|
|
Post by 01aztecgrad on Sept 3, 2012 12:22:57 GMT -8
It seems like many people have just accepted the claim that it's reasonable to assume that going for 2 points is a 50/50 proposition. It's simple enough to do a quick search of NCAA statistics to demonstrate that is a false assumption. Records are available for every year back to 1958, and never has a season had close to a 50% conversion rate. NCAA records2 points success 2007: 42.1% 2008: 37.0% 2009: 40.6% 2010: 40.0% 2011: 43.5% 1 point success 2007: 96.4% 2008: 96.4% 2009: 95.8% 2010: 96.4% 2011: 96.2% So the facts don't support the claim that going for two is the optimal strategy. General population statistics aren't really useful for a specific instance when deciding the best strategy, but they do show that unless you have a reason to believe you have a significant talent advantage over the other team which would lead you to believe your odds are well above average, or you are in a time/point situation that requires going for two, that going for 1 is the best choice. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 3, 2012 12:27:28 GMT -8
Excellent post.
As I mentioned before, even during a season when Abel Perez could barely kick a FG to save his life, he still converted 47 of 49 PATs, or was successful 95.9% of the time.
|
|
|
|
Post by uncledougy on Sept 3, 2012 12:27:41 GMT -8
Exactly!.. not to mention the concept of making 50% gets you the same amount of points doesn't take into consideration game situations. Often times not making a 2pt conversion will hurt you much more than the benefit of making a 2pt conversion. The other team can plan around the 2pt coversion, it just doesn't make sense in anyway, (unless you don't have a kicker!) I am sure we have a disgruntled soccer player somewhere that can put a ball through the uprights....
|
|
|
|
Post by FULL_MONTY on Sept 3, 2012 12:30:42 GMT -8
Going for 2 should be situational not the mo.
Weight your numbers out, a 2 try yields .81 points per attempt versus .96 for the 1 point try.
So for the numbers to work, you have to either believe that you cab be successful 48% of the time or that your kicker is only good for 8 out of 10.
put in 43% when should have been 48%
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 3, 2012 12:39:34 GMT -8
Going for 2 should be situational not the mo. Weight your numbers out, a 2 try yields .81 points per attempt versus .96 for the 1 point try. So for the numbers to work, you have to either believe that you cab be successful 43% of the time or that your kicker is only good for 8 out of 10. Prediction: Zeigler or somebody else at the U-T will figure out a way to parrot that info without attribution. |
|
|
|
Post by laaztec on Sept 3, 2012 12:58:04 GMT -8
Rocky has run the numbers with a Mathematician from SDSU. He has a chart with the odds on it. If SDSU would have made both 2 point conversions everyone on the board would be calling him a genius.
People need to step back from the ledge and let the season play out.
|
|
|
|
Post by aztecjake on Sept 3, 2012 13:09:15 GMT -8
Rocky has run the numbers with a Mathematician from SDSU. He has a chart with the odds on it. If SDSU would have made both 2 point conversions everyone on the board would be calling him a genius. People need to step back from the ledge and let the season play out. I agree, with a small disclaimer. The time is going to come when the place kicker is desperately needed/required and experienced kickers will be essential. The time to gain this experience is during OOC games. |
|
|
|
Post by sdsu1975 on Sept 3, 2012 13:14:56 GMT -8
Rocky has run the numbers with a Mathematician from SDSU. He has a chart with the odds on it. If SDSU would have made both 2 point conversions everyone on the board would be calling him a genius. People need to step back from the ledge and let the season play out. We don't need to "step back." We need to get a f------ kicker!!!!! |
|
|
|
Post by uncledougy on Sept 3, 2012 13:16:45 GMT -8
Rocky has run the numbers with a Mathematician from SDSU. He has a chart with the odds on it. If SDSU would have made both 2 point conversions everyone on the board would be calling him a genius. People need to step back from the ledge and let the season play out. In no way would I refer to Coach's 2 point theory as a genius. I would say he was fortunate if he made both. Even if we make every 2 point try the rest of the season, it's still a bad idea. We'd be fortunate, but always a bad idea. |
|
|
|
Post by bearfoot on Sept 3, 2012 13:21:38 GMT -8
Rocky has run the numbers with a Mathematician from SDSU. He has a chart with the odds on it. If SDSU would have made both 2 point conversions everyone on the board would be calling him a genius. People need to step back from the ledge and let the season play out. You have heard the expression, "garbage in garbage out"? I do not mean that Rocky is garbage, I only mean that we do not know what information Rocky gave the math dude. The larger sample size the more accurate the results. I would have given the dude 2 point numbers for all going back 10 years. I would have asked the dude to do a separate chart adding in all the scoring attempts from 3 and in, since they are similar plays. |
|
|
|
Post by uncledougy on Sept 3, 2012 13:21:38 GMT -8
I am watching 2 high schools in Florida play right now and the high schooler just nutted a 32yd FG...amazes me we don't have a kicker that has some possibilty (above 80%) of converting a 30-39 yd FG. Not sure who was responsible for getting a kicker, but it was an epic failure in the recruiting process.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 3, 2012 13:24:23 GMT -8
Rocky has run the numbers with a Mathematician from SDSU. He has a chart with the odds on it. If SDSU would have made both 2 point conversions everyone on the board would be calling him a genius. People need to step back from the ledge and let the season play out. My dad, may he rest in peace, used to say that if you're doing something nobody else is doing, you want to make sure you have a damn good reason for doing so. I don't remember ever hearing of a HC who made strategy decisions based on data supplied by a professor on his campus. And I'm not on the ledge. As I've said, I'm optimistic that we'll beat Army by double digits. Something like 32-18. However, that doesn't mean I don't think going for two after every TD isn't dumb. |
|
|
|
Post by uncledougy on Sept 3, 2012 13:28:50 GMT -8
I am not calling for Rocky's head, but I am looking forward to a decent BE television contract where we can get a coach that can recruit and put us on the national map. Let's be honest, Rocky isn't going to coach the Aztecs forever, he will be gone someday, whenever that is....I have a hard time believing top recruits will line up to play "Rockyball".
|
|
|
|
Post by localsonly on Sept 3, 2012 14:01:48 GMT -8
Rocky has run the numbers with a Mathematician from SDSU. He has a chart with the odds on it. If SDSU would have made both 2 point conversions everyone on the board would be calling him a genius. People need to step back from the ledge and let the season play out. Fscinating response bordering on knee-jerk appologism. So Rocky has a "Mathermatician"? is that kinda like an edercated person who can read charts and graphs as the premier poster did for this discussion? Did Rocky's chart have different odds? Even had Rocky made both conversions, smart money would have said the odds are still the same, hugely in favor of kicking. Meanwhile, Rocky continues this sketchy notion about not kicking and insults his kickers draining whatever confidence they hardly had. He had better beat Army cuz they kick like a Mule... |
|
|
|
Post by 01aztecgrad on Sept 3, 2012 14:14:44 GMT -8
Rocky has run the numbers with a Mathematician from SDSU. He has a chart with the odds on it. If SDSU would have made both 2 point conversions everyone on the board would be calling him a genius. People need to step back from the ledge and let the season play out. This is the only reference I could find that mentioned what is on the chart: UT link"The chart was developed to help Aztecs’ head football coach Rocky Long calculate the probability of success the team might have if it kicks a field goal, punts, or goes for it on fourth down" Based on this description, I will bet that it's just a generic down/distance chart that isn't specific to the opponent or two point conversions. I don't have a problem with his decision to punt less, or even to try fewer field goals, but the decision to go for two is different. The opportunity cost of not converting on 4th down in an opponents end is usually just 20-30 yards of field position, which doesn't usually have a direct impact on the score of the game, but the cost of going for two is giving up an almost guaranteed point. Even bad kickers are almost automatic on extra points, so a down and distance chart is meaningless on a two point conversion because it wouldn't adequately account for the cost of failing. Aside from situations in which the score dictates going for two, the only other time it would make sense is when you are confident that your offense is superior to your opponents defense and will convert significantly more than 50% of your opportunities, and that you will score enough touchdowns to make it worthwhile. |
|
|
|
Post by Strawberry Puppy Kisses on Sept 3, 2012 14:31:26 GMT -8
I do a lot of business analytics and data mining for San Diego businesses.
I have strong knowledge of statistical theory, application, etc etc.
The data and application of formula to derive a chart illustrating situational probabilities is beginner-level statistics. Models to generate forecasts and other simulations are on the low-end of multivariate statistic complexity. It's not likely there is a "bad data" phenomenon here.
TLDR; We have no kicking game. I'm sure Rocky knows the range is .40 - .50 for 2pt conversion. He clearly feels this is a higher probability than shuffling out the PAT squad for 1pt.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 3, 2012 14:42:16 GMT -8
Mr. Puppy, thanks for the info.
If we have "no" kicking game, who's to blame?
|
|
|
|
Post by 01aztecgrad on Sept 3, 2012 15:18:30 GMT -8
I do a lot of business analytics and data mining for San Diego businesses. I have strong knowledge of statistical theory, application, etc etc. The data and application of formula to derive a chart illustrating situational probabilities is beginner-level statistics. Models to generate forecasts and other simulations are on the low-end of multivariate statistic complexity. It's not likely there is a "bad data" phenomenon here. TLDR; We have no kicking game. I'm sure Rocky knows the range is .40 - .50 for 2pt conversion. He clearly feels this is a higher probability than shuffling out the PAT squad for 1pt. The simple models you describe are useless in predicting the outcome of a specific event. It's like having a probability chart of the expected outcome of a coin flip, and thinking that means that if you flip a coin 9 times with 5 of them being heads, that the next toss is more likely to be tails. It doesn't work that way in random events, let alone for events that aren't random. If they go 0-2 in 2 pt conversions against Washington, 4-4 against North Dakota, and 0-2 against Nevada, would you really believe they would have a 50/50 chance of converting against Alabama because they were 50% for the season? Any model used to justify going for two would have to be opponent specific, and at that point you are introducing data problems, otherwise you could just build a model to predict the outcome of the game and make a killing in Vegas. |
|
|
|
Post by Strawberry Puppy Kisses on Sept 3, 2012 16:16:07 GMT -8
I do a lot of business analytics and data mining for San Diego businesses. I have strong knowledge of statistical theory, application, etc etc. The data and application of formula to derive a chart illustrating situational probabilities is beginner-level statistics. Models to generate forecasts and other simulations are on the low-end of multivariate statistic complexity. It's not likely there is a "bad data" phenomenon here. TLDR; We have no kicking game. I'm sure Rocky knows the range is .40 - .50 for 2pt conversion. He clearly feels this is a higher probability than shuffling out the PAT squad for 1pt. The simple models you describe are useless in predicting the outcome of a specific event. It's like having a probability chart of the expected outcome of a coin flip, and thinking that means that if you flip a coin 9 times with 5 of them being heads, that the next toss is more likely to be tails. It doesn't work that way in random events, let alone for events that aren't random. If they go 0-2 in 2 pt conversions against Washington, 4-4 against North Dakota, and 0-2 against Nevada, would you really believe they would have a 50/50 chance of converting against Alabama because they were 50% for the season? Any model used to justify going for two would have to be opponent specific, and at that point you are introducing data problems, otherwise you could just build a model to predict the outcome of the game and make a killing in Vegas. I didn't suggest any models, or even any particular avenue of statistical analysis. I simply stated that the topic isn't a mind-bender, as it related to the previous response(s). You seem confident in your knowledge; so, good for you. I'm quite confident in mine and am paid for it; so, good for me. That fact alone gives me enough satisfaction to refrain from commenting further on the inaccuracies in your post. Go Aztecs! |
|
|
|
Post by k5james on Sept 3, 2012 16:20:07 GMT -8
It's surreal that there's even a debate about this. Only in Aztecland... It wouldn't be the first time you've been out in left field on a conversation with respect to football... |
|
