jday
Well-Known Member
- Messages
- 9,321
- Reaction score
- 13,284
Earlier today I submitted a video that featured a debate between legendary coach Herm Edwards and the notorious Cowboys homer Will Cain (if you missed it: http://cowboyszone.com/threads/video-is-zeke-the-best-running-back-in-the-nfl.380350/).
My first impression was this is a fine representation of Cowboys hate at its worst on Herm Edwards part. But if you take a closer look at what he said, he had two key points that were very telling about his opinion: 1. Zeke did not have a whole lot of targets/catches in 2016. 2. Zeke’s quarterback (Dak Prescott) was mobile whereas David Johnson’s (the running back that Herm Edwards argued is the #1 all-around back in the league) quarterback (Carson Palmer) is a statue in the pocket and therefore he is also clearly better at picking up the blitz….because his situation dictated that ability is a necessity.
He admitted freely that as a runner, Zeke may be the best; but as an all-around back, Herm put him at fourth, behind David Johnson (#1), Le’Veon Bell (#2) and…wait for it…LeSean “Shady” McCoy (#3). Herm clearly did not watch a whole lot of Cowboys football last year.
Herm’s problem, though, is not a lack of objectivity. His problem highlights the issue I have with stats garnering more credence than they deserve. Essentially, Herm is pointing to factors beyond Zeke’s control, which is (1) how the coaches opted to use him week in and week out and (2) the players the front office has put around him. Because the Cowboys had a Lance Dunbar to use in the passing game, they opted to save tread on Zeke’s tires and use him predominantly in the ground attack. Furthermore, had Romo not been injured, Zeke’s abilities as a pass protector would have had a much better opportunity to be showcased…after all, pass protection may very well be the one truly elite aspect of his game.
Had the Cowboys not rested Zeke on third downs and for entire series in the middle of games and had the Cowboys played him in that last meaningless game against the Eagles and had the Cowboys dispensed with the Dunbar gimmick and just let Zeke carry the load throughout both in the ground game and the pass game last season, this would not even be a debate. Zeke’s value as it compares to other players on other teams is now in question as a result.
But consider that point for a moment; because in it we find enlightenment. If the largest factor behind how stats are accrued is solely dependent on how an offensive or defensive coordinator chooses to use his player, how telling can stats possibly be when comparing one player from one team to another player from a different team? Stats assume that a coach will always use his best player in every situation and in the real world that simply isn’t a reality. Stats assume players face the same conditions, the same teams, the same strength of schedule, the same advantages of surrounding talent, the same litany of factors that decide how a player will perform….and the ugly truth is, that simply is nowhere even close to being true.
Example: Let’s draw from the age old debate of whose better: Emmitt Smith or Barry Sanders. It has long been my contention, as a pure runner, Barry Sanders was better. As an all-around running back, however, Emmitt Smith is the best the game has ever seen (though, Zeke may have something to say about that before it is all said and done…I digress). Let’s say that we have it within our power to clone Emmitt and Barry in their prime for a day and we are going to put that debate to bed once and for all.
We place both players on two different teams in two different states and setup matches for them against (once again) two other different teams, with, by the way, different levels of talent on both sides of the ball. On balance, what we are attempting to recreate is the exact conditions current players are subjected to in determining their overall value as it compares to other players of like position. By now, you should already be able to poke holes all through the logic I am using to determine Emmitt and Barry’s overall value.
You might ask: What if one team they are up against has a great defense and the other team’s defense is crappy. Whoever faces the crappy defense is going to look better! That’s not fair!
But how is that any different from how we determine player’s value today?
You might say: Well, over the course of a season, it all evens out!
Does it? Do we all face the same teams? And do we all face the same teams in the same weather conditions? And do the same set of circumstances play out in every game making coaches largely grant all of their players the same snap count across the board?
You might say: Well, what analyst do is normalize the stats to make them comparable often times dividing their overall yardage (for instance) by their touch count. What you really should be looking at is their average yards per carry, attempt, catch, etc.
That may make things a bit fairer, but don’t player tend to tire the more snaps they get? Could that not negatively impact their overall average over time?
Please don’t misunderstand. I’m not attempting to sale you on the idea that’s all stats are garbage and essentially tell us nothing…I’m merely pointing out that most stats really prove a whole lot of nothing when attempting to compare one player to another player. Unfortunately, football is not played in a controlled environment and as such cannot be used to substantiate any claims that one player is better than the other.
Thoughts?
My first impression was this is a fine representation of Cowboys hate at its worst on Herm Edwards part. But if you take a closer look at what he said, he had two key points that were very telling about his opinion: 1. Zeke did not have a whole lot of targets/catches in 2016. 2. Zeke’s quarterback (Dak Prescott) was mobile whereas David Johnson’s (the running back that Herm Edwards argued is the #1 all-around back in the league) quarterback (Carson Palmer) is a statue in the pocket and therefore he is also clearly better at picking up the blitz….because his situation dictated that ability is a necessity.
He admitted freely that as a runner, Zeke may be the best; but as an all-around back, Herm put him at fourth, behind David Johnson (#1), Le’Veon Bell (#2) and…wait for it…LeSean “Shady” McCoy (#3). Herm clearly did not watch a whole lot of Cowboys football last year.
Herm’s problem, though, is not a lack of objectivity. His problem highlights the issue I have with stats garnering more credence than they deserve. Essentially, Herm is pointing to factors beyond Zeke’s control, which is (1) how the coaches opted to use him week in and week out and (2) the players the front office has put around him. Because the Cowboys had a Lance Dunbar to use in the passing game, they opted to save tread on Zeke’s tires and use him predominantly in the ground attack. Furthermore, had Romo not been injured, Zeke’s abilities as a pass protector would have had a much better opportunity to be showcased…after all, pass protection may very well be the one truly elite aspect of his game.
Had the Cowboys not rested Zeke on third downs and for entire series in the middle of games and had the Cowboys played him in that last meaningless game against the Eagles and had the Cowboys dispensed with the Dunbar gimmick and just let Zeke carry the load throughout both in the ground game and the pass game last season, this would not even be a debate. Zeke’s value as it compares to other players on other teams is now in question as a result.
But consider that point for a moment; because in it we find enlightenment. If the largest factor behind how stats are accrued is solely dependent on how an offensive or defensive coordinator chooses to use his player, how telling can stats possibly be when comparing one player from one team to another player from a different team? Stats assume that a coach will always use his best player in every situation and in the real world that simply isn’t a reality. Stats assume players face the same conditions, the same teams, the same strength of schedule, the same advantages of surrounding talent, the same litany of factors that decide how a player will perform….and the ugly truth is, that simply is nowhere even close to being true.
Example: Let’s draw from the age old debate of whose better: Emmitt Smith or Barry Sanders. It has long been my contention, as a pure runner, Barry Sanders was better. As an all-around running back, however, Emmitt Smith is the best the game has ever seen (though, Zeke may have something to say about that before it is all said and done…I digress). Let’s say that we have it within our power to clone Emmitt and Barry in their prime for a day and we are going to put that debate to bed once and for all.
We place both players on two different teams in two different states and setup matches for them against (once again) two other different teams, with, by the way, different levels of talent on both sides of the ball. On balance, what we are attempting to recreate is the exact conditions current players are subjected to in determining their overall value as it compares to other players of like position. By now, you should already be able to poke holes all through the logic I am using to determine Emmitt and Barry’s overall value.
You might ask: What if one team they are up against has a great defense and the other team’s defense is crappy. Whoever faces the crappy defense is going to look better! That’s not fair!
But how is that any different from how we determine player’s value today?
You might say: Well, over the course of a season, it all evens out!
Does it? Do we all face the same teams? And do we all face the same teams in the same weather conditions? And do the same set of circumstances play out in every game making coaches largely grant all of their players the same snap count across the board?
You might say: Well, what analyst do is normalize the stats to make them comparable often times dividing their overall yardage (for instance) by their touch count. What you really should be looking at is their average yards per carry, attempt, catch, etc.
That may make things a bit fairer, but don’t player tend to tire the more snaps they get? Could that not negatively impact their overall average over time?
Please don’t misunderstand. I’m not attempting to sale you on the idea that’s all stats are garbage and essentially tell us nothing…I’m merely pointing out that most stats really prove a whole lot of nothing when attempting to compare one player to another player. Unfortunately, football is not played in a controlled environment and as such cannot be used to substantiate any claims that one player is better than the other.
Thoughts?