 |
View Old Forum Thread
Old Forum Index » Football Beta Testing » Football Beta Test Discussion » New penalty accept/decline logic
Admin
Right now there are several tests, one of them being yards gained vs penalty yards, and the sim simply picks whichever is higher. Thus, the gain (which was likely 5.1 or 5.2 yards) is higher than the 5.0 yard penalty. Obviously this logic is very simplstic. A 5.5 yard gain should be ignored in favor of a 5.0 yard penalty plus replaying the down. A 6 yard gain, even, should probably be declined in favor of 5 yards penalty plus replaying the down. But where exactly are the lines drawn?
I found some data on expected chance of making a first down in each situation, from 1 to 30 yards to go and from 1st to 3rd down. I've put all this into a table and now I can directly compare situations to each other to determine which is "better".
So let's say it's 2nd and 12 and an 11 yard pass is completed, but the defense commits a 5 yard penalty. Would you rather have 2nd and 7 or 3rd and 1? The old logic would have picked 3rd and 1 hands down, but the new logic says that the chance of converting from 2nd and 7 is 62% while the chance of converting on 3rd and 1 is 58%, so the new logic would accept the penalty.
Right now I am testing this on my Dev server, but I'd like to get your opinions, especially of the example situation. What would you do if you were the coach?
--Chris
New penalty accept/decline logic
June 26, 2011 at 06:23PM View BBCode
We've had a few situations where the accept/decline logic appears to fall down, i.e a 5 yard penalty vs a 5 yard gain results in a decline when any sane coah would accept the penalty to get the down over again.Right now there are several tests, one of them being yards gained vs penalty yards, and the sim simply picks whichever is higher. Thus, the gain (which was likely 5.1 or 5.2 yards) is higher than the 5.0 yard penalty. Obviously this logic is very simplstic. A 5.5 yard gain should be ignored in favor of a 5.0 yard penalty plus replaying the down. A 6 yard gain, even, should probably be declined in favor of 5 yards penalty plus replaying the down. But where exactly are the lines drawn?
I found some data on expected chance of making a first down in each situation, from 1 to 30 yards to go and from 1st to 3rd down. I've put all this into a table and now I can directly compare situations to each other to determine which is "better".
So let's say it's 2nd and 12 and an 11 yard pass is completed, but the defense commits a 5 yard penalty. Would you rather have 2nd and 7 or 3rd and 1? The old logic would have picked 3rd and 1 hands down, but the new logic says that the chance of converting from 2nd and 7 is 62% while the chance of converting on 3rd and 1 is 58%, so the new logic would accept the penalty.
Right now I am testing this on my Dev server, but I'd like to get your opinions, especially of the example situation. What would you do if you were the coach?
--Chris
KLKRTR
But I'm wondering if there is a way to have a slider like we do with 2-pt conversion daring..
For example, I could shift it to the left if I wanted my penalties to favor more yardage, and I would shift it to the right if I wanted my penalties to favor more downs.
Now, how would that work?
Say, for example, I wanted more yardage always, regardless of the down. If I put that on my slider, it could add a value (let's say 5% here) to the percentage for a situation.
In the above example, with a neutral slider (neither left nor right):
2nd & 7 is 62% after accepting the penalty
3rd & 1 is 58% without the penalty
The penalty would be accepted.
If my slider favored yardage over down, the numbers would be adjusted as such:
2nd & 7 is 62% after accepting the penalty
3rd & 1 is 63% without the penalty
The penalty would be declined.
If my slider was only half-way to the yardage side, you could add 2.5% (for example):
2nd & 7 is 62% after accepting the penalty
3rd & 1 is 60.5% without the penalty
The penalty would be accepted.
These numbers wouldn't reflect the actual chances of converting, but rather, it would simulate the confidence a coach has in either strategy (or in neither and sticks to playing the percentages exactly).
June 26, 2011 at 06:42PM View BBCode
I would actually go for the 3rd and 1....But I'm wondering if there is a way to have a slider like we do with 2-pt conversion daring..
For example, I could shift it to the left if I wanted my penalties to favor more yardage, and I would shift it to the right if I wanted my penalties to favor more downs.
Now, how would that work?
Say, for example, I wanted more yardage always, regardless of the down. If I put that on my slider, it could add a value (let's say 5% here) to the percentage for a situation.
In the above example, with a neutral slider (neither left nor right):
2nd & 7 is 62% after accepting the penalty
3rd & 1 is 58% without the penalty
The penalty would be accepted.
If my slider favored yardage over down, the numbers would be adjusted as such:
2nd & 7 is 62% after accepting the penalty
3rd & 1 is 63% without the penalty
The penalty would be declined.
If my slider was only half-way to the yardage side, you could add 2.5% (for example):
2nd & 7 is 62% after accepting the penalty
3rd & 1 is 60.5% without the penalty
The penalty would be accepted.
These numbers wouldn't reflect the actual chances of converting, but rather, it would simulate the confidence a coach has in either strategy (or in neither and sticks to playing the percentages exactly).
tworoosters
New Orleans pick up 10 yards forcing a fourth but also moving into FG range, there is a 10 yard penalty on New Orleans which is declined and New Orleans kick a 43 yard FG to increase the lead to six.
I'm wondering if, in real life, LA would've accepted that penalty to force 3rd and 24 from the 47 ?
June 26, 2011 at 09:57PM View formatted
You are viewing the raw post code; this allows you to copy a message with BBCode formatting intact.
There's another aspect to be considered as well, in the [url=http://footballbeta.simdynasty.com/boxscore.jsp?boxscoreid=3124]wild card game[/url] with 2 minutes to go New Orleans have 3rd and 14 from the LA 37, with a three point lead.
New Orleans pick up 10 yards forcing a fourth but also moving into FG range, there is a 10 yard penalty on New Orleans which is declined and New Orleans kick a 43 yard FG to increase the lead to six.
I'm wondering if, in real life, LA would've accepted that penalty to force 3rd and 24 from the 47 ?
RichNYC1
June 27, 2011 at 01:32AM View BBCode
They have a chart for this just like they do for the 2 pt. conversion. Im looking for it...
Admin
Promising idea... in implementation it might go too far, and I don't think I could display a clean chart like I do on the 2-point conversion settings, but I am thinking at some point this has to be adjustable. The 2-point chart is based on ideal math, too, which NFL coaches don't always use. Even though this penalty strategy is mathematically optimal, in its default configuration it needs to look like what NFL coaches actually do.
--Chris
June 27, 2011 at 04:24AM View BBCode
Originally posted by KLKRTR
I would actually go for the 3rd and 1....
But I'm wondering if there is a way to have a slider like we do with 2-pt conversion daring..
For example, I could shift it to the left if I wanted my penalties to favor more yardage, and I would shift it to the right if I wanted my penalties to favor more downs.
Now, how would that work?
Say, for example, I wanted more yardage always, regardless of the down. If I put that on my slider, it could add a value (let's say 5% here) to the percentage for a situation.
In the above example, with a neutral slider (neither left nor right):
2nd & 7 is 62% after accepting the penalty
3rd & 1 is 58% without the penalty
The penalty would be accepted.
If my slider favored yardage over down, the numbers would be adjusted as such:
2nd & 7 is 62% after accepting the penalty
3rd & 1 is 63% without the penalty
The penalty would be declined.
If my slider was only half-way to the yardage side, you could add 2.5% (for example):
2nd & 7 is 62% after accepting the penalty
3rd & 1 is 60.5% without the penalty
The penalty would be accepted.
These numbers wouldn't reflect the actual chances of converting, but rather, it would simulate the confidence a coach has in either strategy (or in neither and sticks to playing the percentages exactly).
Promising idea... in implementation it might go too far, and I don't think I could display a clean chart like I do on the 2-point conversion settings, but I am thinking at some point this has to be adjustable. The 2-point chart is based on ideal math, too, which NFL coaches don't always use. Even though this penalty strategy is mathematically optimal, in its default configuration it needs to look like what NFL coaches actually do.
--Chris
Admin
Right now the penalty logic works like this:
- Determine if the penalty is a "half to goal" penalty
- If offensive penalty and there's been a defensive touchdown or a safety, decline
- If defensive penalty and the offense has scored a touchdown, decline
- If it's 3rd or 4th down AND it's not a loss of down penalty AND it's an offensive penalty AND the offense didn't make a first down AND the offense didn't score, decline the penalty (to force a turnover on downs or a 4th down situation)
- If it's an offensive penalty AND there was a turnover on the play, decline the penalty
- If the penalty is against the defense AND the penalty is not enforced at the spot or from the spot of the infraction, and there has not been a turnover or a defensive touchdown, and none of the above rules have already applied:
--- If it's an automatic first down and the offense did not make first down and did not score, accept it
--- If the offense made a first down, decline unless the penalty moves the ball farther
--- If neither of those two conditions was true, use the new logic
- If the penalty was against the offense and there was a turnover, decline
--Chris
June 27, 2011 at 07:23AM View BBCode
I've put the new logic in place... I agree that we need some method of controlling it, but for now please keep an eye out for logical flaws as I had to change quite a bit of the penalty logic.Right now the penalty logic works like this:
- Determine if the penalty is a "half to goal" penalty
- If offensive penalty and there's been a defensive touchdown or a safety, decline
- If defensive penalty and the offense has scored a touchdown, decline
- If it's 3rd or 4th down AND it's not a loss of down penalty AND it's an offensive penalty AND the offense didn't make a first down AND the offense didn't score, decline the penalty (to force a turnover on downs or a 4th down situation)
- If it's an offensive penalty AND there was a turnover on the play, decline the penalty
- If the penalty is against the defense AND the penalty is not enforced at the spot or from the spot of the infraction, and there has not been a turnover or a defensive touchdown, and none of the above rules have already applied:
--- If it's an automatic first down and the offense did not make first down and did not score, accept it
--- If the offense made a first down, decline unless the penalty moves the ball farther
--- If neither of those two conditions was true, use the new logic
- If the penalty was against the offense and there was a turnover, decline
--Chris
KLKRTR
June 27, 2011 at 05:59PM View BBCode
With the new penalty logic, are we going to add the Personal foul yardage to the end of the play?
RichNYC1
Thats a good point...
I like this plan, Chris. Been looking for that chart and cant find it, but this seems like a good solution
June 28, 2011 at 01:03AM View BBCode
Originally posted by KLKRTR
With the new penalty logic, are we going to add the Personal foul yardage to the end of the play?
Thats a good point...
I like this plan, Chris. Been looking for that chart and cant find it, but this seems like a good solution
KLKRTR
Anyone studied Game Theory before? I actually signed up for a class on it a couple of years ago, but I ended up switching out of it. Just found an article using it with football, and it's a pretty awesome read. If you find it, would probably help people with their strategies.
-------------------------------------------------------------
Now, I also found something called "Expected Points". Basically, this guy did a study of how many points a team would be "expected" to score if they had a certain down at a certain point. So while mathematically optimal conversion numbers are what you're using now, I think this may be an easier (and more feasible) way to give players some choice in what do to do about questionable penalty acceptances/declines. (Link: [url]http://www.advancednflstats.com/2009/09/4th-down-study-part-1.html[/url])
[img=50x50]http://farm3.static.flickr.com/2675/3688474051_95851a0ed5_o.png[/img ]
This Graph shows the data for first down.
Here's the example this guy uses:
"We can also look at each option in terms of Expected Points, the average point advantage an offense can expect given a particular combination of down, distance and field position. In this case, the 1st and 5 is the better option. Between the 20s, where the EP curve is nearly linear, the average value of 2nd and 2 situations is 0.2 EP less than the average value of 1st and 5 situations (accounting for being 3 yards further back).
The advantage for the 1st and 5 holds throughout the range of field positions between the 20s. As an example, a 1st and 5 at midfield is worth 2.5 EP. A 2nd and 2 at an opponent?s 47, is worth 2.3.
The 1st and 5 is so valuable simply because it gives you an extra down. It?s essentially an opportunity for a free shot downfield, which is likely why the conversion probabilities are equal but the EP values are not. The defense is really in a bind because it has to guard equally against any play type. That?s true of 2nd and short as well, but the 1st and 5 gives the offense an additional down of unpredictable abandon.
Coaches should accept the penalty on gains anything short of 8 yards, simply on the basis of the probability of conversion. And on an 8-yard gain, coaches should accept the penalty on the basis of expected points. A 9-yard gain is essentially a wash, and a 2nd and 1 can be very lucrative if taken advantage of.
One caveat: This analysis is based on the assumption that between the 20s, 1st and 5s are roughly equally advantageous over a 1st and 10 at the same field position. (There just aren't that many 1st and 5s to be certain.) But as I'll show in a future post, it's possible this hasn't been the case. In fact, the data might suggest that 1st and 5s have actually been less valuable than a 1st and 10 at the same field position. "
Thoughts?
June 28, 2011 at 03:09AM View BBCode
Basically we just need this: [url]http://www.sciencedaily.com/releases/2006/04/060420232621.htm[/url]Anyone studied Game Theory before? I actually signed up for a class on it a couple of years ago, but I ended up switching out of it. Just found an article using it with football, and it's a pretty awesome read. If you find it, would probably help people with their strategies.
-------------------------------------------------------------
Now, I also found something called "Expected Points". Basically, this guy did a study of how many points a team would be "expected" to score if they had a certain down at a certain point. So while mathematically optimal conversion numbers are what you're using now, I think this may be an easier (and more feasible) way to give players some choice in what do to do about questionable penalty acceptances/declines. (Link: [url]http://www.advancednflstats.com/2009/09/4th-down-study-part-1.html[/url])
[img=50x50]http://farm3.static.flickr.com/2675/3688474051_95851a0ed5_o.png[/img ]
This Graph shows the data for first down.
Here's the example this guy uses:
"We can also look at each option in terms of Expected Points, the average point advantage an offense can expect given a particular combination of down, distance and field position. In this case, the 1st and 5 is the better option. Between the 20s, where the EP curve is nearly linear, the average value of 2nd and 2 situations is 0.2 EP less than the average value of 1st and 5 situations (accounting for being 3 yards further back).
The advantage for the 1st and 5 holds throughout the range of field positions between the 20s. As an example, a 1st and 5 at midfield is worth 2.5 EP. A 2nd and 2 at an opponent?s 47, is worth 2.3.
The 1st and 5 is so valuable simply because it gives you an extra down. It?s essentially an opportunity for a free shot downfield, which is likely why the conversion probabilities are equal but the EP values are not. The defense is really in a bind because it has to guard equally against any play type. That?s true of 2nd and short as well, but the 1st and 5 gives the offense an additional down of unpredictable abandon.
Coaches should accept the penalty on gains anything short of 8 yards, simply on the basis of the probability of conversion. And on an 8-yard gain, coaches should accept the penalty on the basis of expected points. A 9-yard gain is essentially a wash, and a 2nd and 1 can be very lucrative if taken advantage of.
One caveat: This analysis is based on the assumption that between the 20s, 1st and 5s are roughly equally advantageous over a 1st and 10 at the same field position. (There just aren't that many 1st and 5s to be certain.) But as I'll show in a future post, it's possible this hasn't been the case. In fact, the data might suggest that 1st and 5s have actually been less valuable than a 1st and 10 at the same field position. "
Thoughts?
Hamilton2
June 28, 2011 at 03:52PM View BBCode
I think I should get extra credit for inviting you ... overachiever! ;)
KLKRTR
I just enjoy stuff like this. If I could actually get paid to do it, I'd consider leaving science for the sports world. Even with the two science majors, I still love the math: math minor and planned MBA lol. I'm not an overachiever, just spending time on something I enjoy.
June 28, 2011 at 07:12PM View BBCode
Haha, thanks..I just enjoy stuff like this. If I could actually get paid to do it, I'd consider leaving science for the sports world. Even with the two science majors, I still love the math: math minor and planned MBA lol. I'm not an overachiever, just spending time on something I enjoy.
Admin
--Chris
June 29, 2011 at 05:11AM View BBCode
A lot of things I've done are based on research I've done at that site... it's a gold mine. Even if I don't use the data directly, it makes me think. In fact, it was research there that led to the new penalty logic.--Chris
Pages: 1