|
Post by fantom on Nov 13, 2015 13:24:09 GMT -6
You guys are right, it does not matter where you coach. Good football is good football, and Texas does not have a monopoly on it. I apologize for that comment, I was wrong. We will just have to agree to disagree here on the statistical stuff. I do think agree that winning the kicking game is vitally important. Of course it is. And if you think that statistical goals are worthwhile and, more importantly, have the time and manpower to do it, have at it. I can also help you get off the hook for the Texas comment. If you're at a 6A in Texas I assume you have a pretty big staff. In effect, you DO have GA's.
|
|
|
Post by coachd5085 on Nov 13, 2015 14:28:15 GMT -6
You guys are right, it does not matter where you coach. Good football is good football, and Texas does not have a monopoly on it. I apologize for that comment, I was wrong. We will just have to agree to disagree here on the statistical stuff. I do think agree that winning the kicking game is vitally important. No, the stat stuff really isn't that important, for the very reasons explained in this thread. They simply show you HOW you did...but we already have that metric in place. Its called the scoreboard.
|
|
scottc
Sophomore Member
Posts: 149
|
Post by scottc on Nov 16, 2015 14:25:57 GMT -6
A book called "Saturday Morning Wakeup Call" (Warning: heavy math and statistics formulas) Really gets into this theory of these types of stats at the high school level. The Points Per Possession (PPP) stat really is one that should be used by the following criteria according to this book and my understanding:
The opponents PPP is a stat used to determine if you should go for it in a certain situation based on field position you are risking compared to the PPP differential by punting and the field position gained. If your punter nets 30 or less many times you are better off [mathematically speaking] going for it than punting because of the PPP equity gain is not greater than your make % on 4th down.
similar to poker in # of outs compared to the pot odds you get if you are chasing a flush or straight draw etc.
example: assuming you have an average defense (all numbers are made up to show the author's thinking)
You are facing 4th and 2 at the -18. Your avg net punt is say 30 yards. your opponent has a PPP of 5.9 [theoretical] when starting in plus territory and a 6.1 PPP if starting +20 or better [theoretical]. On the flip side your defense PPP is 4.8 when starting on our end and up to 5.3 when the opp starts on the +20 or better. The % difference in these is 18.6% on a punt and 13.1% on a go for and not make it. If your 4th and 2 make % is 40% then the odds favor you going for it based on the equity of you making compared to the equity you give up in PPP % compared to not making it on 4th down or the equity gained in punting the ball 18.6%.
The data says its a losing proposition if you punt it there. Not saying I agree or disagree because being brought up in the 70s where field position etc were etched into my brain. However, this is the same line of thinking Belichick used in his infamous 4th and 2 call on -28 vs. the Colts and Manning a few years back on SNF or MNF. Better the opp offense then the more aggressive you should be based on the PPP stat on 4th down. It has really caused me to really think hard on what might be "automatic" punts in situations. If you are a numbers guy I suggest you read the book mentioned. Also says that teams do not go for 2 enough or onside kick enough.
|
|
|
Post by Chris Clement on Nov 16, 2015 15:30:25 GMT -6
That sounds more like Expected Points, which is extremely useful, but can't really be divined for high school.
|
|
|
Post by coachd5085 on Nov 16, 2015 19:56:03 GMT -6
A book called "Saturday Morning Wakeup Call" (Warning: heavy math and statistics formulas) Really gets into this theory of these types of stats at the high school level. The Points Per Possession (PPP) stat really is one that should be used by the following criteria according to this book and my understanding: The opponents PPP is a stat used to determine if you should go for it in a certain situation based on field position you are risking compared to the PPP differential by punting and the field position gained. If your punter nets 30 or less many times you are better off [mathematically speaking] going for it than punting because of the PPP equity gain is not greater than your make % on 4th down. similar to poker in # of outs compared to the pot odds you get if you are chasing a flush or straight draw etc. example: assuming you have an average defense (all numbers are made up to show the author's thinking) You are facing 4th and 2 at the -18. Your avg net punt is say 30 yards. your opponent has a PPP of 5.9 [theoretical] when starting in plus territory and a 6.1 PPP if starting +20 or better [theoretical]. On the flip side your defense PPP is 4.8 when starting on our end and up to 5.3 when the opp starts on the +20 or better. The % difference in these is 18.6% on a punt and 13.1% on a go for and not make it. If your 4th and 2 make % is 40% then the odds favor you going for it based on the equity of you making compared to the equity you give up in PPP % compared to not making it on 4th down or the equity gained in punting the ball 18.6%. The data says its a losing proposition if you punt it there. Not saying I agree or disagree because being brought up in the 70s where field position etc were etched into my brain. However, this is the same line of thinking Belichick used in his infamous 4th and 2 call on -28 vs. the Colts and Manning a few years back on SNF or MNF. Better the opp offense then the more aggressive you should be based on the PPP stat on 4th down. It has really caused me to really think hard on what might be "automatic" punts in situations. If you are a numbers guy I suggest you read the book mentioned. Also says that teams do not go for 2 enough or onside kick enough. Here is the one thing that seems to get lost when topics like this come up: those type of statistics assume a never ending string of opportunities for which to apply the optimum strategy. That is not the case though, and I am not sure that aggregating the data gives you the best basis from which to derive a plan. You bring up poker, so I would counter with the reality that tournament poker decisions can be different than cash game decisions even when the hands are the same. I do agree though, that there is a great deal of benefit from now thinking through previously "automatic" decisions.
|
|
scottc
Sophomore Member
Posts: 149
|
Post by scottc on Nov 17, 2015 9:02:39 GMT -6
That sounds more like Expected Points, which is extremely useful, but can't really be divined for high school. They get into expected points too by field position plus they break it down by PPP based on field position start as well...
|
|
scottc
Sophomore Member
Posts: 149
|
Post by scottc on Nov 17, 2015 9:17:34 GMT -6
A book called "Saturday Morning Wakeup Call" (Warning: heavy math and statistics formulas) Really gets into this theory of these types of stats at the high school level. The Points Per Possession (PPP) stat really is one that should be used by the following criteria according to this book and my understanding: The opponents PPP is a stat used to determine if you should go for it in a certain situation based on field position you are risking compared to the PPP differential by punting and the field position gained. If your punter nets 30 or less many times you are better off [mathematically speaking] going for it than punting because of the PPP equity gain is not greater than your make % on 4th down. similar to poker in # of outs compared to the pot odds you get if you are chasing a flush or straight draw etc. example: assuming you have an average defense (all numbers are made up to show the author's thinking) You are facing 4th and 2 at the -18. Your avg net punt is say 30 yards. your opponent has a PPP of 5.9 [theoretical] when starting in plus territory and a 6.1 PPP if starting +20 or better [theoretical]. On the flip side your defense PPP is 4.8 when starting on our end and up to 5.3 when the opp starts on the +20 or better. The % difference in these is 18.6% on a punt and 13.1% on a go for and not make it. If your 4th and 2 make % is 40% then the odds favor you going for it based on the equity of you making compared to the equity you give up in PPP % compared to not making it on 4th down or the equity gained in punting the ball 18.6%. The data says its a losing proposition if you punt it there. Not saying I agree or disagree because being brought up in the 70s where field position etc were etched into my brain. However, this is the same line of thinking Belichick used in his infamous 4th and 2 call on -28 vs. the Colts and Manning a few years back on SNF or MNF. Better the opp offense then the more aggressive you should be based on the PPP stat on 4th down. It has really caused me to really think hard on what might be "automatic" punts in situations. If you are a numbers guy I suggest you read the book mentioned. Also says that teams do not go for 2 enough or onside kick enough. Here is the one thing that seems to get lost when topics like this come up: those type of statistics assume a never ending string of opportunities for which to apply the optimum strategy. That is not the case though, and I am not sure that aggregating the data gives you the best basis from which to derive a plan. You bring up poker, so I would counter with the reality that tournament poker decisions can be different than cash game decisions even when the hands are the same. I do agree though, that there is a great deal of benefit from now thinking through previously "automatic" decisions. You make a great point here about long term. Which is a cash poker theory--making the right decision over and over in the long term eventually will be profitable. A single game may be more like a tournament now that I think about it. At that point it depends on the situation in the game i.e early 1st qtr late 4th qtr etc. plus the quality of opponent etc.
|
|
pistola
Sophomore Member
Posts: 193
|
Post by pistola on Nov 17, 2015 11:34:30 GMT -6
You guys are right, it does not matter where you coach. Good football is good football, and Texas does not have a monopoly on it. I apologize for that comment, I was wrong. We will just have to agree to disagree here on the statistical stuff. I do think agree that winning the kicking game is vitally important. No, the stat stuff really isn't that important, for the very reasons explained in this thread. They simply show you HOW you did...but we already have that metric in place. Its called the scoreboard. Not that important TO YOU.. its all about preference. you can say stats and stuff aren't important and they just showed you HOW you did it, but some people thats important to. it helps them to visually see the numbers
|
|
|
Post by rwb32497 on Nov 18, 2015 7:43:05 GMT -6
I did the Points Per game breakdown for our last two seasons following our last regular season game. In 2014 our average was 2.33. In 2015 our average was 2.03. I feel like our average would be better this season if it wasn't for special teams play. We have given the opponent the ball on our 40 after punts and kickoffs 13 times this year. We gave up 6 TD's and one FG after in those 13 times. Overall points this season, we have given up 68 less points compared to last season.
|
|
|
Post by silkyice on Nov 18, 2015 9:01:18 GMT -6
No, the stat stuff really isn't that important, for the very reasons explained in this thread. They simply show you HOW you did...but we already have that metric in place. Its called the scoreboard. Not that important TO YOU.. its all about preference. you can say stats and stuff aren't important and they just showed you HOW you did it, but some people thats important to. it helps them to visually see the numbers I think coachd5085 means that they aren't that important because they do CHANGE anything he does. In baseball if it you can show my that our winning percentage (or on base percentage or runs per inning or whatever) goes up if we wait to swing until we already have a strike, then that is valuable information. But if I just know that one game my points per play was 0.85 and another it was 0.45, I am not sure how that helps me in football. One more example, let's say my most productive play is draw. Does that mean I should run it more? Maybe. Maybe the reason it is productive is because I am only calling it on good downs against teams that are rushing the passer. Or I ran it a bunch against our worst opponent. It would not be smart to run it against a team that sits on the line and doesn't rush the passer no matter what the "stats" say.
|
|
|
Post by Chris Clement on Nov 18, 2015 22:07:00 GMT -6
Understanding the difference between descriptive and predictive statistics is essential to understanding when to use which numbers. PPP is useful in evaluating performances, it's largely descriptive, and it's good to help contextualize performances under different circumstances. If you score 18 points per game and allow 12, and an upcoming opponent allows 35 and scores 50, this allows me to get some handle on which team is "better." If I choose to go no huddle this year I can see if my offense was actually more effective.
As regards the idea that you should run more of the plays that gain the most, one must consider how this numbers were gotten, but if you have a bunch of five step plays and one gains way more (assuming a useful number of trials), there's a very good chance that you should be calling that play more often.
|
|
|
Post by silkyice on Nov 23, 2015 12:10:05 GMT -6
Understanding the difference between descriptive and predictive statistics is essential to understanding when to use which numbers. Boom
|
|