General thoughts: scoring options

[WoVeU]

OK choose you are getting all up in my wheel house, weighted probabilities. (Engineering Major, Minored in ECO) I love these things, for simple math I have found nothing more valuable. But the value is found (much like differential equations, not so simple) in the boundary conditions...and of course defining the process and rigor!

Maybe after 9 hours of sleep in 2 days my brain is not engaged or I scrolled with preconceived expectations. I am missing something, or something is amiss (I think you were trying to short hand some English to not use a full page.)

Ignoring the clean fractional percentiles (that always give me pause) and the chunky increments and accepting the definitions of short, medium, and long...still...

TD with PAT kick:
short: (0.333)(0.95)*7 + (0.333)(0.05)*6= 2.31435 points
average: (0.25)(0.95)*7 + (0.25)(0.05)*6= 1.7375 points
long: (0.1)(0.95)*7 + (0.1)(0.05)*6= 0.695 points

In the first nomial I get the weighted prob. is that for a TD but why 7 if you have not treated the PAT as given, but the 2nd I expect to take care of the word "with" and addition...but how is it 6.

I expected to see for a short TD w/ PAT kick: (0.333)6...ok, 1/3 chance of braking the plain...yielding 2pts. But the PAT is a given not conditional now, so we add (0.95)1 = 0.95
thus...Short TD w/ PAT: (0.333)6 + (0.95)1 = 2.95pts {Definition: Weighted Probability (profit) leveraging <=10 yard position from goal ; Condition: Predetermined TD attempt and Predetermined PA kick.}
The problem here is the envelope of the method, 1st if it is predetermined that means a 4 down consumption will be used to secure it, this is fine if the (0.333) is given for the 4 down case (it can't reflect a 3 down consumption w/ the negotiable option to attempt the field goal.) And then the latter is fine for the PAT if we are saying we are taking the hard to miss 1 point.} And if we take care of that .333 for which case 3 or 4 downs the definition and conditions presupposes the General of the offense (HC or OC) is amassing these returns.

In the observational case you have to include outcomes to get really meaningful data. SO for at or inside the 10

[P% for using 3 downs*(P% for TD in 3 downs)*6pts] + [P% for PAT kick*(P% for making Kick)*1pt] + [100% - (P% for PAT kick)*(P% for 2pt Conv)*2pts] + P% for FG*3pts
+ [100%-(P% for using 3 downs)*(P% for TD in 3 downs)*6pts] + [P% for PAT kick*(P% for making Kick)*1pt] + [100% - P% for PAT kick*(P% for 2pt Conv)*2pts]

This is for the Definition: Total Scoring Probabilities ; Conditions: At or inside the 10 yard line

Admittedly this still does not take into consideration the differentiated losses. Not a huge deal here, TO on downs leaves the ball in good defensive position (in general not factoring in penalty possibilities, still a very real probability.) And does not account for an actual TO and the resulting probabilities for where the other offense will gain possession, and they could in fact score).

As said I may have missed the premise or at least part of it.

[Spence]

You are making something relatively simple - lots harder. No reason to over think it.

[donovan]

So I watched it and there was a link to this play......tough two points.

http://www.youtube.com/watch?v=nf2wgeIRDPo&NR=1

[WoVeU]

You are making something relatively simple - lots harder. No reason to over think it.

This is why I addressed choose.

[Spence]

You are making something relatively simple - lots harder. No reason to over think it.

This is why I addressed choose.

I know, so is he.

[WoVeU]

So I watched it and there was a link to this play......tough two points.

http://www.youtube.com/watch?v=nf2wgeIRDPo&NR=1

Cool. But it was a touch back...no points. The defense gained 20 yards of field position out of it.

[WoVeU]

Ok, you collected with the conditional given, duh.

And kept everything in singular approach...which I took to be the case.

[billybud]

You guys are leaving out the "situational factors"...

When deciding whether to go for a two point conversion, the time left in the game is an important factor to add to the equation. Also the relative score...your lead or how much you must catch up.

I have always like this site's analysis...I have linked their two point conversion chart.

http://www.footballcommentary.com/twoptchart.htm

[billybud]

Take a look at this model which describes odds for going for it on 4th and taking the 2 point conversion....

http://www.footballcommentary.com/blunders.htm

[silverfox]

OK

[WoVeU]

OK

Surprise, surprise...BSU was the most frequent 2-point leader last season:

http://ncaafootball.fanhouse.com/2010/1 ... t-edition/

also see link to http://cfbstats.com which was used as the reference for the PAT data

Before I even got down to this post I was considering team types and situations for going for 2. And Boise State popped in my head. If I were going to bet on any team converting the deuce, I would take Boise. I'd like to see them go for 2 every time. You have to cover the whole field against Boise, it is very hard to prevent them gaining 3 yards. Why is college football using the 3 yard line anyway. I like the 2 yard line. And I'd like to see the 1 point conversion done away with.

[WoVeU]

Did anybody read the portion in that article about "Unlucky U", N. Texas!!!
Just terrible events!

[RazorHawk]

OK

Surprise, surprise...BSU was the most frequent 2-point leader last season:

http://ncaafootball.fanhouse.com/2010/1 ... t-edition/

also see link to http://cfbstats.com which was used as the reference for the PAT data

You need to compute the numbers with the SOS factored.

[billybud]

Why not go for two when you are 40 point favorites? A missed point here and there won't be crucial.

I would expect more risk taking when the downside of failure is negligible.

[donovan]

Why not go for two when you are 40 point favorites? A missed point here and there won't be crucial.

I would expect more risk taking when the downside of failure is negligible.

Me too, which of course, makes it not risky at all.