Week 3 Discussion Thread

[Ness]

Originally Posted by Rosebud

I'm calling shenanigans on this one.

That's fine. Hell, I think a lot of what you say is pure garbage.

Believe what you like. As will I.

[Rosebud]

Originally Posted by Ness

That's fine. Hell, I think a lot of what you say is pure garbage.

Believe what you like. As will I.

So does Caddy, but he's terrible at being a human being to.

[njx9]

Originally Posted by abaddon41_80

What reality is being ignored? Just curious.

context. there is absolutely no context to raw stats, and in a sport where context overwhelmingly matters, the stats are, ultimately, worthless without it.

[cmarq83]

Originally Posted by Splat

http://profootballtalk.nbcsports.com...-in-san-diego/



Should be a great game to...

Don't try and change the subject! There is a good 3-4 pages left of fighting to be done about mediocre current and former 49ers players like Manny Lawson and Alex Smith. That takes precedence! We could have a good Donte Whitner, Anthony Davis, or Michael Crabtree conversation ahead of us!

[Ness]

Originally Posted by Rosebud

So does Caddy, but he's terrible at being a human being to.

Well hey it looks like you got an apprentice in the making :)

[Raiderz4Life]

Originally Posted by Rosebud

So does Caddy, but he's terrible at being a human being to.

I like you RB =)

And Aussieland will do that to anyone so don't hold it too much against Caddy.

[abaddon41_80]

Originally Posted by Rosebud

It's ludicrous argument so what do the stats matter? If you just wanted to present a useless factoid that'd be one thing, but you seem to think these prove anything other than how desperately you want Alex Smith not to be a middle of the road starting QB, which they don't

Why is it a ludicrous argument if the stats support it, though? People in this topic have already said that stats are supposed to be used as support for an argument. If Eli can make the jump from middle of the road starting QB after 87 starts then why can't Alex Smith?

context. there is absolutely no context to raw stats, and in a sport where context overwhelmingly matters, the stats are, ultimately, worthless without it.

I don't know what context you need to go along with the stats when you are given the team's record and how good the team around the QB is. Isn't the NFL only about your team winning games? If two quarterbacks have a similar record, similar stats and similar help from their defense and running game what else is there to compare?

[AntoinCD]

Originally Posted by njx9

context. there is absolutely no context to raw stats, and in a sport where context overwhelmingly matters, the stats are, ultimately, worthless without it.

This is a direct result of the fantasy craze where numbers=worth and value.

It is far too easy to look at a box score and decide for yourself how good that player is.

Using stats does have a place when discussing the NFL as at the end of the day it is a production driven business, however there is way more that needs to go into making a decision.

For instance, hearing that one QB threw for 300 yards, 3TDs and 1INT while the other threw for 250, 2TDs and 0INTs tells us nothing in terms which guy played better.

We need to know so many variables

Did one guy play in a dome while the other played in the Meadowlands in November?

Did one guy throw to great WRs when the other threw to scrubs?

What situations did each guy face? Was one consistently facing 3rd and long while the other QB's team won on first and second down all day?

Football is too complex to directly compare players to others. Everyone does it in some way and at times it is a fun exercise, however based on different schemes, different options, different situations etc there are virtually no fair ways to compare players

[nrk]

How many times has Eli lost his status as a starter? 0

[Ness]

Originally Posted by nrk

How many times has Eli lost his status as a starter? 0

Speaking of out of context...

[njx9]

Originally Posted by abaddon41_80

I don't know what context you need to go along with the stats when you are given the team's record and how good the team around the QB is. Isn't the NFL only about your team winning games? If two quarterbacks have a similar record, similar stats and similar help from their defense and running game what else is there to compare?

seriously? because throwing 47 1-yard passes and watching your running back dodge the entire defense for 90+ yard scores all day will make it look like you're an incredible qb. until you actually watch the game and realize the qb couldn't throw downfield if his life depended on it. football does not break down to simple numbers, and it never will. this isn't track. we can't just look at 100/200 times and figure out who's a pretty good sprinter.

Originally Posted by AntoinCD

This is a direct result of the fantasy craze where numbers=worth and value.

It is far too easy to look at a box score and decide for yourself how good that player is.

Using stats does have a place when discussing the NFL as at the end of the day it is a production driven business, however there is way more that needs to go into making a decision.

For instance, hearing that one QB threw for 300 yards, 3TDs and 1INT while the other threw for 250, 2TDs and 0INTs tells us nothing in terms which guy played better.

We need to know so many variables

Did one guy play in a dome while the other played in the Meadowlands in November?

Did one guy throw to great WRs when the other threw to scrubs?

What situations did each guy face? Was one consistently facing 3rd and long while the other QB's team won on first and second down all day?

Football is too complex to directly compare players to others. Everyone does it in some way and at times it is a fun exercise, however based on different schemes, different options, different situations etc there are virtually no fair ways to compare players

thank you for saying that better than i was about to. and absolutely, stats can have a place. but that place is solely in support of direct visual evidence, NEVER as the entirety of your argument.

[Ness]

Too bad tonight's game can't start now.

[AntoinCD]

I think tonight's game will all come down to the fact that Carolina will not be able to disrupt Manning enough up front while the Giants d-line should have fun against the Panthers' front.

I really like Carolina's LBs and am looking forward to see Keuchly for the first time, but when it comes down to it Manning should be able to sit back and work the Giants down the field enough.

Stopping the Panthers' rushing attack will be a challenge but overall I think the Giants win by 7-10 points

[abaddon41_80]

Originally Posted by AntoinCD

This is a direct result of the fantasy craze where numbers=worth and value.

It is far too easy to look at a box score and decide for yourself how good that player is.

Using stats does have a place when discussing the NFL as at the end of the day it is a production driven business, however there is way more that needs to go into making a decision.

For instance, hearing that one QB threw for 300 yards, 3TDs and 1INT while the other threw for 250, 2TDs and 0INTs tells us nothing in terms which guy played better.

We need to know so many variables

Did one guy play in a dome while the other played in the Meadowlands in November?

Did one guy throw to great WRs when the other threw to scrubs?

What situations did each guy face? Was one consistently facing 3rd and long while the other QB's team won on first and second down all day?

Football is too complex to directly compare players to others. Everyone does it in some way and at times it is a fun exercise, however based on different schemes, different options, different situations etc there are virtually no fair ways to compare players

seriously? because throwing 47 1-yard passes and watching your running back dodge the entire defense for 90+ yard scores all day will make it look like you're an incredible qb. until you actually watch the game and realize the qb couldn't throw downfield if his life depended on it. football does not break down to simple numbers, and it never will. this isn't track. we can't just look at 100/200 times and figure out who's a pretty good sprinter.

I see your point, to an extent. The problem is that all of these things; receivers making the plays for the QB, playing in a dome versus in the wind, etc; tend to level out for every QB over the course of a season. We are talking about five seasons. If memory of Stats I serves, you only need around 30 observations from a sample for it to become approximately normal. We are talking about 2500+ pass attempts.

How many times has Eli lost his status as a starter? 0

The only time Alex Smith kind of lost his status as a starter was when he got injured in 2010. He got healthy but Singletary, in his vast wisdom, decided to keep Troy Smith. I wouldn't call it Smith being benched and even if you would, does Mike Singletary benching you mean anything? If anything I would consider Singletary benching a player a compliment to that player.

[Matthew Jones]

I'm pretty confident in the Giants winning tonight as well. The Panthers may be an up-and-coming team, but it's hard to pick a team with a top-ten pick last year over the champs.

[njx9]

Originally Posted by abaddon41_80

I see your point, to an extent. The problem is that all of these things; receivers making the plays for the QB, playing in a dome versus in the wind, etc; tend to level out for every QB over the course of a season. We are talking about five seasons. If memory of Stats I serves, you only need around 30 observations from a sample for it to become approximately normal. We are talking about 2500+ pass attempts.

that couldn't be less true in the nfl. how many OCs did alex smith have in that time? how many different schemes? how many different offensive philosophies? how often was he playing with a game out of reach, or trying to preserve a close victory? those things absolutely do not even out across multiple teams over multiple years. statistical analysis of something like the nfl takes a massively more complex regression analysis than you're trying to present. take a look at the DVOA stuff, and even then hardly anyone agrees on its practical value.

[Ness]

Even though the Panthers have had trouble the last couple of weeks putting a cap on rushing yards against their defense, I still like the Panthers tonight. I think Newton has a good game.

[K Train]

i think newton rolls over them

[abaddon41_80]

Originally Posted by njx9

that couldn't be less true in the nfl. how many OCs did alex smith have in that time? how many different schemes? how many different offensive philosophies? how often was he playing with a game out of reach, or trying to preserve a close victory? those things absolutely do not even out across multiple teams over multiple years. statistical analysis of something like the nfl takes a massively more complex regression analysis than you're trying to present. take a look at the DVOA stuff, and even then hardly anyone agrees on its practical value.

Different coordinators and different systems and different offensive philosophies can be seen in the stats. If one QB has 300 yards on 10/30 passing he is clearly playing in a system that throws a lot of deep passes. If a QB has 300 yards on 35/50 passing he is likely playing in an offense that emphasizes shorts completions. If one QB throws for 300 yards with no touchdowns but his team still scores four offensive touchdowns and has 400 total yards then he probably doesn't throw much near the goalline. If another QB throws for 200 yards with three touchdowns while his scores four offensive touchdowns and has 400 total yards then he probably throws a lot more at the goalline.

[nrk]

Originally Posted by abaddon41_80

The only time Alex Smith kind of lost his status as a starter was when he got injured in 2010. He got healthy but Singletary, in his vast wisdom, decided to keep Troy Smith. I wouldn't call it Smith being benched and even if you would, does Mike Singletary benching you mean anything? If anything I would consider Singletary benching a player a compliment to that player.

Yup, see how stats don't tell the whole story?

[abaddon41_80]

Originally Posted by nrk

Yup, see how stats don't tell the whole story?

I don't see what that has to do with stats ...

[AntoinCD]

Originally Posted by abaddon41_80

Different coordinators and different systems and different offensive philosophies can be seen in the stats. If one QB has 300 yards on 10/30 passing he is clearly playing in a system that throws a lot of deep passes. If a QB has 300 yards on 35/50 passing he is likely playing in an offense that emphasizes shorts completions. If one QB throws for 300 yards with no touchdowns but his team still scores four offensive touchdowns and has 400 total yards then he probably doesn't throw much near the goalline. If another QB throws for 200 yards with three touchdowns while his scores four offensive touchdowns and has 400 total yards then he probably throws a lot more at the goalline.

10/30 for 300 yards could also indicate a guy who is left with multiple 3rd and long situations and/or is playing from a big hole and has to throw. It could also indicate teams playing prevent late in the game where they know the QB needs big plays and are willing to give up passes underneath and give the guy the middle of the field but not get out of bounds. There are just too many possibilities for any specific statistic

[abaddon41_80]

Originally Posted by AntoinCD

10/30 for 300 yards could also indicate a guy who is left with multiple 3rd and long situations and/or is playing from a big hole and has to throw. It could also indicate teams playing prevent late in the game where they know the QB needs big plays and are willing to give up passes underneath and give the guy the middle of the field but not get out of bounds. There are just too many possibilities for any specific statistic

My point is that over the course of five seasons these situations are going to level out for quarterbacks.

[Raiderz4Life]

Sweet baby Jesus plz let the game start so we can stop talking bout alex ******* smith.

[Saints-Tigers]

Alex Smith will never be the reason that a team is really good, but at least at this point he's good enough to not drag a dominant roster down far.