer; it even has a dissenter who is more inclined to agree with the anti-Schottenheimer camp.

Re #98
That'd be me, in #22, when I started us down this whole path in the first place, by pointing out Marty has a much better track record than Jeff Fisher, except in close games in the playoffs. Yet, my subjective opinion is that Fisher is widely regarded as one of the best coaches in the league, while Marty comes in for a great deal of ridicule from some quarters, as you can see from the previous Marty XP thread. Napoleon's supposed comment about preferring lucky generals to good ones seems apposite.

Well, I'm just glad to see Will mocking someone other than me for a change :-) Although he's yet to bust out the word "generalizations".

18 games is more than one regular seasons' worth. If a team won between 3 and 5 games in a regular season I guess Will is saying we're not allowed to say they're a bad team or their coach did a bad job. Whatever.

Oh BTW I did point out way back in #48 that Marty was very good at turning around bad teams. He just can't get them any playoff wins after he's made them into contenders. At least, not in the last 14 years.

Hell, Marty's so bad in the playoffs he can't even win the big one in the era of teams and players who can;t win the big one finally winning the big one (Red Sox, White Sox, Cowher, Peyton, Dungy). Sample size schmample size. It sure didn't take 18 opinions from me for Will to get a bad impression of me did it? As I recall, it only took one, maybe two. Methinks Will should follow his own advice...

Dolfan, I have no impression of you other than it is amusing when you, to use a very complicated word, generalize. I'll simply note that my mocking tone in this thread was adopted in response to someone who stated that he laughs at opinions that differ with his. I'd say a mocking tone in response isn't very notable.

As to 18 being more than a season's worth of games, see the piece I linked to. One of the number crunchers addresses that obliquely, in noting that the 18 games would have been much more significant if they had come in succession.

Finally, it is duly noted that you think sample size is a concept to be mocked, which is fine, but it does cause one to wonder why a person with that belief would spend a lot of time at a site where the concept of sample size undergirds the vast majority of articles posted. Not for the first time, you have puzzled me.

As a Dolphins fan, I'd have taken Marty in a heartbeat. The chance to lose a home playoff game would be riches indeed. As it stands we'll see what his OC is about.

Will, some people neither like nor want to understand the simple underpinnings of that very recent science we call statistics. If we could convert the world, Las Vegas would likely go bankrupt and most football fans would have absolutely nothing to say.

And who suffers for it? Poor, beleaguered, misunderstood Marty Schottenheimer.

Oh, gosh, bravehoptoad, I have no desire to covert the world, I have fun in Vegas, and I think poor ol' Marty will muddle through somehow. I really don't mind it at all when people denounce statistical analysis. I'm certainly not a professional number cruncher, although I respect the science a great deal. I just find it interesting that people who don't like that way of looking at things spend a lot of time at a site like this.

So Will, how many playoff games would constitute an adequate enough sample size for you? 30 games? 40? You do realize that nobody in NFL history has coached more than 36 playoff games, right?

My problem with sample size is this: if a guy kills someone else in cold blood for the fun of it, why should I have to wait until he kills ten more people before I can call him a murderer?

Besides, it's 18 games over the course of 22 years we're talking about here, not 3-4 games over one or two years. There's a consistent pattern with Marty, whether you want to admit it or not. And waiting for him to coach 18 more playoff games in order to see if some hypothetical 13-5 record happens is pointless--he'd have to coach until his 90's to do that.

Three of Marty's 5 playoff wins were by 4 points or less. How do we know he didn't get lucky in those games and they really should've been close losses like the rest? Maybe his bad playoff record by rights should be even WORSE. Makes more sense to me than saying most of his 13 playoff losses were flukes.

Dolfan,

After 1 murder, somebody is a murderer.

We could look at Marty's career record in week 5s of the season. It wouldn't tell us much about his overall ability, but the sample size relevance would be similar to that of 18 games of 22 seasons.

Bill Callahan is 15-17 in the regular season and 2-1 in the postseason. That means he coached an average of 1.5 playoff games per season that he has coached. Which record is more indicative of his abilities?

Bud Grant is 158-96-5 in the regular season but 10-12 in the postseason. Which is more indicative of his abilities?

DolFan's post #108 got me thinking. If Marty's never stomped anyone in the playoffs, I wonder what his stomp percentage in the regular season looks like. If stomps are predicative of a team doing well in the post season, then they should apply to Marty too right?

The argument is that Marty is too conservative in the playoffs and particularly while playing with the lead. Isn't that the criticism of every coach that losses?

When Andy Reid was losing all of those NFC championship games, wasn't he Mr. Too conservative? The fans were going nuts the year they "opened it up" against Atlanta with the lead in the 4th quarter. Wasn't Cower, and Marv Levy and all the other guys that lost in the playoffs labeled "too conservative"?

It's a losers curse. If your losing and you get soundly beat, then you stink as a playoff coach OR you just had a less talented team. If it is a close game and the other team wins, they you were too conservative. If your winning and they come back then you were too conservative and should have won.

If a team stinks, like say the Cleveland Browns, sometimes fans have sympathy for the coach because the "no talent" argument pops up and they say that coach never really had a fair shot. However, if a good coach makes an average team artifically good, fans hold it against them when that artificially good team losses to a good or great team ( say Dallas last year). I guess you could simplfy that logic by saying that getting too good too fast, or artificially raising your expectations isn't always a good thing for the life of a coach.

Wins raise expectations. Ideally as a coach, you want to get better each and every year. When Jim Haslet, John Fox, Mangini, Del Rio had some early success, then people will get "bored" with a 9-7 team because they feel that team should be getting better. If your team keeps racking up more wins than expected, it is good in the short term, but can create backlash in the long term if you won't be getting better.

Dolfan, the very fact that you speak of meaningful patterns, within a sample size of 18, indicates that you don't understand the concept of sample size, or you simply choose to reject the concept as a matter of course. That's fine, of course, but in terms of analysis, someone who doesn't accept sample size as a meaningful concept can't really have a useful conversation with someone who does; it is as if one of us is speaking Russian, and the other Chinese. There just isn't any overlap in mutually agreed-upon terms to allow for an exchange of ideas. I don't mean this pejoratively. It's just a reality of communication.

If you'll see the thread, I specifically stated that playoff records are of extremely limited use for evaluating coaches, because the sample sizes are too small. Regarding your second paragraph, you don't seem to understand that we don't need to establish a pattern of murders to call someone a murderer, and thus punish him One suffices.

Like I said, it is as if we are speaking different languages.

I think Marty said during his tenure with the Skins that he tried to have great defense and great special teams. One thing I have noticed is the consistently mediocre to bad wr core. Marty seems to believe that with an advantage on D and ST you will win most of the time. And that seems to be how his teems are built. I don't know if he is too conservative, but the lack of big plays on offense in the playoffs is almost remarkable. I mean, I have no idea how many 40+ yard pass plays any of his teams have had in the playoffs, but I am thinking not a lot. I am not saying that the big pass downfield is the only way to win a playoff game, its not. It just seems like many playoffs wins are partially due to the "WOW" plays. That seems to be a component that is lacking on a marty team.

#111:

So true!

#110:

Marty does have one playoff stomp, a 17-point win over the Colts in '87. His other playoff win was an 8-pointer over the Oilers in which the Chiefs trailed 13-7 in the 4th.

Speaking of the playoffs, Wikipedia has a description of every playoff game of the past 40 years and some earlier years (link should be in my name, I have it on '85-86 but at the bottom you can pick whatever year you like).

#112:

So it's just as I suspected, you're saying that NO playoff games sample size will ever be big enough for you. This way no coach can ever be judged by his playoff performances. I'm sorry, but that won't work for me. Sports itself is a zero sum game. There are clear winners and losers, and certain people and teams are clearly better than others, and conversely some are clearly worse. As usual, we'll just have to agree to disagree. I don't "hate" you nor am I "out to get" you. You just happen to say a lot of things I disagree with, that's all.

#113:

You may have something here. In Cleveland Kosar wasn't exactly known for his big arm, and in KC and SD the WR corps when Marty was there was decidedly mediocre and nondescript. Perhaps Marty is conservative by nature because he's never had that much to work with. But even then after a certain number of years one would have to wonder why that situation wasn't improved. Seems like the Chargers had just as much of a chance at remaking their WR corps this offseason as the Patriots did.

Yes, Dolfan, I know. You've rejected the concept of sample size, and the scientific method generally, as useful tools. Hey, it's a free country. Go nuts.

Will, those things have their purpose but they are not the be-all, end-all of judging things and situations. They are just *a* tool, not *the* tool. Sometimes you just have to put the numbers aside and just go by what you see. And what I (and a lot of others apparently) see is Marty choking in the playoffs year after year with teams that are in almost every case the superior one on paper. He can't win in the playoffs with the worst team, the best team, or any team in between.

Again, if anything he lucked out by having Montana on his side in '93 because otherwise there wouldn't be ANY season in which Marty won 2 playoff games in a row. And even THEN Montana won those games in spite of Marty. I was there, I still remember.

Yeah, I know, Dolfan, you see things. Like I said, it's a free country.

"...Marty choking in the playoffs year after year with teams that are in almost every case the superior one on paper."

You have got to be kidding me.
No way, just no way has that been true.
First of all, what does superior on paper mean?
Last year?
OK sure, I will give ya last year.
But when were the Chiefs and Browns ever superior on paper?
And don't even try 97, thats a case of maybe if you bend the stats every which way they come out slightly better than GB or DB.

There's a lot of stats getting thrown around (and ignored) here but there's one thing I saw slip by that I think should be explored. It was stated that Schottenhemier is 0-4 after a bye, which is fairly remarkable, since the winning percentage of the 1-2 seeds is usually pretty good. Going back to when the NFL went to the 12 team playoff system, playoff teams coming off a bye have gone 53-15, a 0.78 winning percentage.

The odds of going 0-4 in this situation then are 0.22^4, 0.00234 ! If Marty's bye-receiving teams actually had a 78% chance of winning those 4 games, the probability of seeing this outcome are 0.2%!

It's been a LONG time since I took a stats class, but I feel like this could be statistically significant. Will, what do you think?

I vaguely remember that you could construct some sort of statistical statement using "degrees of freedom" when the sample size was small, but the wikipedia entry for "degrees of freedom" didn't help. Am I remembering that wrong?

These other teams will be sorry they passed on Lane Kiffin. He is next great Raiders head coach. Sorry to tell you all that Raiders are back to winning ways.

zip, you really should go read the thread that Fat Tony linked to in #68. The entire thread is worth reading, but your issue is addressed starting about #320. Bottom line: Yes there is reason to make an impressionistic argument that Marty is a bad playoff coach, and as I said in that thread, I have no objection to it. Heck, I make the same sort of impressionistic argument against Mike Martz as a head coach. My only argument is that people are better off not going too far, and asserting that they can empirically demonstrate that Marty is a bad playoff coach with any degree of confidence.

For the record, I'm a big Marty fan and fall on the "small sample size" side of the argument vs. the "playoff choker" side. But let me throw out a contrary thought.

Re Marty and WRs: first, I think Marty values wr blocking more than most fans (perhaps because he is a run-oriented coach). Since there is no blocking stat to factor into wr performance ratings, it often appears that Marty has lesser wr corps than others because we only measure their pass catching performance.

Second, in the salary cap era - approximately half of Marty's head coaching career - almost every team has a weakness somewhere. The essence of a moneyball strategy is find something (Walks/OBP - wr blocking - etc) that makes particular types of players undervalued and create a strength buy spending less than the market to get those players.

Third, Marty is often quoted as saying that a player's most important ability is his "avail-ability" - that is, he prizes durable players who play thru injuries. Again, possibly a successful moneyball approach. It seems to work in the regular season where having good players available every week beats having better players who might miss a few games. But in the playoffs, teams he encounters there have often survived injury and show up with all their better players.

It is just possible that, as famously quoted in the linked article, that this is an aspect of Marty's sh@t that doesn't work in the playoffs.

It's also possible that if he'd had another year in SD, that like Polian/Petyton/Dungy, his sh@t would have finally worked.

Haven't read the whole thread yet, but I wanted to share this while I'm thinking of it. To Brian and McNBGA:

You both make valid points about playoff records, but neither of you are quite right. McNBGA is right in that you can't just look at playoff W/L because the games are played sequentially until a loss occurs, but I don't think the claim that he is "8 games under 0.500" is quite right. He is actually, by my calculations, between 6.375 and 7.1875 games "uder 0.500".

A little probability theory: The number of wins a coach will attain in a playoff appearance is almost (but not quite) a geometric progression. I.e., if you assume that a team has probability p of winning any given playoff game, then the probability of winning 0 games in a playoff appearance is 1-p, the probability of winning one game is p(1-p), of two games is p^2 (1-p), and so forth (it's not quite geometric because you stop after the SB regardless of who wins, so the progression is truncated). There's actually two slightly different distributions, depending on whether you have a 1st round bye or not.

If you know p you can calculate the "Expected number of wins" a team will have in any playoff appearance. To use McNBGA's example, consider p=50, i.e. every playoff team has a 50% chance of beating any other playoff team (which should represent a team that we expect to be "at 0.500"). In this case, the expected number of wins in a given playoff experience is 0.9375 (no bye) or 0.875 (bye) (it is less than 1 because an average team will lose its first playoff game 50% of the time and play no more that season. It's lower for bye teams because there are fewer possible games they can play that will be a win). Hence, in 13 playoff appearances, Marty should have between 11.375 and 12.1875 wins, if his teams had really had a 50% chance of winning each game. He actually has 5 wins. So he's about 6-7 games under 0.500 over the span of 18 playoff games. I don't know if this sample size is large enough to say he's a bad playoff coach, but it's certainly more significant than Brian's 4-game estimate.

I'm not making a stand on whether this implies he's a bad playoff coach or not (my personal opinion is that whoever said he's good at motivation and execution, but weaker on strategy was right, so is slightly a weaker playoff coach than say Belichick or Dungy or Shanahan. But that's a subjective viewpoint--I don't have statistics to back that up). I just wanted to set the record straight about playoff performance and performing under or above 0.500.

MJK,

That was fascinating. I'm not sure I understand all of it, but it seems to support me (that Marty is well under .500) more so than Brian (that Marty is only a little under .500).

One interesting outcome of your math is that if every playoff team has a 50% chance of winning each game, then a record which reflects the expected win probability will actually be slightly under .500 (actually between .467 and .484, depending on byes). Obviously not every game is 50-50, but over enough time these things even out. (I'm sure Will is going to say that there isn't enough time to even things out, and he's right, but let's ignore that for now.)

In any case, while I have argued that Marty's poor playoff performance is meaningful, I absolutely agree with Will that his regular season performance is both excellent and more conclusive. If he had just won a single Super Bowl, he would be Hall-worthy. (Yes, a single game like that does matter, and no, it's not fair, but that's how the system works. Marty isn't a truly great coach if he couldn't win or even reach 1 SB in 13 opportunities.)

18 games is more than one regular seasons’ worth. If a team won between 3 and 5 games in a regular season I guess Will is saying we’re not allowed to say they’re a bad team or their coach did a bad job. Whatever.

We're not talking about a regular season, where you play a lot of bad teams, we're talking about the playoffs, where everyone has a good team.

Let's pick a good-but-not-great team from last season: the Saints. Let's say the Saints played every Super Bowl winner and runner-up from the last eight years. Would winning only five of those games mean the Saints were a bad team, or Payton a bad coach?

The odds of going 0-4 in this situation then are 0.22^4, 0.00234 ! If Marty’s bye-receiving teams actually had a 78% chance of winning those 4 games, the probability of seeing this outcome are 0.2%!

It’s been a LONG time since I took a stats class, but I feel like this could be statistically significant. Will, what do you think?

The first of his bye losses was in 1985 when the 8-8 Browns lost in the first round to the 12-4 Dolphins. This was before the 12-team playoff system. I don't think Marty's playoff suckitude was responsible.

In 1997 the 13-3 Chiefs lost to the 12-4 Broncos. The Broncos actually had a slightly higher DVOA that year, for whatever it's worth. Maybe they should have won this game, but that Chief team only won 3 regular season games against teams with winning records. Denver went on to win the Super Bowl, so they may have been the better team.

The 2004 Chargers lost to the Jets. The Jets were nearly 7% better than the Chargers in DVOA. But this is a game they really should have won I think. I'll blame Marty for this one.

The 2006 Chargers were pretty good. They had a great DVOA despite playing a weak schedule. The Patriots were also very good. Again, I think the Chargers should have won this game.

So, in his bye games, he was 0-4. There's no way he should have been expected to win the 1985 game, so that adjusts his expected record to 3-1. And if the typical bye team wins 78% of the time, we can lop another win off his expectation and say that in his 4 bye playoff appearances he should have gone 2-2.

That still puts him 2 games off expectation, which still isn't good.

I can't believe there is a debate about what "games under .500" means.

If a team finishes a season 4-12, they are 8 games under .500.

You subtract the number of wins from the number of losses to determine games under .500. The term represents the number of consecutive games a team would need to win in order to get back to .500.

Except that with the playoffs, if you win a game, you play another. So their is a variable number of total games.

Richie,

Not only did you not understand how "games under .500" doesn't really apply to the playoffs, you also got the definition wrong. 4-12 adds up to a 16-game season. A .500 record over 16 games is 8-8, so according to the normal definition (which again is not useful here), 4-12 is 4 games under .500, not 8 games under, because the team would have had to win 4 more games (and thus lose 4 fewer).

Hes 0-4 in the divisional round after a BYE. All other home teams in the divisional round are 36-4.

I count 95, 97, 06.
Did I miss a year?
And there have been 68 2nd round games since 1990.
Those three games were super tight losses.
I totally forgot what happened in 95 with the Colts.
The one play I remember is a KC linebacker dropping an easy int near the end zone.
The Colts end zone, looked like a walk in.
Now that I think about it, I have a really hard time remembering any plays from 97?

To be clear, I'm not debating the term "games under 0.500". Obviously, in the literal sense, going 5-13 in the playoffs is going 4 games under 0.500. However, the problem arises because people often use the term "going 0.500" to imply that a team is playing "average"--i.e. that it has a 50% chance of winning any given game.

However, "going 0.500" in the playoffs, i.e. having an 7-7 playoff record, for example, is NOT necessarily indicative that your team was "average" (i.e. that you had a 50% chance of winning any given game) among playoff teams, for the reasons that McNBGA outlines. For example, if you went 7-7 in the playoffs, that could mean that you had seven playoff appearances and each time won your first game and lost your second. However, an "average" playoff team with a 50% chance of winning any given playoff game would actually be expected to win between 6.125 and 6.5625 games, so winning 7 would be performing a little bit better than average.

But even that's not the whole story, because a 7-7 playoff team could have won two SB's, one as a top seed (3 wins) and one as a wildcard (4 wins), and had 7 other "one and done" wildcard playoff appearances. An "average" playoff team with a 50% chance of winning any playoff game would be expected, in these nine playoff appearances, to win 8.375 games (if I did the math right), so only winning 7 would mean the team had UNDERPERFORMED "average" performance, even though it won two SB's.

In my explanation above, I was using the term "under" or "over 0.500" to imply that the team in question was performing worse or better than having a 50% chance to win any given game. However, Scott de B.'s point is worth keeping in mind--a team that performs at a "0.500" level in the playoffs is probaby a team capable of performing significantly above 0.500 in the regular season.

Re 125:

That's true, that an "average" playoff team (with a 50% win probability) will be expected to perform slightly less than 0.500, except for one thing that I outlined in post 132--SB wins. If you don't win the SB, each playoff appearance gives one loss (at some point) and an expected 0.9375-0.875 wins, which leads to a sub-0.500 playoff record. However, if a team happens to win a SB, then it will get 3-4 wins for that playoff appearance and no losses. So even though each playoff is expected to give less than 1 win, each appearance doesn't have to give a loss. Hence it's possible for a team to have a 0.500 record and have actually underperformed its expectation, if there were some SB wins in there (another way of putting it, in more rigorous probability terminology, is that given that a team has not won the SB, its expected win percentage is less than 0.500 if it had a 50% chance of winning every playoff game. Given that you won the SB at some point, that implies that your record contains at least three wins and no losses, so the expected record may be above 0.500). All that the math implies is that this hypothetical 50% team is expected to win 0.875-0.9375 games for each playoff appearance, but whether that playoff appearance involves a loss depends on whether the team wins the SB or not (which has a probability, by my estimation, of 0.625-0.125).

Yes, probability theory can be surprising, but is fun. :-)

I'll poke at the numbers and see if I can predict the actual expected win percentage for this theoretical 50% team...

And incidentally, if you're wondering why it works out that the expected wins per playoff appearance for a 50% team is less than 1, its because of this:

A 50% team will lose its first game in half of all its playoff appearances, which means for this to average out to at least 1 win per playoff appearance, you have to win two games in all the other appearances. However, in half of these appearances you will lose your second game.

OK, I'm sure most non-probabilicians are past caring right now, and I apologize for cluttering up the thread, but in case there are any math people that care, out of curiosity I just did the math further to calculate what the expected win percentage for a 50% team in the playoffs actually is. Now I know there are cases in probability theory where it is not appropriate to operate on the mean (i.e. "reason on the average") in the manner I'm doing, and I don't remember what those conditions are, so this may be wrong. But from what I just calculated, it turns out that, with no information about whether or not a team has won a SB, the expected win percentage for a team that wins 50% of the time is... 0.500. Shocking, I know.

This does not contradict my earlier statement. A wildcard team with a 50% chance in every game is expected to win 0.9375 games each time it appears on the playoffs, so if it never wins a SB, it will have a win percentage of less than 0.500. But that team has roughly a 6% chance of winning a SB, so sooner or later, if enough games are played, they will have a playoff appearance with no losses and 4 wins, which will offset the sub 0.500 percentage from the earlier appearances and give them a 0.500 winning percentage. The same holds true for a top-seeded team.

So the final answer is this. With no information about whether a team has won a SB or not, you expect a team to have 0.875 wins for each 1-2 seeded playoff appearance, and 0.9375 wins for each 4-6 seeded playoff appearance, if it were "average". However, you would expect its win percentage to be 0.500, because not all playoff appearances result in a loss. In Marty's case, we KNOW he has never won a SB, so given that piece of information we expect a sub-0.500 win percentage if his team truly had a 50% chance of winning every game. Just not as sub-0.500 as his actually is.

#42: "Schottenheimer is 5-13 in the playoffs. That’s really only 4 games below .500. We would expect, in general, a good coach to have a 50/50 shot at winning any given playoff game. Doing the math, 1 in 10 coaches with a 50/50 shot at each game would be 5-13 after 18 games."

To build on what Brian said above, I looked at all the coaches I could find who had coached in 18+ playoff games. There were 12 by my count. Some had very good records, some had very bad records, and Marty was the only one with a 5-13 record, although other coaches were not far off. So, just by random chance, we would expect that at least one of those twelve coaches would have a 5-13 record. And there was one such coach. If there were also coaches who were great in the regular season and then "choked" in the playoffs consistently, then we would expect 2 or more coaches to have a record as bad as Marty's - 1 coach who had that record by bad luck, and at least 1 who really was that bad in the playoffs. But that's not what we see.

Bad luck is the explanation that fits the data best, and it fits far better than the idea that some coaches are just natural born losers in the postseason.

Alex,

I think you make some bad assumptions. If there are only 12 coaches in this sample, then how do we know how many of them should be "chokers"? What percentage of coaches are chokers? Is it 1 in 10? 5 in 10? 1 in 20? You don't believe there is such a thing, but you try to prove that there is a lack of evidence without justifying what evidence we expect to find. If playoff chokers are rare, then finding only Marty proves nothing. Maybe Marty is the only playoff choker who was good enough in the regular season and coached long enough to get to 18 games. Explain to me why you would expect to find X number of chokers in a sample of 12 coaches with lots of playoff experience, and then I'll believe you when you say that you didn't find enough. (For the record, I would expect to find very few, because coaches who lose in the playoffs get fired and don't have the opportunity to play 18+ games.)

The second bad assumption is that every coach has a 50% chance of winning a playoff game and it is chance which determines the result. Why is it not reasonable to suppose that a good regular season coach could put his team in a bad position to win playoff games? Playoff games are not like regular season games because, for example, they are always against quality opponents. Why can't Marty's strategies be successful against poor and average competition but unsuccessful against good competition? I'm thinking in particular of his "get a lead early and sit on it" style which is much less effective against good teams because 1) good teams have a better shot at coming back into the game, and 2) you're less likely to get that early lead in the first place against a good team.

Postulating that things happen by chance, and then finding the right statistical result that would be predicted if things happen purely by chance, does not prove that things actually happen purely by chance. Remember, it is easier to be bad at something than good. They are not equally achievable.

#137: "If there are only 12 coaches in this sample, then how do we know how many of them should be “chokers�?"

As long as one of the coaches in the sample is a "choker", we should find at least two coaches with a record at least as bad as 5-13. You claim that Marty is a "choker", so there should be at least one other coach who has a record of 5-13 or worse due to bad luck. There isn't, so my conclusion is that there are no chokers in the sample, and that Marty is the one with a 5-13 record due to bad luck.

"Why can’t Marty’s strategies be successful against poor and average competition but unsuccessful against good competition?"

Aside from the fact that everyone's strategies are less successful against better opposition (after all, that's why it's called better opposition), his record against playoff teams in the regular season is roughly .500, so there's no reason to think that he isn't an effective coach against quality teams. Pretty much every coach has a worse record against playoff teams than against bad teams.

"The second bad assumption is that every coach has a 50% chance of winning a playoff game and it is chance which determines the result."

No, the assumption was that, over a long career, most coaches that get to the playoffs many times will sometimes have the better team, and sometimes have the worse team, and that these situations will be roughly equally likely. And that's fairly reasonable in most cases. If you look at the playoff careers of most of the coaches who have 18+ playoff games, you'll find that they had superior teams in some games, and inferior teams in others. But in general, on average, you would expect most of them to win half of their playoff games, if the outcome were determined solely on the skill of the two teams. It's not just determined by skill, however. There is an element of luck involved in football.

Given that we should expect a coach to win about half his playoff games over the long run, and that simple bad luck should push at least one in ten coaches to a record of 5-13 or worse, it's entirely reasonable to assume that an additional "choker" should add at least one other coach with a 5-13 record or worse. The fact that there is not another coach with such a record is evidence, although of course not conclusive evidence, that there aren't any "chokers" in the sample.

Re 56: "What makes you say this? Aside from you being a Steelers fan?"

Beating the #1, #2, and #3 seeds in the AFC (all on the road) and then the #1 seed in the NFC in the SB makes me say that. Correct me if I am wrong, but that is the one and only time it has ever happened.

Alex,

You have an incorrect understanding of how statistics generate predictions. The fact that 1 in 10 coaches would, by chance, have a 5-13 record over 18 playoff games does not automatically mean that you can expect to find such a record in a sample of 12 coaches with 18+ games. First of all, 12 coaches is too small a sample size for you to make any prediction other than that their records should very roughly approximate a normal curve. Predicting that a certain statistical outlier should exist in such a small group based on probability is going too far. If you had 50 coaches and predicted that about 5 would have such a record, that would be much more reasonable. Secondly, coaches who have the opportunity to coach 18+ games are neither a random nor representative sample. As I said, unsuccessful playoff coaches get fired. I would expect coaches with that many games to have records which skew above average (although again, only 12 is not a big enough sample to really see such effects). So I am not expecting to see lousy records for that reason as well. The lack of more than one coach with a 5-13 record is evidence of nothing. There could easily be one choker and zero statistically predicted outliers.

Also, you didn't adequately address my second criticism in post 137. Yes, I agree that over time the odds of any coach winning a playoff game average out to 50%. However, that does not mean that the outcome is random. You assume that playoff games are determined overall by chance. As Will Allen mentioned, that might be fair to say in games where the score is within 4 and it might not, but it is not true of games with a higher margin of victory. (You have to tread carefully here. If games won by 4 or less don't count, then Bill Belichik doesn't get credit for winning any Super Bowls.) In any case, Marty's record of 2-5 in games determined by greater than 4 points is clearly not due to fluke bounces of the ball. You said that games are determined by two things: The skill of the players and chance. But you forgot that games are also determined by the quality of the coaches.

I am not saying definitively that Marty is a bad playoff coach. I think he is, but the evidence is weak, I admit. However, Alex is wrong in that there is no evidence at all to suggest that he is not a bad playoff coach (besides his regular season record). The statistically likelihood of 1 coach in 12 having a 5-13 record purely by chance does not indicate whether it actually happened by chance or not. It only indicates that chance is one possible explanation.

#140: "I am not saying definitively that Marty is a bad playoff coach. I think he is, but the evidence is weak, I admit."

Then you're being a whole lot more reasonable than most people. I don't expect everyone to just completely forget all their beliefs about Marty in the playoffs, but it'd be nice if they wouldn't act so certain that he's awful. You seem willing to do that, so I don't have a problem with you.

"The statistically likelihood of 1 coach in 12 having a 5-13 record purely by chance does not indicate whether it actually happened by chance or not. It only indicates that chance is one possible explanation."

Ok, but one possible explanation is better than nothing, and most people aren't even willing to consider that he might be a good playoff coach with bad luck. The fact that you believe this is even a possible explanation makes you far more intelligent than most people who get involved in Marty-playoff arguments, and it's really all I was after here.

I'm pretty skeptical of all this math being thrown around here. Football games are not analagous to coin flips or at-bats, and I think it's legitimate to draw conclusions about performances like Marty's in playoff games.

Football is more like the social sciences than the hard sciences, in that tons of subjective, unquantifiable, and, to those of us watching these games on TV, unobservable, phenomena contribute to the final result. As in the social sciences, where there is a rich tradition of critique of the reliance on stats, I think football analysis would benefit more from an approach like that of, for example, Clifford Geertz, than it does of some of the--forgive me--snooty-sounding insistence on sample sizes this and that.

Regarding Marty, I think it's a case of quarterbacking, for the most part. The guy has rarely fielded a team in a playoff in which he had the superior QB. He's a bit like Bill Cowher in that regard (though Cowher tended to consistently win at least one game per year).

Marty is a very good coach, in that he repeatedly led teams quarterbacked by guys like DeBerg and Bono to the playoffs. But the style of play that lets that happen tends to be conservative--it's easier to coach the DeBergs and Bonos to not make mistakes than it is to coach them to routine 300-yard passing days. Get them to hit the occassional third down, avoid turnovers, and play defense.

That approach works well in the regular season, but it's one with a small margin for error when you're playing a good team. One or two screwups that give the opponent an easy score, or that squander your own scoring opportunities, is all it takes, and that has what has happened to many of Marty's teams. So, yeah, Marty's been a victim of bad luck, but the type of teams he's had have been more vulnerable to bad breaks.

Coaches in trouble:

Wade Phillips. Win with a-holes (one in particular) or be gone.

Marinelli. These guys should do a bunch better than 3-13.

Favre's overseer.

Childress. Obviously.

Gruden.

Fox.

The stiff in Oakland.

The guy with the big water bill. Hell he ought to quit now.

The stiff in San Dee if they dont make the playoffs.

Tomlin if he lays an egg.

Marvin if he has a few more players go to jail.

Jauron.

These guys likely quit if it doesnt go well. Or even if it does. Gibbs. Big Show. Dungy. Belichick. Billick.