Thread: Penalties

"Del" wrote:

It's nice but it is relative to the situation. In football where 1 penalty could make or break a game, 1.32 seems significant.

Every penalty is significant, so it is kind of a misleading term.
When we say that there is no significant difference among the quarters, what we are really saying is that there is not a large enough difference to be able to confidently say that the difference was anything other than random.
Confidence level in this case looks to be about 95% (cant' say for sure without seeing the tables).

So basically we are less than 95% sure that the difference in number pf penalties was caused by anything other than chance.

"NodakPaul" wrote:

"Del" wrote:

It's nice but it is relative to the situation. In football where 1 penalty could make or break a game, 1.32 seems significant.

Every penalty is significant, so it is kind of a misleading term.
When we say that there is no significant difference among the quarters, what we are really saying is that there is not a large enough difference to be able to confidently say that the difference was anything other than random.
Confidence level in this case looks to be about 95% (cant' say for sure without seeing the tables).

So basically we are less than 95% sure that the difference in number pf penalties was caused by anything other than chance.

Ah I see what you are saying.

"Del" wrote:

"Zeus" wrote:

"josdin00" wrote:

The thing that frustrated me most about last season was the shear amount of penalties that were getting called on us. I don't know if it was just a one year problem of everyone figuringout the new systems, or whether it's more serious than that. I went through the play by plays on NFL.com and logged this info. This is just the penalties that cost us yards.

Wow, Josdin - great stuff.

Pokes a couple holes into Del's contention that Birk was getting called for tons of penalties last year.


Glad to see that the best player on the team, Antoine Winfield, doesn't get flagged too often.

=Z=

Actually it doesn't.

Matt Birk was called for multiple illegal shifts/illegal procedures for lifting the ball off the ground 2 alone in the the Greenbay game during week 16. I have the game film if anyone is interested, in fact I have all the games if anyone is interested. I got my info off the games not NFL.com which lists the penalties as MIN-ILLEGAL SHIFT.

He lifted the ball cost us 5 yards 2 Times in one game, and 2 Other times during the season. That is 8 penalties for our Pro Bowl center. NFL.Com is not a reliable source for penalties. I take it straight off the game film.

All he is counting is:

Week 1- Hold
Week 1- Ineligible downfield
Week 2- False Start
Week 4- Face Mask

Which is fine, and it is great information and illustrates we have a problem, but it is not all inclusive.

The illegal shifts for when the ball is lifted off the ground and reset are oddballs not attributed to anyone.

According to the game logs, there were 5 penalties that weren't attributed to a specific player. There were 2 illeagal shifts in week 16, like you mentioned, and if they were both on Birk, that puts him at 6 penalties for 35 yards. That's tied for third for most penalties on the team, and 5th for penalty yards.

The other "team fouls" were an Illeagal contact on the defense, 12 men on the field on the defense, and a "Illeagal block above the waist" on a kick return. (I have no idea why they don't have a player mentioned for the illeagal contact and the block in the back.)

"Del" wrote:

"NodakPaul" wrote:

"Del" wrote:

It's nice but it is relative to the situation. In football where 1 penalty could make or break a game, 1.32 seems significant.

Every penalty is significant, so it is kind of a misleading term.
When we say that there is no significant difference among the quarters, what we are really saying is that there is not a large enough difference to be able to confidently say that the difference was anything other than random.
Confidence level in this case looks to be about 95% (cant' say for sure without seeing the tables).

So basically we are less than 95% sure that the difference in number pf penalties was caused by anything other than chance.

Ah I see what you are saying.

Statistics are just a tool.
Sometimes something shows up as statistically significant but it has no significance in the terms of what you are studying.
For example, if you have a P < 0.01 you are just saying that you are 99% sure that what you are seeing is due to the treatment effects and not just the inherent variation.

If I am studying the phase of the moon in relation to a stallion recovering from castration I could conceivably come up with a statistical significance related to that event.
The reality may or may not be related to the phase of the moon and the experiemental design may have been flawed from the onset of the experiment.
It wasn't a true testable hypothesis.

The problem with using meaningful statisitics in a team sport is there are so many reasons a game was lost or won that it's difficult to pin it on one of the supposed indicators.
A game mey be lost because the TE blew his block and the gain was only 6 yds instead of a TD or the LG missed his pull or....

I think what Del was sayin is that with or without the statistical significance a single penalty is often enough to lose a game on.
Was it the only reason?
No.

"josdin00" wrote:

"Del" wrote:

"Zeus" wrote:

"josdin00" wrote:

The thing that frustrated me most about last season was the shear amount of penalties that were getting called on us. I don't know if it was just a one year problem of everyone figuringout the new systems, or whether it's more serious than that. I went through the play by plays on NFL.com and logged this info. This is just the penalties that cost us yards.

Wow, Josdin - great stuff.

Pokes a couple holes into Del's contention that Birk was getting called for tons of penalties last year.


Glad to see that the best player on the team, Antoine Winfield, doesn't get flagged too often.

=Z=

Actually it doesn't.

Matt Birk was called for multiple illegal shifts/illegal procedures for lifting the ball off the ground 2 alone in the the Greenbay game during week 16. I have the game film if anyone is interested, in fact I have all the games if anyone is interested. I got my info off the games not NFL.com which lists the penalties as MIN-ILLEGAL SHIFT.

He lifted the ball cost us 5 yards 2 Times in one game, and 2 Other times during the season. That is 8 penalties for our Pro Bowl center. NFL.Com is not a reliable source for penalties. I take it straight off the game film.

All he is counting is:

Week 1- Hold
Week 1- Ineligible downfield
Week 2- False Start
Week 4- Face Mask

Which is fine, and it is great information and illustrates we have a problem, but it is not all inclusive.

The illegal shifts for when the ball is lifted off the ground and reset are oddballs not attributed to anyone.

According to the game logs, there were 5 penalties that weren't attributed to a specific player. There were 2 illeagal shifts in week 16, like you mentioned, and if they were both on Birk, that puts him at 6 penalties for 35 yards. That's tied for third for most penalties on the team, and 5th for penalty yards.

The other "team fouls" were an Illeagal contact on the defense, 12 men on the field on the defense, and a "Illeagal block above the waist" on a kick return. (I have no idea why they don't have a player mentioned for the illeagal contact and the block in the back.)

That is true of NFL. Com no doubt.

What is also true of NFL.com is they have a Illegal formation attributed to B.Mckinnie during week two when it was actually Wiggins, and a week 7 holding on M. Johnson that was B. Mckinnie.

Point being it is not the end all of accuracy, in fact the most accurate would be the media books for the Vikings, just like the one Webby sent us a few years back. Actually if you still have yours and you look up the stats from it and match them to the stats on NFL.com you will see vast inaccuracies in all aspects weather it is team stats or personal stats.

The media guide is put together from the actual stats taken from the game and the field. Things like penalties, sacks, and tackles are often corrected in official statistics after the game and put into media guides. NFL.com does not bother to do that. In fact a refferee can call the wrong player (it happens all the time) and the penalty is credited to that one player and after the game when reviewed it is officially changed.

It's no knock on you or your work because it looks close enough, but Matt Birk did end the season with 8 penalties no doubt in my mind. The most accurate proof I can give you other then mailing you the game film is the media guide of which I do not have yet.

"Prophet" wrote:

"Del" wrote:

"NodakPaul" wrote:

"Del" wrote:

It's nice but it is relative to the situation. In football where 1 penalty could make or break a game, 1.32 seems significant.

Every penalty is significant, so it is kind of a misleading term.
When we say that there is no significant difference among the quarters, what we are really saying is that there is not a large enough difference to be able to confidently say that the difference was anything other than random.
Confidence level in this case looks to be about 95% (cant' say for sure without seeing the tables).

So basically we are less than 95% sure that the difference in number pf penalties was caused by anything other than chance.

Ah I see what you are saying.

Statistics are just a tool.
Sometimes something shows up as statistically significant but it has no significance in the terms of what you are studying.
For example, if you have a P < 0.01 you are just saying that you are 99% sure that what you are seeing is due to the treatment effects and not just the inherent variation.

If I am studying the phase of the moon in relation to a stallion recovering from castration I could conceivably come up with a statistical significance related to that event.
The reality may or may not be related to the phase of the moon and the experiemental design may have been flawed from the onset of the experiment.
It wasn't a true testable hypothesis.

The problem with using meaningful statisitics in a team sport is there are so many reasons a game was lost or won that it's difficult to pin it on one of the supposed indicators.
A game mey be lost because the TE blew his block and the gain was only 6 yds instead of a TD or the LG missed his pull or....

I think what Del was sayin is that with or without the statistical significance a single penalty is often enough to lose a game on.
Was it the only reason?
No.

You are absolutely correct, especially about non testable hypothesis.

However, in this case we weren't talking about the significance of penalties in regards to the outcome of the game.
In fact, we weren't talking about the significance of penalties at all.

Someone had mentioned that there were a lot more penalties in the second quarter as compared to the first.
However, Purplexing pointed out (and I agreed with) the fact that the number of penalties in the second quarter was not significantly greater than any of the other quarters.

By the way, since we are on the subject of stats,the one thing that annoys me more than anything else is when someone tries to infer cause from correlation. I see the media do it all the time, and it is just wrong.

21 in the first quarter.

42 in the second quarter.


I guarantee a coach is looking at that and trying to find out why. A math teacher may try to figure out wether it is worth looking at or not, but if your team is penalty prone going into the half then that is a problem.

Statistically there may be no significance, but there is substantial significance in application to the game. The fact remains this team is penalized more in the 2nd quarter then the rest.

"Del" wrote:

I guarantee a coach is looking at that and trying to find out why. A math teacher may try to figure out wether it is worth looking at or not, but if your team is penalty prone going into the half then that is a problem.

LMAO.
OK Del, you got me there. ;D
Most people would try to figure out why it happened, I was trying to decide if it was statistically a problem or not.

I guess that is why I am an engineer and not a coach! ;D

"NodakPaul" wrote:

"Del" wrote:

I guarantee a coach is looking at that and trying to find out why. A math teacher may try to figure out wether it is worth looking at or not, but if your team is penalty prone going into the half then that is a problem.

LMAO.
OK Del, you got me there. ;D
Most people would try to figure out why it happened, I was trying to decide if it was statistically a problem or not.

I guess that is why I am an engineer and not a coach! ;D

Maybe that is why you drive a porsche and I drive a voltswagon

For example, if you have a P < 0.01 you are just saying that you are 99% sure that what you are seeing is due to the treatment effects and not just the inherent variation.

So that is what my High School Football coach meant.


He said, the reason you guys have more penalties in the 2nd and 4th quarters is a direct reflection of your conditioning.
You guys are getting tired therefore you guys need to run more.
;D