You know the exact formula for RPI?  Wow, I didnt think anyone knew that.  I am impressed.

Posted by sully712 on 9/10/2010 3:04:00 PM (view original):
You know the exact formula for RPI?  Wow, I didnt think anyone knew that.  I am impressed.

serious question
is the RPI formula on here different than what it is in real life?

Posted by _hannibal_ on 9/10/2010 8:44:00 AM (view original):
Billy, I'm not trying to be ridiculous; I tried to include every case so that you (or whomever) would be able to pick one that makes the strongest argument.  I'm not expecting it to affect every situation.

Let me pick at your first example.  First, A does not know that the prestiges are identical.  Since one could be a high B- and the other a low B-, the prestige factor introduces at least a +-30% to the calculation.  Ignoring that for the moment, secondly, A does not know how much carry over money B has.  If B has $2000 more carry over than A in your scenario then B has the advantage.  Or if B spent $2000 less on FSS.  Finally, a very minor point, you have the example backwards, you were supposed to have more battles in the real HD world, not less.

the uncertainty of the prestige factor can be dealt with simply. player A assumes the prestige is identical. this is an equally reasonable assumption in both cases and yields the same result.

the uncertainty of scouting, carry over, etc are all similar, if player A makes an assumption on them, an attempt to quantify, and those assumptions are the same in both cases (why wouldn't they be?), then the same results hold.

you could make the same kind of arguments for any of the uncertainties. say prestige was certain, and HV:CV was not. then, when a player attempts to quantify HV:CV, either he has the uncertainty of prestige or he doesn't - clearly with that single difference you can come up with a million examples where the guy might of battled and might not of.

a better example would have the coach feeling like it was a close battle, and not knowing who would win. because of this uncertainty, and because he likes the player, he goes for it. if you take away the uncertainties, then half the time (roughly) he knows he is losing, and wouldn't battle, so he would have half the chance of battling in that case as he would with the uncertainty.

Posted by wfudd on 9/10/2010 3:17:00 PM (view original):

Posted by sully712 on 9/10/2010 3:04:00 PM (view original):
You know the exact formula for RPI?  Wow, I didnt think anyone knew that.  I am impressed.

serious question
is the RPI formula on here different than what it is in real life?

its the same. it been posted on these forums a hundred times too.

 - You haven't missed it, I don't claim to have done so.  I'm not trying to convince anyone I'm trying to understand the opposite viewpoint.

But you are ignoring it when people tell it to you.  I've said (and numerous people have echoed it) that there is already too much known information in recruiting and I don't think that they should give us any more than they already do.  I think it will lead to fewer battles and make recruiting even more formulaic than it already is.  You come back and say you don't think so.  Well, if we're not going to change your mind and you're not trying to change anyone's - then this has gone about 5 pages too long.

I'd rather they introduce more variables, like they do with HBD free agency - they already have preferences based on distance from home and favorite school - they could add desire for immediate playing time/immediate starts, preference for potentially getting into the postseason, preference for playing at a school with a big football tradition or a school where the basketball players are the BMOC, preference for an academically prestigous university, etc.

Posted by coach_billyg on 9/10/2010 3:33:00 PM (view original):

Posted by wfudd on 9/10/2010 3:17:00 PM (view original):

Posted by sully712 on 9/10/2010 3:04:00 PM (view original):
You know the exact formula for RPI?  Wow, I didnt think anyone knew that.  I am impressed.

serious question
is the RPI formula on here different than what it is in real life?

its the same. it been posted on these forums a hundred times too.

I was just trying to make a point that even knowing that formula does not really help out.  Its not like you can know what your opponents are going to do when you schedule games, etc...

 - Recruiting money per scholarship, per postseason game, recruiting costs for various distances, costs per recruits for FSS, RPI.  There are probably more.

Items 1-3 there aren't ration driven, they are set amounts.  Your budget is a formula you know and you can figure the cost per recruit with simple division. 

Being able to specifically decide that HVs are more effect to 220 miles (for example) and CVs are the way to go beyond that is more like revealing that a 80 ATH/70 REB player will, other things being equal, be a superior rebounder than a 40 ATH/95 REB player. 

"Recruiting money per scholarship, per postseason game, recruiting costs for various distances, costs per recruits for FSS, RPI.  There are probably more."

Really? These things are not formulas that anybody who knows simple math wouldn't be able to figure out. You have 3 open scholarships at D3 you had $0 carryover and now you have 9k to recruit, how much money do you get per recruit? Same thing could be down with recruiting money, so instead of 9k you get 11k, and your conference played 8 NT games and 0 PI games, clearly you would get 3k extra per game, and split 24k up among the 12 teams. The costs for distance is right there, the actual formula for determining those costs are not known we just know the prices because we absolutely HAVE to know the prices or else recruiting would be impossible. Same goes for cost per recruit, you look at the total recruits divide it by the price and you find out the cost per recruit. The RPI is also a slightly tweaked version of the real life RPI formula that's common knowledge. RPI isn't an HD only thing.

The formulas I'm talking about are the real formulas that are game exclusive and that aren't 100% necessary for us to know, like the formula used for rankings the relationship each rating has in production, prestige values and how much value each recruiting technique has. These things are unique to HD and we don't NEED to know them exactly to do well and these are things that with some effort we can come close to figuring out.

> the uncertainty of the prestige factor can be dealt with simply. player A assumes the prestige is identical. this is an equally reasonable assumption in both cases and yields the same result.

Billy I think you are missing the point here.  You can't assume the prestiges are identical -- you do not have that information.  Or are you proposing that we assume that coach A acts in a less than optimal fashion?  If that's the case then you can conclude anything.  If you assume the actors are rational, then the addition of the certainty of the CV:HV ratio is not dispositive because of the uncertainty of the other factors.

> I think it will lead to fewer battles and make recruiting even more formulaic than it already is.  You come back and say you don't think so.  Well, if we're not going to change your mind and you're not trying to change anyone's - then this has gone about 5 pages too long.

Nobody's forcing you to participate.  I've tried to get clarification on the reasoning behind why people think it will decrease battles.

> I'd rather they introduce more variables, like they do with HBD free agency - they already have preferences based on distance from home and favorite school - they could add desire for immediate playing time/immediate starts, preference for potentially getting into the postseason, preference for playing at a school with a big football tradition or a school where the basketball players are the BMOC, preference for an academically prestigous university, etc.

I think that sounds great.  I wouldn't mind if they even made the relative worth of HVs and CVs vary by recruit.

> The costs for distance is right there, the actual formula for determining those costs are not known we just know the prices because we absolutely HAVE to know the prices or else recruiting would be impossible.

Actually you don't.  They could make recruiting one giant black box if they wanted to: not tell you how much money you have, how much anything costs and just tell you if you have enough money to do what you are asking to do.

Also the recruit money formula isn't as straightforward to crack as you assume, especially if you are not given the criteria beforehand.  (Nobody ends up with exactly $9k in D3 because everyone gets some postseason money and a little FSS money too.)

> The formulas I'm talking about are the real formulas that are game exclusive and that aren't 100% necessary for us to know, like the formula used for rankings the relationship each rating has in production, prestige values and how much value each recruiting technique has. These things are unique to HD and we don't NEED to know them exactly to do well and these are things that with some effort we can come close to figuring out.

Well I'm sorry I wasn't psychic enough to pick one of the formulas you were thinking of, which coincidentally are the same as the set of formulas which would not satisfy your question.

I'd actually like more unknowns - like I think it would be great if all player ratings that we see in recruiting were in fact just within 5 plus or minus of the actual rating - the good looks like a 70 but when he turns up he could be anywhere from 65 to 75

horse, dead, kick it again.

hannibal, your responses are getting more outlandish. Did you really just type:

> The costs for distance is right there, the actual formula for determining those costs are not known we just know the prices because we absolutely HAVE to know the prices or else recruiting would be impossible.

hannibal: "Actually you don't.  They could make recruiting one giant black box if they wanted to: not tell you how much money you have, how much anything costs and just tell you if you have enough money to do what you are asking to do."

 

You can't be serious, can you? This would basically end recruiting for all intents and purposes. You would have no way to strategize, no way to budget, no way to plan. No way to day anything, basically. Just totally ludicrous. You yourself admit you can't come up with a legitimate reason to reveal what you want to reveal, and these are your arguments/justifications in response to what everyone else is saying? 

Posted by sully712 on 9/10/2010 3:38:00 PM (view original):

Posted by coach_billyg on 9/10/2010 3:33:00 PM (view original):

Posted by wfudd on 9/10/2010 3:17:00 PM (view original):

Posted by sully712 on 9/10/2010 3:04:00 PM (view original):
You know the exact formula for RPI?  Wow, I didnt think anyone knew that.  I am impressed.

serious question
is the RPI formula on here different than what it is in real life?

its the same. it been posted on these forums a hundred times too.

I was just trying to make a point that even knowing that formula does not really help out.  Its not like you can know what your opponents are going to do when you schedule games, etc...

i thought you were being sarcastic, actually.

also, knowing the rpi formula helped me considerably, in coming up with my strategy for scheduling. so i would disagree there. you can definitely anticipate what your opponents are going to do when you schedule, that is really the point of scheduling, isn't it?

Posted by _hannibal_ on 9/10/2010 4:30:00 PM (view original):
> the uncertainty of the prestige factor can be dealt with simply. player A assumes the prestige is identical. this is an equally reasonable assumption in both cases and yields the same result.

Billy I think you are missing the point here.  You can't assume the prestiges are identical -- you do not have that information.  Or are you proposing that we assume that coach A acts in a less than optimal fashion?  If that's the case then you can conclude anything.  If you assume the actors are rational, then the addition of the certainty of the CV:HV ratio is not dispositive because of the uncertainty of the other factors.

> I think it will lead to fewer battles and make recruiting even more formulaic than it already is.  You come back and say you don't think so.  Well, if we're not going to change your mind and you're not trying to change anyone's - then this has gone about 5 pages too long.

Nobody's forcing you to participate.  I've tried to get clarification on the reasoning behind why people think it will decrease battles.

> I'd rather they introduce more variables, like they do with HBD free agency - they already have preferences based on distance from home and favorite school - they could add desire for immediate playing time/immediate starts, preference for potentially getting into the postseason, preference for playing at a school with a big football tradition or a school where the basketball players are the BMOC, preference for an academically prestigous university, etc.

I think that sounds great.  I wouldn't mind if they even made the relative worth of HVs and CVs vary by recruit.

> The costs for distance is right there, the actual formula for determining those costs are not known we just know the prices because we absolutely HAVE to know the prices or else recruiting would be impossible.

Actually you don't.  They could make recruiting one giant black box if they wanted to: not tell you how much money you have, how much anything costs and just tell you if you have enough money to do what you are asking to do.

Also the recruit money formula isn't as straightforward to crack as you assume, especially if you are not given the criteria beforehand.  (Nobody ends up with exactly $9k in D3 because everyone gets some postseason money and a little FSS money too.)

> The formulas I'm talking about are the real formulas that are game exclusive and that aren't 100% necessary for us to know, like the formula used for rankings the relationship each rating has in production, prestige values and how much value each recruiting technique has. These things are unique to HD and we don't NEED to know them exactly to do well and these are things that with some effort we can come close to figuring out.

Well I'm sorry I wasn't psychic enough to pick one of the formulas you were thinking of, which coincidentally are the same as the set of formulas which would not satisfy your question.

hannibal, if you aren't figuring coaches act in a less than optimal fashion, then you haven't been playing the same game i have. and besides, with imperfect information, you have to make some sort of judgement on that unknown and go from there. it is completely reasonable, and completely rational, for a coach to assume the prestige is the same when the letter is the same, for planning purposes. im not really sure what you are getting it. its no more complicated than this - uncertainty drives battles. the type of uncertainty is irrelevant. if you took away all uncertainty, you would know for sure if you would win a battle or not ahead of time, and would only enter battles you would win - and they would always be uncontested, because the other guy would know he would lose, and would spend his money elsewhere.

really, until you are willing to put out any argument for why uncertainty does not increase battles over the no uncertainty case, there is no point of continuing.

I am going to go in a slightly different direction here and go back to the ping pong balls.  If the argument is that it is 95% predictable now, and releasing these formulas makes it 98% predictable, and that more predictability is bad, wouldn't that seem to imply that more unpredictability was good?  Wouldn't the ping pong balls add uncertainty, even if it was only another 5 or 6% wouldn't that be a good thing?  Or is the argument that 95% is exactly the right amount of uncertainty?  It would certainly stimulate battles, every top 25 player would have 5 teams on them.  It might even be argued that this would open recruiting into something resembling the ability for a mid major or even a local minor to win a real recruit once in a while and it definitley would result in some of the distance advantage going away.  I mean, all of a sudden, UCLA would actually have a shot, a significant shot at getting a recruit from New York or North Carolina.  Duke would have a shot at one in Kansas or the 'Cuse would have a shot at that great ballhandler out of Texas.  And yes, Florida International might actually get one, yes one, player that might actually get them out of the cellar of their small time league and get them to the dance where they will be a low seed and still probably get taken out in the first or second round.  But they would have a chance, however small.  Teams like Idaho and Hawaii would not be sooo difficult to recruit for and teams like Duke or Pittsburgh would not be quite as easy.

Now with the disclaimer that I do not think that I am a great or even good recruiter, I actually hate recruiting in this game, I think it is among the worst aspects of this game.  It is so important that coaching comes in a way, way way distant 12th out of two items.  It is extremely predictable and truly it is either ridiculously easy or ridiculously hard.  I would wager that there is more than a 99% predictability in it now and that one percent has nothing to do with the formulas but with timing and intestinal fortitude (read guts and bluff).  In fact, I think it is such an important part that it should really be called recruiting dynasty.  D3 recruiting is all about random chance but also about school location and prestige.  Advantage to vets, whether known formulas or not.  D2 recruiting is all about formulaic recruiting.  By the time you are in D2, you either have learned most of that or you got lucky and moved up quickly.  Pretty danged equal recruiting chances if you stay at least 4 or 5 years although with the wrong team, it can be very discouraging if it is your first one.  D1, well some of you know my opinion about D1 but it pretty much sucks.  It will always suck because of the way it is designed and maintained.  Veteran coaches in general want it to remain the same for various reasons and I do not fault them for it as they worked their way up through the system.  However, you do not see a lot of new ones working up through it because it is so difficult to move from low D1 to mid D1 to High D1 and you cannot move your team or conference up to the elite (although you can get close as Dalt did)  but the game is designed to never allow this.  This means that at D1, recruiting is always rigged to the longer term vets and that is why there is so much predictability - they have the teams with the prestige to win, period.  That one thing overrides all of the other things and basically makes all other questions pretty much moot.  If any team in RL had done what Dalt did with his team, they would have been elite for now and a long long time after.   But, it did not work that way and it will not because there is opposition to that from veteran coaches (as well as support for it).

Even with that said, I do not see any reason to release the exact CV/HV information nor do I see any reason not to.  It is the way it is and I think this is one of those 6 of one vs a half dozen arguments that does not really have a correct answer which is why it is taking so long to resolve it.