Post by Reignman on Feb 12, 2017 15:39:55 GMT -6
I like to compile my own data by watching the games sometimes. Sometimes I have to because the information I'm looking for just isn't available at the local stat site. And I like to be as accurate as possible, but sometimes I run into things that aren't as objective and require a little judgement. Like this play for example.
I'm collecting some info on pass plays and part of that info is how far the ball traveled through the air. So how would you guys score an over throw like the pic above (or even an under throw for that matter)? Would you go by where the ball landed? In this case the 4 yard line? The LOS was the 34 yard line so this would go down as a 30 yard pass? Or maybe you go by where the intended target is when the ball arrived? In this case Johnson at the 8 yard line? That would make it a 26 yard pass.
Common sense tells me it should go by where the target is, after all that's where the QB should be trying to throw the ball. He should be trying to complete a 26 yard pass here but put a little too much on it. Not to mention a receiver never really catches the ball where it would have hit the ground. On the other hand, maybe Johnson got slowed up by the defender? Had he been there, maybe he catches it at the 6 and it goes down as 28 air yards?
The data I have from a site that specializes in these types of statistics, gave Hill credit for 29 air yards on this pass, which puts the ball at the 5, so I'm not exactly sure what they go by. They must give the QB the benefit of the doubt on passes. Perhaps if Johnson is there it could have been caught at the 5.
Here's another example.
This looks like a 9 yard pass to me. It is if Diggs catches it. Maybe 10. But it's wide of Diggs and lands at the 39 yard line. The stat experts have it down as 14 air yards. Ok yeah technically it was in the air for 14 yards, but it was only meant to be 9-10. To be fair Diggs was interfered with on this play (which wasn't called), so it probably would have been at most 12.
Last year I did all of the Vikings games before I had access to some expert data, so I just went by where the receiver was when the ball arrived, unless the receiver was impeded, then I went by where the ball would have been caught, which is usually 1-2 yards ahead of where the ball lands, depending on the trajectory. Now I'm not so sure. Should I stick with my own formula or try to figure out how the experts do it?
The expert way bothers me a bit though, like on that Diggs pass. I really don't see how that can go down as 14 yards lol. It tells me the experts are padding incompletes with an extra 4-5 yards simply because a receiver wasn't there to stop the balls forward momentum. For example. In this one game so far I've documented 24 completions and 17 incompletions. I'm never off by more than 1 yard from what the experts have on any completion, and in most cases we agree. However on the incompletions, it's all over the place. I'm plus or minus 1-2 yards on every one of them (overthrow/underthrow?), and sometimes 3 or 4, or even 5 like on the Diggs play from above.
On this play I have to question their sanity however. They say the ball traveled 1 yard through the air lol. Ummm yeah, that's clearly 5. I gave it 6 because Rudolph was the intended target, but the defender knocked it down before it got there. Which sounds more accurate?
They gave this one 10 yards. Again in what world is this 10 yards downfield? If this was a 3rd down play it would count as a past the sticks throw when it's clearly not. Rudolph could have easily been tackled short of the sticks. Rudolph bobbles it for a bit and it lands incomplete at 10. Is that what they go by? Dumb! If it misses Rudolph it probably goes 12. You see the defender going for a potential diving INT in that area.
I don't think my OCD will allow me to go along with the "experts" on incompletions. Their method doesn't seem to make much sense. Maybe they have a bunch of different people compiling the data and needed a way to remove subjectivity.
I'm collecting some info on pass plays and part of that info is how far the ball traveled through the air. So how would you guys score an over throw like the pic above (or even an under throw for that matter)? Would you go by where the ball landed? In this case the 4 yard line? The LOS was the 34 yard line so this would go down as a 30 yard pass? Or maybe you go by where the intended target is when the ball arrived? In this case Johnson at the 8 yard line? That would make it a 26 yard pass.
Common sense tells me it should go by where the target is, after all that's where the QB should be trying to throw the ball. He should be trying to complete a 26 yard pass here but put a little too much on it. Not to mention a receiver never really catches the ball where it would have hit the ground. On the other hand, maybe Johnson got slowed up by the defender? Had he been there, maybe he catches it at the 6 and it goes down as 28 air yards?
The data I have from a site that specializes in these types of statistics, gave Hill credit for 29 air yards on this pass, which puts the ball at the 5, so I'm not exactly sure what they go by. They must give the QB the benefit of the doubt on passes. Perhaps if Johnson is there it could have been caught at the 5.
Here's another example.
This looks like a 9 yard pass to me. It is if Diggs catches it. Maybe 10. But it's wide of Diggs and lands at the 39 yard line. The stat experts have it down as 14 air yards. Ok yeah technically it was in the air for 14 yards, but it was only meant to be 9-10. To be fair Diggs was interfered with on this play (which wasn't called), so it probably would have been at most 12.
Last year I did all of the Vikings games before I had access to some expert data, so I just went by where the receiver was when the ball arrived, unless the receiver was impeded, then I went by where the ball would have been caught, which is usually 1-2 yards ahead of where the ball lands, depending on the trajectory. Now I'm not so sure. Should I stick with my own formula or try to figure out how the experts do it?
The expert way bothers me a bit though, like on that Diggs pass. I really don't see how that can go down as 14 yards lol. It tells me the experts are padding incompletes with an extra 4-5 yards simply because a receiver wasn't there to stop the balls forward momentum. For example. In this one game so far I've documented 24 completions and 17 incompletions. I'm never off by more than 1 yard from what the experts have on any completion, and in most cases we agree. However on the incompletions, it's all over the place. I'm plus or minus 1-2 yards on every one of them (overthrow/underthrow?), and sometimes 3 or 4, or even 5 like on the Diggs play from above.
On this play I have to question their sanity however. They say the ball traveled 1 yard through the air lol. Ummm yeah, that's clearly 5. I gave it 6 because Rudolph was the intended target, but the defender knocked it down before it got there. Which sounds more accurate?
They gave this one 10 yards. Again in what world is this 10 yards downfield? If this was a 3rd down play it would count as a past the sticks throw when it's clearly not. Rudolph could have easily been tackled short of the sticks. Rudolph bobbles it for a bit and it lands incomplete at 10. Is that what they go by? Dumb! If it misses Rudolph it probably goes 12. You see the defender going for a potential diving INT in that area.
I don't think my OCD will allow me to go along with the "experts" on incompletions. Their method doesn't seem to make much sense. Maybe they have a bunch of different people compiling the data and needed a way to remove subjectivity.
... and neither is this Chris guy.