NFL Penalty Tracker
10/15/2012 12:02 pm
One thing I would like to add, and perhaps you guys could help come up with/tweak the formula is a "killer penalty" number.
So, say, a defensive offsides that gives the offense the ball back is a bigger deal than a defensive offsides on 1st and 10.
I don't think coming up for rules on a yes/no system would be too hard, but I'd love if we could actually come up with some sort of a number.
Something that took down, distance, if it gave the team a first, yardage, etc into and came up with something that Rated 0-10, or whatever, 0 being declined, 10 being a leaping penalty on 4th and 98.
 | Scott - 6225 Posts 10/15/2012 @ 12:22:31 PM |
 |
|---|---|---|
| I like where this could lead, but it could the data could easily be misinterpreted. It's probably obvious but being the benificiary of lots of penalties probably doesn't mean that the Refs are in your corner, just like being called for a lot of penalties probably doesn't mean the refs are out to get you. Likewise, a "killer penalty" like an offsides on a failed 3rd down attempt that gives you a first down might have as much to do with offending team being undisciplined as it does with the beneficiary having a QB that is really good at drawing teams offsides. So you might have a team that draws lots of defensive pass intereference penalties, and seems to have lots of "killer penalties" that they benefit from, but that could just mean that other teams are doing everything they can to keep up with or slow down that team and penalties arise from it. Other than that, I like where this is heading. I'm already drawing up plans for my backyard pool I'm going to install from the money we get when ESPN pays us for this (I'm in on the payroll, right?) |
Yours:
 
Rated 0 times. |
|
| Scott screwed with this at 10/15/2012 12:34:23 pm | ||
|
| Scott - 6225 Posts 10/15/2012 @ 12:32:12 PM |
|
|---|---|---|
|
Any defensive penalty on 3rd down should be weighted slightly higher than a 1st or second down penalty, and even more heavily if that 3rd down penalty results in a first down. Any defensive penalty that eliminates a turnover (like a RTP penalty on a play where the QB throws an INT or fumbles the ball). Any 4th down defensive penalty is pretty bad. Obviously the yardage of the penalty matters, like if the penalty results in a first down. I would say a 3rd down penalty that doesn't result in a first down is still worse than a 4th down penalty that doesn't result in a first down because the 3rd down penalty would result in replaying 3rd down, where the offense is still going to run a play and attempt to get a first down; whereas a 4th down penalty that doesn't result in a first down, unless it results in a 4th and 1, is still likely going to result in the offense punting. Any penalty on a field goal attempt is pretty bad, either offense or defense. It's bad on defense because it at best gives the offense a second try at a field if the first one was no good, and at worst it gives the offense a first down deep inside enemy territory. On offense, a penalty will result in having to re-attempt the field goal try, and from 5-10-15 yards further away, that is if you made the first attempt and the defense accepts the penalty. I think the time of the game matters too. A penalty on 4th down in the 1st qtr can be made up for throughout the game. A defensive penalty on 4th and 3 that results in a first down with the offense trailing by 2 and the ball at the opponents 40 yard line with 37 seconds left might be a back breaking penalty that sort of decides the game. I have no numbers thought up yet, but there would be many scenarios to consider, if that's where you're going with this. |
Yours:
Rated 0 times. |
|
 | Jeremy - 9475 Posts 10/15/2012 @ 12:45:29 PM |
|
|---|---|---|
|
Well, all of those things apply to penalties in general too. The goal wasn't really to find out who has the refs in their pocket, but rather to find some way to quantify the fact that all 5 yard penalties are not the same. For now it pretty much as to be things that can be derived from JUST down/distance, the penalty, and the yardage. I'd have to manually go through the system and define which penalties have auto firsts, but it seems like that would be vital. Things like "what it undid", especially things like undoing a turnover, or led to a TD would be hard to impossible to do automatically. Same with "on a field goal". I might be able to get out of the play data easily enough what it undid for offensive penalties. (So like a hold that brings back a 40 yard run.) But even that can't be promised. For the most part it needs to be something that just considers the down, and the distance/penalty yardage (which for the most part will just mean first down or not) Quarter/Time could be a factor, though I'm not sure it should. Score probably could be as well, but might be complicated. The main thing is that it needs to be an equation/algorithm. Something like first down = 10 (10 + $penalty_yards/15 + blah - blah) / blah = Killer Penalty Factor It can't be something we evaluate/rule on, even if it means a low score for the occational brutal penalty, or a high score for sometihng that backed a punt up 5 meaninglessish yards. Although, I suppose that brings me to maybe needing to pull field position data. Now I remember why I gave this up.  |
Yours:
Rated 0 times. |
|
| Jeremy screwed with this 2 times, last at 10/15/2012 12:48:59 pm | ||
 | Jeremy - 9475 Posts 10/15/2012 @ 12:57:03 PM |
|
|---|---|---|
|
It might be the case we need a completely different system for offensive and defensive penalties too. Or maybe something that, for now, only looks at defensive penalties. Those are much more straight forward. You're giving the offense ___. Offensive penalties are harder. You generally get to replay the down, a penalty on 3rd down is often worse than one on 4th, but mostly even if I can figure out that a hold cost you a 40 yard run/pass, you might have only gotten that because of the hold, so it's murky as to what it truly costs. If an offense accepts a penalty that gets them a 1st down, it's clear they didn't otherwise earn it. That isn't necessarily true in reverse. |
Yours:
Rated 0 times. |
|
| Jeremy screwed with this 3 times, last at 10/15/2012 1:28:31 pm | ||
|
| Scott - If you aren't enough without it, you'll never be enough with it. 10/15/2012 @ 01:30:01 PM |
|
|---|---|---|
|
I'm trying to decide if the "what it leads to" should matter. I'm thinking now, accept for the case of an offensive penalty on a field goal that negates 3 points, and then possibly considering whether the FG retry was good or not. But even considering the retry might be unnecessary. I mean, committing a penalty that takes points off of the board is pretty bad. The fact that you still ended up making the ensuing field goal doesn't negate the killer-ness of a penalty. In a sense, the same goes for defensive penalties. If you commit a personal foul on a third down stop and then the offense scores a touchdown 4 plays later, that doesn't make the foul worse than if the offense throws an interception the very next play. It might be logical to conclude that without that penalty the TD wouldn't have happened, but regardless of the actions that follow, the act itself was pretty killer, and the only thing that really was "directly" related is anything that happened on that play and that play only (like a holding call that negates a 40 yard run, or a FG that is negated). It's kind of like not bringing a change of clothes or rain gear with you camping is pretty stupid. Just because it didn't rain on your trip doesn't change the fact that you were stupid for not bringing a pancho. |
Yours:
Rated 0 times. |
|
|
| Scott - 6225 Posts 10/15/2012 @ 01:32:41 PM |
|
|---|---|---|
| I agree with your analysis of offense penalties, and that thought was in the back of my mind too. A 40 yard run looks huge and you feel screwed when a holding call brings it back, but the hold probably resulted in the 40 yard run, not "the 40 yard run was ruined by the hold". |
Yours:
Rated 0 times. |
|
|
| Jeremy - 9475 Posts 10/15/2012 @ 01:32:44 PM |
|
|---|---|---|
| One other thing we count account for is an avoidability factor for each penalty. So, something like a false start is fairly inevitable. PI is avoidable, but will probably happen over the course of playing good coverage. Personal Foul penalties are, like 90% of the time, just someone being a total dumbass, or otherwise having nothing to do with the play. They're just a totally superfluous gift to the other team. |
Yours:
Rated 0 times. |
|
|
| Jeremy - I hate our freedoms 10/15/2012 @ 01:37:43 PM |
|
|---|---|---|
|
Scott Wrote - Today @ 01:30:01 PM I'm trying to decide if the "what it leads to" should matter. I'm thinking now, accept for the case of an offensive penalty on a field goal that negates 3 points, and then possibly considering whether the FG retry was good or not. But even considering the retry might be unnecessary. I mean, committing a penalty that takes points off of the board is pretty bad. The fact that you still ended up making the ensuing field goal doesn't negate the killer-ness of a penalty. In a sense, the same goes for defensive penalties. If you commit a personal foul on a third down stop and then the offense scores a touchdown 4 plays later, that doesn't make the foul worse than if the offense throws an interception the very next play. It might be logical to conclude that without that penalty the TD wouldn't have happened, but regardless of the actions that follow, the act itself was pretty killer, and the only thing that really was "directly" related is anything that happened on that play and that play only (like a holding call that negates a 40 yard run, or a FG that is negated). It's kind of like not bringing a change of clothes or rain gear with you camping is pretty stupid. Just because it didn't rain on your trip doesn't change the fact that you were stupid for not bringing a pancho. Agreed, but we really couldn't do that other stuff if we wanted to anyway. I only have the information for the plays that got penalized. I could re grab every game, I suppose, but it would be very complicated even if I had the data for the whole drive. If for no other reason than the data isn't made to be used like this. The logic to turn english descriptions of the plays into the penalty data is screwy enough, and that would be nothing compared to a generic "what happened" parser. Plus what if the defense gives an offense 2 more sets of downs on one drive. I think the best we could do here is use field position, assuming I can get that. You're more likely giving away points if you give a team 3 more downs on the 15 than the 50, but then each gift is what it is. (From there we more or less assume the worst.) |
Yours:
Rated 0 times. |
|
| Jeremy messed with this at 10/15/2012 1:40:29 pm | ||
|
| Scott - If you aren't enough without it, you'll never be enough with it. 10/15/2012 @ 01:54:07 PM |
|
|---|---|---|
| what is your source of all this information? (unless, of course, that's a deeply guarded secret that if revealed would compromise the entire endeavor and the lives of those involved) |
Yours:
Rated 0 times. |
|
|
| Scott - Get Up! Get outta here! Gone! 10/15/2012 @ 01:58:22 PM |
|
|---|---|---|
|
Now I'm thinking if the situation matters at all. Maybe it does. But since some penalties are automatic first downs, why would we treat a non-automatic first down penalty worse just because it gave the offense a first down? In other words, a 5 yard offsides penalty on 3rd and 10 is a free play for the offense, but if the offense fails it's still 3rd down. on 3rd and 5 or less, it's a first down regardless. But, does that really matter? Or is that the point of this whole thing? Is the point to assume that a penalty that gives a team a first down (or some other extra-ordinary positive result) when they otherwise didn't earn it inherently has more impact than a penalty that doesn't provide a first down-equivalent return to the offense? I'm not challenging anything here, more thinking outloud. |
Yours:
Rated 0 times. |
|
| Scott edited this at 10/15/2012 1:58:48 pm | ||
|
| Jeremy - Pie Racist 10/15/2012 @ 02:04:01 PM |
|
|---|---|---|
This is what I get:{ |
Yours:
Rated 0 times. |
|
| Jeremy edited this at 10/15/2012 2:05:23 pm | ||
|
| Scott - Resident Tech Support 10/15/2012 @ 02:08:00 PM |
|
|---|---|---|
| I meant, where you pulling the information from? Is it just a query that pulls the play-by-play information from espn or something? |
Yours:
Rated 0 times. |
|
|
| Scott - Resident Tech Support 10/15/2012 @ 02:10:59 PM |
|
|---|---|---|
| Oh, is that mass of code something that you are given? In other words, you didn't write that code, it's just something that is provided from somewhere? |
Yours:
Rated 0 times. |
|
| Scott messed with this at 10/15/2012 2:11:22 pm | ||
|
| Jeremy - 9475 Posts 10/15/2012 @ 02:11:33 PM |
|
|---|---|---|
| NFL.com. It's more than play by play data, but that's part of it. Each individual play that had a penalty is all I store though. |
Yours:
Rated 0 times. |
|
|
| Jeremy - 9475 Posts 10/15/2012 @ 02:26:32 PM |
|
|---|---|---|
| I am given that info about the play. Ideally anything and everything we do can be derived either from just that data, or from information that we know about the penalty itself. (In this case, that Unnecessary Roughness is an automatic first down.) |
Yours:
Rated 0 times. |
|
| Jeremy perfected this at 10/15/2012 2:28:45 pm | ||
|
| Jeremy - Robots don't say 'ye' 11/21/2012 @ 10:10:20 AM |
|
|---|---|---|
| http://www.nflpenalties.com/ Stepping up in the world! |
Yours:
Rated 0 times. |
|
|
| Scott - 6225 Posts 12/03/2012 @ 10:53:59 AM |
|
|---|---|---|
|
http://www.nflpenalties.com/player/d-barclay-green-bay-packers?year=2012 Apparently Packers rookie right tackle, Don Barclay (making is NFL debut yesterday) is super human? He apparently played for 5 different teams on Sunday, committing 9 penalties, including an offensive pass interference call. I'm not sure what is more impressive, the fact that he played for 5 different teams, or the fact that he committed penalties for teams that were playing each other. Is Don Barclay the real-life incarnation of Bill Brasky? Is his family seal a barracuda eating Neal Armstrong, too? |
Yours:
Rated 0 times. |
|
| Scott edited this at 12/03/2012 10:55:08 am | ||
|
| Jeremy - Robots don't say 'ye' 12/03/2012 @ 10:35:17 PM |
|
|---|---|---|
| He was a busy guy. |
Yours:
Rated 0 times. |
|
Leave a Comment of your very own
| Name: | |||
| Comment: | |||
| |||
Sign up, or login, to be able to upload files for Nutcan.com users to see.
Fetching Nutcan Content....
Rated 0 times.