Good article IMHO.
Good article IMHO.
Barnwell has disrespected us at just about every turn since he started writing for Grantland, but I have to commend him for owning up to it and eating crow...every single time.
1) teams can change in season. replacing Cam wi Caldwell was a huge upgrade, tho it took a couple games to get going.
2) I've seen plenty of games where the 'better' team lost cuz of one tipped pass, fumble, bad call. part of a Super Bowl run is to get a few breaks.
good observation on holding in the end zone = automatic safety. gonna give one anyway so...
And you know, if Cam was still the OC, and our Oline didn't change, and all our hurt players didn't get healthy in time, then we would have stayed inconsistent, and faltered, and maybe not even win the division and been lucky to even get in the playoffs.
Even though I look at stats too, because if the same players are on the field every game for a specific team, then looking at their stats can help to quantify them.
But, if you look at a team like this years' Ravens, then alot of those stats can be thrown out.
So many players hurt along the way.
Backups to backups playing sometimes in crucial moments of games this season.
Young players thrown into early action...and they developed and learned and became battle harded, so to speak.
Then players coming back at different times...
Then OC being tossed out at the right time...
And then becoming healthy enough and a game plan good enough to win the division against the Giants.
The stats did not reveal how that game played out beforehand.
And everryone coming together, getting healthy (resting during last game of season), Caldwell being comfortable, emotional lift given by Ray, all the rookies and 2nd year players AND this years' acquired free agents gelling AND Oline shift...the stats do not account for all of these changes and what it means.
And that is why the "experts", looking at stats and earlier season performances came to wrong conclusions.
The Ravens team that took to the fields during the post season didn't resemble the team that started the season or at anytime during it (well maybe the first game - vs. Bengals).
And as I listened to the "experts" tell their conclusions, I knew they were wrong as soon as it came out of their mouths.
They didn't do their homework.
And they ARE bitter that the Ravens made them look stupid.
And that's why I see all the negative press.
Occasionally teams outperform what the advanced stats suggest. There's no shame in that. Statistical analysis is all about the likelihood of outcomes, and measuring what you can measure, and ignoring the rest.
And it's not like advanced stats have some bias against the Ravens. A few years ago when the Ravens snuck into the playoffs at 9-7, Football Outsiders had us as the number one team in football.
I respect advanced stats because they challenge our beliefs and show us how flawed your eyes can be. they also back up common sense ideas (going for it more on fourth down, running to win as BS, running on 3rd/4th and short more, etc).
I loved the section about Joe. Even if you think the Ravens got lucky occasionally (and few SB winners didn't) there is no denying that what Joe did was historic.
Agree its much harder to quantify football, though he says a few things that a coach might want to look into in terms of the conventional wisdom on going for 2, etc.
But the biggest things that are not accounted for are things like the ones mentioned above -- (1) the Oline shuffle at the start of the playoffs -- it was like getting a brand new LT (and a very good one), added to the roster all of a sudden, not to mention upgrading at LG and RT in the process., then (2) the Caldwell Effect -- and you could break that into an A and B -- (2A) -- the seemingly better play calling and play sequencing, liberating the offense (and thus making comparison to the offense we had most of the season not longer applicable), and (2B) the basic fact that there is no playcalling history with Caldwell for anybody to get a read on. All those "stats" (like Jimmy Smith noting that the 49ers throw to Crabtree 52% of the time inside the red zone) go out the window, because just having a new OC - ANY new OC -- means the tendencies and scouting of tendencies is out the window. Then there's (3) the Jacoby Jones Wild Card Factor. How do you account for a guy who catches lightning in a bottle like he does? He catches one ball, but the one he catches he's totally behind the defense AGAIN. Throw in a perfectly blocked kickoff return, and boom -- there's 14 points from two plays really out of nothing, out of nowhere. No way to account for those kinds of plays -- you can't say "on average Jones will add 3.8 points per game" -- you either get the big play -- and all 7 or 14 points -- or you don't and get zero (which you have to go and find elsewhere).
Those are just three things, but you also have to look at something that goes TOTALLY overlooked by media these days, and it's something that USED TO be, back in the day, a routine part of the conventional wisdom. That is, the Ravens are a more experienced and battle tested team, and therefore a team you would expect to hold its nerve in a tight situation in a high profile game. It is no surprise to me that the Ravens executed in key moments and the 49ers didn't. That's experience, big game experience, and the Ravens have a boatload of it -- probably as much or more than anyone in the league.
Was a "stat" shown of how many playoff games the collective rosters of each team had played in? And won?
In college, experience is always a big factor in evaluating odds -- you can use "total career starts on the offensive line" as a strong predictor of games there. But in the NFL, pundits have almost totally written "playoff experience" out of the storyline in favor of whatever trendy flavor of the month excites the viewers.
Yet anybody who has ever competed in any sport at anything resembling a high level knows that no matter how talented the younger, upstart team may be, the wily, grizzled vet team who's been there a bunch, and who once were the young upstarts themselves, will always have the whammy over the younger guys, and it is something teams work very hard at trying to get their young players to keep out of their heads. But it is hard, because you know the other guy knows how to find a way in a big game, while you, the young guy, deep down wonder whether you can -- and there is no substitute in this world for "knowing you can, because you've already done it." Having no doubt in your abilities is vastly superior to believing in yourself, or imagining you can do it. All the pop psychology and phony bravado in the world can't get you over that hump - you have to have been there and done that, to have that mental edge. And that is something none of this statistical analysis accounts for.
At a rough guide though, simply totaling up each side's collective playoff experience would be useful in getting a better perspective on forecasting the game (which is what the stat heads want to do). The fact that you don't see it incorporated into any of these articles suggests that these guys are doing a lot of analysis on the wrong things, which is why they can never get it right when it comes to a team like the Ravens.
Good article, but I have one beef with Barnwell:Typical total misunderstanding of probabilities. He's essentially arguing the "Law of Averages" while forgetting that the "law" only applies after many, many, many trials. If going 6 for 8 on 3rd down in the first half had any bearing at all how they would fare the next time they faced the situation, you would expect their success to continue--since the only real data you'd have at that point suggests that the probability of converting on the next 3rd down is close to 75%.Quote:
No team in the league can pick up 75 percent of their third-down conversions over any appreciable span of the time, and the Ravens were no exception.
What I find curious is that right up top he admits that 16 games is not a large enough sample to make accurate judgments about at team--then he argues for a "reversion to mediocrity" after just 8 trials. Bizarre.