Wednesday, May 16, 2012

The West Coast Offends

One of the things people get wrong with the San Francisco 49ers is the whole West Coast-offense thing.

The way the legend goes is that Bill Walsh came to town with a phenomenal new scheme called the West Coast offense and revolutionized a moribund attack and suddenly it was okay to say bad words on records. Or something like that.

What a bunch of rubbish.

The fact is that the San Francisco 49ers have always run a West Coast offense, in one or another of the ways that term has come to be defined. The West Coast offense out of the shotgun? Buck Shaw and Frankie Albert ran that in 1951. Five-yard dumpoffs to the running backs? That's how Hugh McElhenny got to Canton. McElhenny finished his career with 264 receptions. The only contemporary running backs who caught as many passes were Bobby Mitchell and Frank Gifford, and they spent most of their careers as actual flankers. Big possession receivers? San Fran had two of the best in six-foot-three R.C. Owens and six-foot-four Bernie Casey. Passing to the tight end? Dave Parks led the league in receptions in 1965, with John Brodie passing and Jack Christiansen in charge.

Brodie was a prototypical West Coast offense quarterback: short drop, quick release, short patterns, limited mobility. Brodie led the league in passing yards three times. He led the league in yards per completion once -- 1961, a complete aberration of a season.

From the time the 'Niners entered the league until 2002, the last good year of their run, the San Francisco offense was always more multidimensional than most, always full of backs who could catch, always willing to use the short pass as a surrogate run.

Like television or the iPhone, the West Coast offense was less revolutionary than evolutionary, a successful synthesis of elements that had already been tried. The only difference between the West Coast offense and the Run 'n' Shoot  is that the West Coast offense worked in the NFL.

The Run 'n' Shoot might have worked in the NFL; there was nothing inherently flawed about it. All it needed to succeed was the right coach and the proper personnel. The West Coast offense had Bill Walsh, and after a couple of years it had the people to make it work. But perhaps most importantly, it had a team used to such stuff.

Sunday, May 6, 2012

Of Course, You Know This Means WAR

Yeah, I’m back. Don’t get all excited or anything.

The hot stat in baseball is replacement value. It’s expressed in different ways, including WAR, or “wins above replacement,” but its basic premise is this: if you replaced player X on your team with a statistically average player, would he perform better or worse than the player he replaced – and by extension, would the team be better or worse?

Of course, this being a baseball statistic it has all kinds of tendrils – would the team be better offensively? Defensively? Would he hit in the clutch? Would he get on base? Would he throw out base-stealers? Would he give up the gopher ball in late-and-close situations? – but the idea of replacement value really boils down to how close a given player comes to the norm.
Replacement value is semi-self-fulfilling. Successful teams have more above-the-norm players, but it’s not clear whether those teams are successful because they have more above-the-norm players, or more of those teams’ players are above the norm because they’re on successful teams. Success begets performance and vice versa, and don’t forget: “Mother and Child Reunion” was first a chicken-and-egg dish served in a Chinese restaurant.
Football being less stat-driven than baseball (though not for lack of want-to), replacement value is not the hot topic it is on the diamond. Replacement value is less valuable in football because three-quarters of an NFL roster is on the field for one-third of a game or more. Many of those players are going to be at or below replacement value. It’s unavoidable. It’s the definition of the mean in a  world where all the children are above average.
(With all the situational offenses and defenses sometimes you’re better-served looking at the replacement value within a position on a given team – moving from the designated run-stuffer to the designated pass-rusher on defense or looking at the first-down back versus the third-down back. One of the reasons the Giants are so good is that Brandon Jacobs and Ahmad Bradshaw are both above replacement. Ditto for Osi Umenyiora and Jason Pierre-Paul.)

In football, replacement value is good for answering tough questions at the top of the pyramid. Who’s better – Peyton Manning or Tom Brady? How can you tell?
The baseball answer to that, delivered from the pulpit of replacement value, is to take Manning and Brady away from their teams, replace them with a player whose replacement value is at the norm, and see what happens.
That’s a loaded question. Take Albert Pujols out of the St. Louis lineup and there’s an Albert Pujols-shaped hole that can't be filled by Matt Holliday or Lance Berkman. The Cardinals still have to bat nine; they can slip in a Skip Schumaker and go into run-manufacturing mode, but it’s a partial transformation at best. They’re not going to go all bunt-y and start running Berkman like he was Michael Bourn.
Contrast that with football. The most crushing blow to an offense in 2011 outside of Peyton Manning occurred when the Vikings lost Adrian Peterson. How did the ‘Queens react? By not running the same offense at all. Joe Webb got the keys, the passing attack flipped around, the running attack was stood on its head, and while it wasn’t more successful it wasn’t markedly less successful. The point is that given resources and resourcefulness it can be done.
Indianapolis, of course, did none of those things, because it wasn’t built to do them. Joseph Addai and Donald Brown are running backs whose chief asset is not running the ball 30 times a game. They’re the Trevor Hoffman fastball, effective not because it’s used but because it might be used the next pitch … or maybe the pitch after that. The offense built around Peyton Manning could not be rebuilt around anyone else, certainly not Curtis Painter, Dan Orlovsky, or Kerry Collins, and the defense wasn’t good enough to deliver a win when the offense couldn’t outscore the opposition. So on the surface you’d guess that Peyton Manning had the greatest replacement value of any player in recent times.
Not so fast; we still have the Brady-Manning question to answer. Why did the Patriots not disintegrate when they lost Tom Brady in 2008, while the Colts went kablooey when they lost Manning in 2011? Defense is a big part of the answer, and having a defensive coach as head coach. The Indianapolis defense was shredded by injuries in 2011; New England in 2008 had a great defense, with Richard Seymour, Vince Wilfork, Mike Vrabel, Tedy Bruschi, Rodney Harrison, Jerrod Mayo, and Adalius Thomas all playing at a high level. They weren’t the 16-0 defense of the year before, but they were a top-10 defense, and the best defense the Pats have had since.
Even so, the offense was the league’s sixth-best unit, with the same crappy running game that New England has always had. Good receivers to be sure – Wes Welker and Randy Moss – but not really better than Wayne and Garcon and Dallas Clark in Indy. So you think maybe Matt Cassel was better than Curtis Painter?
Maybe a little, but maybe not that much. If Tom Brady’s replacement value is significantly lower than Peyton Manning’s, replacing Brady with Cassel doesn’t hurt nearly as much as replacing Manning with Painter-slash- Collins-slash- Orlovsky. So we come back to the original conclusion: Manning is better than Brady.

Yes, to the extent that his replacement value is higher. You may want Brady over Manning in a big game, and that’s your right. Replacement value doesn’t go there.

All of which is the long way ‘round to the big question: What player is/was the most indispensable to his team?

The simplest way to look at this is to examine the change in record from the year before the player left to the year he left, and then if he went to a new team the difference between the year before the player came and the year he came. A couple of simple change equations, acknowledging that the loss of one player does not define any team’s season.

Let’s work with the top 50 players of all time according to Pro Football Reference, and let’s look at changes that occurred in the meat of that player’s career, if that’s possible.

It gets dicey. Carl Eller and Johnny Unitas changed teams at the tail-end of their careers, when they were close to being replacement-level players. That’s not the same as Marshall Faulk changing teams after three years or T.O. being dealt to the Eagles in the prime of his career.
This little work is split in two, with a list of players who went from one team to a different team this time, and a list of players who didn’t go anywhere saved for the next episode, along with a big wrapup.

Now, the list of players who switched teams.
Lest you think the transition from Brett Favre to Aaron Rodgers was smooth and productive, guess again. The Packers got more than 50 percent worse; the Jets got 125 percent better. And the QB aside, the Favre-led team was nearly indistinguishable from the Rodgers-led team; it wasn’t like the defense collapsed or the offense was gutted by injuries. Favre at that stage of his career was that much better than Rodgers at that stage of his career. It was a big difference.

The difference between Warren Moon and Billy Joe Tolliver was likewise huge. Unlike the calculated move Green Bay made from the old pro (Favre) to his successor (Rodgers), Houston went from a productive star, to a 28-year-old journeyman who could serve as the poster boy for sub-replacement-level quarterbacks. If Curtis Painter were older, less intelligent, and longer off the tee he would be Billy Joe Tolliver. Given that, a precipitous decline was inevitable.
Interesting stuff abounds in these numbers. Love him or hate him, T.O. made the Eagles better and the ‘Niners worse – though he had the opposite effect when he went from Philly to Dallas three years later. Randy Moss, meanwhile, has had the opposite effect everywhere he's gone. Randy Moss is a player whose teams get worse when he arrives. That doesn't mean teams aren't good when Moss arrives, or that they don't prosper at some point during his tenure. They just don't get better when he arrives – and they get better when he leaves.
You can see how two safeties, Paul Krause and Emlen Tunnell, stabilized young defenses, and how three great defensive backs – Rod Woodson, Ronnie Lott, and Herb Adderley – were unable to do much for established defenses.
You can also see that L.T. is a good player. The Chargers may have underestimated how good – or perhaps they simply forgot that addition by subtraction is still subtraction. How could the Chargers without Tomlinson possibly be a better team than the Chargers with him?
And then there's Marshall Faulk. When he went from the Colts to the Rams he pulled off the amazing combination of making his new team nine games better, while the team he left got 10 games better. There has never been an NFL trade that made each team that much better.

This is good, but there's still the Peyton Manning situation to consider. And we will ... next time. 

And I promise there will be a next time.