Metrics are for.... | Page 4 | FinHeaven - Miami Dolphins Forums

Metrics are for....

If the objective data tells us the same thing we're seeing on TV, that's sure great, but when we're highly emotionally invested in seeing one thing and not another, which happens a great deal more when watching players for our own team (as opposed to Kaepernick, for example), then in my opinion we should sure allow the objective data to disconfirm our subjective perceptions with regularity and without issue.

I love metrics and the correct use of stats, I just hate when people focus on one stat like it's the end all be all of stats, for example YPA, that's a team number. Not a individual variable.
 
Really? The guy that just broke the franchise record for road playoff wins, on a team that had hall of famers Steve Young and Joe Montana no less. The guy that could very easily go to back to back Super Bowls in his first two seasons as a starter. The guy thats 23-7 as a starter and 4-1 in the playoffs. Really? You're really going to take a stand against the YPA stat using that guy, Colin ****ing Kaepernick, as proof the stat is meaningless?

Young only played in one road playoff game in his entire career. As for montana, If you count super bowl wins as a road game then no crapernick hasn't won more road playoff games.
 
I love metrics and the correct use of stats, I just hate when people focus on one stat like it's the end all be all of stats, for example YPA, that's a team number. Not a individual variable.
I'm happy to agree to disagree with anyone who looks at the following table and fails to perceive a correlation between net YPA and the consensus perceptions of individual quarterbacks' ability:

http://www.pro-football-reference.c...pos_is_db=Y&draft_pos_is_k=Y&draft_pos_is_p=Y
 
Let me give you a example of what I'm talking about.

When I was the starting S for FDHS here in Summerville SC, we were going up against Wando High, who was a primarily quick pass, heavy run Om they used screens every drive to get there fast playmakers in space. Yes it's high school but there was nfl talent out there. They were the heavy favorite due to there explosive offense. The Qb had a 10.5 YPA, Awsi just blew a load at that number, anyways like I said earlier they were a heavy run with quick screen passing game, yes SCREEN, throughout the first q they were hitting there screen passes with ease until about halftime. Once we got adjusted to it the whole second half they barely moved the ball down the field due to stacking the box on that peewee football gameplan. His finished the game with a 5.4 YPA due to his playmakers getting tackled instantly.

What I'm getting to is, a qb that threw nothing but screens had a 10 YPA, guys who catch the ball are just as important as the guy throwing it.

Don't look at one number and think it's the qbs problem.
 
I don't think anyone is arguing that it's worthless number that has no meaning. Colin Kaepernick is just a great example of a player who has a really high YPA yet isn't a particularly good quarterback.

Take Young and Montana for instance. Kaepernick's got a higher career YPA than Joe Montana, I believe. You'd really take Kaepernick over Montana? :lol: Likewise, Steve Young had an adjusted yards per attempt of just 5.0 in Tampa. That's awful. Was Steve Young a terrible quarterback? How then does he pull off an incredible 7.9 adjusted yards per attempt in his 13 years in San Francisco? Do you think that coaching and supporting cast matter? Because I do. I think they matter a lot.

All I can say is that if you truly believe that one statistic means everything, and tells you everything you need to know about an individual player, then you must also entertain the possibility that David Carr is a significantly better quarterback than Andrew Luck. But I know you don't.

Personally, I prefer passer rating (PR) to YPA, I think PR is a much stronger indicator of overall QB skill but I certainly wouldn't try to use Kaepernick as an example to discredit the YPA stat. Kaepernick doesn't score particularly highly in PR despite having an extremely high YPA and YPA is a big component of the PR stat, which suggests he's well above average at certain aspects of the stat but well below average at other aspects of the PR stat.

Just based on PR stats Kaepernick still gets an inconclusive grade from me but he is above average so far. Both Young and Montana had multiple seasons with a PR of 100+ which is the ultimate indicator of QB play for me. Here's an interesting nugget: Dan Marino only had one season with a PR of 100+ and Russel Wilson already has two. Another interesting stat is every QB with the exception of Phillip Rivers that has at least two seasons with a PR of 100+ has won a Super Bowl.
 
Personally, I prefer passer rating (PR) to YPA, I think PR is a much stronger indicator of overall QB skill but I certainly wouldn't try to use Kaepernick as an example to discredit the YPA stat. Kaepernick doesn't score particularly highly in PR despite having an extremely high YPA and YPA is a big component of the PR stat, which suggests he's well above average at certain aspects of the stat but well below average at other aspects of the PR stat.

Just based on PR stats Kaepernick still gets an inconclusive grade from me but he is above average so far. Both Young and Montana had multiple seasons with a PR of 100+ which is the ultimate indicator of QB play for me. Here's an interesting nugget: Dan Marino only had one season with a PR of 100+ and Russel Wilson already has two. Another interesting stat is every QB with the exception of Phillip Rivers that has two seasons with a PR of 100+ has won a Super Bowl.

IMO it's hard to compare players from past to present.

Also Wilson has a good PR, but he also completed and attempted a small amount of passes.

I've always wondered what RT YPA would be with the average amount of sacks..
 
I've always wondered what RT YPA would be with the average amount of sacks..
Well wonder no more, because the correlation between sacks and YPA between 1994 and 2013 is exactly zero, and game-to-game in 2013, the correlation between the number of sacks Tannehill took and his YPA was -0.05.

In other words, you're talking about two variables that are completely unrelated for quarterbacks in general and for Tannehill in 2013 in particular.

This is precisely why we should use objective data to form conclusions, as opposed to watching Tannehill on TV alone and believing he would perform much better (with regard to YPA at least) if he were sacked less. While it may make us feel better to believe he would play better if he were sacked less, that belief isn't supported by anything objective.
 
Roy, I got ya. The reason we bring up YPA is because others continue to place extreme emphasis on it. In Awsi's case, I understand why he does it, as efficiency statistics are very useful for laying wagers. In shouright's case, he simply parrots things that he reads and then regurgitates them, often with very humorous results. In fact, you could say that he sometimes "ckparrots" things which he doesn't understand.

My argument is simply this: football is about production. Nothing else matters. And as someone who views football as a game of production, I think that statistical analysis should supplement scouting and film evaluation in any well-run front office.

However, efficiency statistics do not tell you if a higher level of production can be achieved by a player; efficiency statistics tell you whether or not a high level of production is being achieved by a player. That's why I bring up players like Steve Young, Rich Gannon, Jake Plummer, etc. These are all quarterbacks who had the physical and mental gifts necessary to produce at a high level (to varying degrees), yet early in their careers they were unable to do so.

Young was never going to be an All-Pro in Tampa. Rich Gannon was never going to be great running a Marty Ball offense. Jake Plummer was never going to go 13-3 and be a Pro-Bowler in Arizona.

When I see Ryan Tannehill on film, I see a quarterback with great physical tools who can throw the ball accurately and with anticipation. I see a quarterback who can read a defense better than most. I see a quarterback who can run an offense. I see a quarterback who had four fourth quarter comebacks this season, which is pretty good. And all of this brings me back to something Greg Cosell said a couple of months ago: Ryan Tannehill is executing the offense. The problem (IMO) is this: look at the offense he is being asked to execute.

I don't know if Bill Lazor will be a good hire or a bad one, but I do know that Mike Sherman was not. And Sherman's body of work at every stop in his career (save for one year in Houston... under Gary Kubiak) bears that out, IMO.
 
kaepernicks a heavy product of everyone else around him...that ypa thing is severely overrated with him...he can make some good throws the back shoulder down the seam to davis for the td a few weeks ago i think it was was straight money but you take away that run game and defense and he's a qb that most teams would look to upgrade on...he's a one read primarily qb who takes off and leaves clean pockets all the time cause he doesnt go thru progressions well...what helps him last though is he's not stupid he doesnt run to contact and he's got a long stride in full gallop so the speed is deceptive

he's damn lucky on that keep td vs carolina that wasn't a safety coming full bore across the formation with a clean look in the gap at the los who overran it cause had it been a head hunter he would have been destroyed...

but yeah kaeps heavily a product of what they have around him...from the defense to the running game to the oline which is as physical as you will find in football...

and yeah futurescout anything that has cutler worth more wins than tom brady needs to be burned in a burn barrel..i understand metrics and stats have their place sure but these numbers in a vacuum stuff that this board now gets pounded with aren't worth a damn...they just seem to get worse
 
Statistics, keep statisticians in business. Nothing more.
 
And when someone is no one but a fan on a message board (myself included), he ought to recognize his potential for bias with regard to what he believes he's seeing on his television screen, and use objective data to confirm or disconfirm the conclusions he makes via that medium.

I agree. I think it is important to use advanced analytics and film. They both are powerful tools.

Sent from my SPH-L710 using Tapatalk
 
The thing is none of it makes sense. There are to many variables that can't be determined by some statistics professor, who doesn't know the first thing about football. I don't care what anyone says numbers are deceiving and don't tell the whole picture. Film is the closest thing you can get. "Analytics" and "Metrics" are just the flavor of the month, everyone is caught up on the Billy Beane book/movie. This isn't baseball. Football is to damn complex for some formula or statistic. I wonder how much analytics Bill Belichick or Ozzie Newsome use?
I have a bit of a connection (acquaintance is an ex-pro QB who knows Belichick well!) and I can tell you that Bellichick DOES use them. He certainly doesn't use them the way the way Shou/Gravity does but they do have their place. Coaches look for matchup advantages and statistical tendencies and metrics have a lot to do with creating them.
 
Roy, I got ya. The reason we bring up YPA is because others continue to place extreme emphasis on it. In Awsi's case, I understand why he does it, as efficiency statistics are very useful for laying wagers. In shouright's case, he simply parrots things that he reads and then regurgitates them, often with very humorous results. In fact, you could say that he sometimes "ckparrots" things which he doesn't understand.

My argument is simply this: football is about production. Nothing else matters. And as someone who views football as a game of production, I think that statistical analysis should supplement scouting and film evaluation in any well-run front office.

However, efficiency statistics do not tell you if a higher level of production can be achieved by a player; efficiency statistics tell you whether or not a high level of production is being achieved by a player. That's why I bring up players like Steve Young, Rich Gannon, Jake Plummer, etc. These are all quarterbacks who had the physical and mental gifts necessary to produce at a high level (to varying degrees), yet early in their careers they were unable to do so.

Young was never going to be an All-Pro in Tampa. Rich Gannon was never going to be great running a Marty Ball offense. Jake Plummer was never going to go 13-3 and be a Pro-Bowler in Arizona.

When I see Ryan Tannehill on film, I see a quarterback with great physical tools who can throw the ball accurately and with anticipation. I see a quarterback who can read a defense better than most. I see a quarterback who can run an offense. I see a quarterback who had four fourth quarter comebacks this season, which is pretty good. And all of this brings me back to something Greg Cosell said a couple of months ago: Ryan Tannehill is executing the offense. The problem (IMO) is this: look at the offense he is being asked to execute.

I don't know if Bill Lazor will be a good hire or a bad one, but I do know that Mike Sherman was not. And Sherman's body of work at every stop in his career (save for one year in Houston... under Gary Kubiak) bears that out, IMO.



Good points. Not many qb;s had the leisure to sit and learn behind Montanna for like 4 years and your right Young would have been Archie Manning or worse in Tampa.
I personally think Tannehill will be fine but our line is pure crap and not protecting the qb ususally results in bad results. I don't need a metric to show me that. I know that usually when a qb gets sacked 7 times turnovers happen. That's simple football.
 
Good points. Not many qb;s had the leisure to sit and learn behind Montanna for like 4 years and your right Young would have been Archie Manning or worse in Tampa.
I personally think Tannehill will be fine but our line is pure crap and not protecting the qb ususally results in bad results. I don't need a metric to show me that. I know that usually when a qb gets sacked 7 times turnovers happen. That's simple football.

Coaching and scheme matters a lot. Go back and look at the year by year splits on efficiency statistics for QBs like Rich Gannon and Matt Hasselbeck, for instance. Look at the complete metamorphosis that Rich Gannon underwent when Jon Gruden got his hands on him. Look at what happened to Matt Hasselbeck in Seattle after Holmgren left the building. Once Greg Knapp got his hands on Hasselbeck, his efficiency numbers fell off the face of the earth.

Nick Foles has done well in Philadelphia and deserves credit for his play, but do you really think he would be enjoying such incredible efficiency if we swapped him into the Dolphins offense under Mike Sherman?

I can tell you he wouldn't have a 9.0 yards per attempt statistic or a top three all time passer rating in an offense in which he's being asked to throw quick flare-outs to Dion Sims and tight comebacks/outs to Brian Hartline and Mike Wallace with a pocket that's usually crumbling in his face. Oh, and if the first read ain't wide open and they're covering McCoy, just take off if you got some room. And folks, that isn't a criticism of Foles or some attempt to discredit him. That's a credit to GREAT coaching and a QB who understands what's asked of him -- and can do it well.

Incidentally, this is why I am very skeptical of the Lazor hire. From what I'm hearing, Lazor is going to keep a lot of the stuff from Mike Sherman's system in place, or at least it sounds that way to me. That's not good.

We should have fired Joe Philbin if that's what it would take to bring in someone who actually understands how to get production out of his players.
 
Back
Top Bottom