Gil you don't suck at writing at all! You cover every angle possible! And thoroughly explain what's on your mind. I just can't keep up, myself. It's a lot to take in. But for those with a brain that is larger than a thumbtack like mine, you offer lots of information to process!

Thanks for explaining, makes perfect sense.

So are turnovers occurring 100% of time by the player chosen to shoot and never really passing turnovers? And its completely window dressing when it says bad pass out of bounds or even loses the ball in the back court (because you're not shooting from the backcourt)? Same thing for "passing out of double team"?

Where does passing impact if a turnover occurs? Based upon what you said, passing has no impact on whether a turnover occurs. So a guy with 20 passing will have the same number of turnovers as a guy with 100 passing with all other things remaining equal?

Posted by WardoYT on 1/30/2020 8:03:00 AM (view original):

Posted by Benis on 1/30/2020 6:59:00 AM (view original):
Thanks for explaining, makes perfect sense.

So are turnovers occurring 100% of time by the player chosen to shoot and never really passing turnovers? And its completely window dressing when it says bad pass out of bounds or even loses the ball in the back court (because you're not shooting from the backcourt)? Same thing for "passing out of double team"?

Where does passing impact if a turnover occurs? Based upon what you said, passing has no impact on whether a turnover occurs. So a guy with 20 passing will have the same number of turnovers as a guy with 100 passing with all other things remaining equal?

The way I got it is everything is a dice roll which multiplies by 5 a turnover equals 25% which makes the player invisible to the odds we do not have the perfect engine and never will be. I guess that's why I am always going deep on rotation and no hidden tricks.

Good lord you are a worthless human being

"So are turnovers occurring 100% of time by the player chosen to shoot and never really passing turnovers? And its completely window dressing when it says bad pass out of bounds or even loses the ball in the back court (because you're not shooting from the backcourt)? Same thing for "passing out of double team"?

Where does passing impact if a turnover occurs? Based upon what you said, passing has no impact on whether a turnover occurs. So a guy with 20 passing will have the same number of turnovers as a guy with 100 passing with all other things remaining equal?"


1. Before the possession starts there is a decision that is something like "Is there a turnover before the shot" this is why the press is effective.
2. Passing impacts the Shot-Foul-Turnover decision, the sim decides WHO is going to have to make that decision and then WHAT decision happens. The higher the pass (individual and team) the lower the likelihood of a TO.

Does that help BEnis?

Posted by Trentonjoe on 1/29/2020 8:57:00 PM (view original):

Posted by shoe3 on 1/29/2020 6:23:00 PM (view original):
I’m not sure I understand what gil is saying here. I go back and forth on two interpretations, one I mostly agree with, and the other I don’t.

1) Assists are correlated with things like good passing (especially guards), but are not themselves the source of “good open looks”. Assist totals and FG% are affected by passing; passing, especially in guards, can increase FG%, reduce turnovers, and thereby increase assists. Agree. I think this is probably what he’s getting at, but that just might be because I agree with it.

2) Assists have no relationship to the outcome of each possession, ie in line with what dogg seems to say above, you could take away all the assists and the outcomes of each possession would remain the same. I don’t agree.

I agree with 1 and 2.

Schrödinger’s dish.

i just bumped a thread from a long time ago that has my (maybe right) theory on how pass is used. It's titled: How each possession is determined.

Yeah, I remember that conversation from when you bumped it 3 years ago. I think basically we’re on the same page, with each other and with gil. I think assist totals are an indicator of efficiency. I think, as in real life, it’s kind of nonsense to try to think about an assist leading to a field goal. The field goal is determined, then the assist is recognized later, if applicable; but the assist requires the field goal, not the other way around. IOW the assist doesn’t cause the field goal; rather, a field goal allows for the possibility of the recognition of an assist.

I do think the assist totals and field goal percentage have a relationship, which is why I don’t agree with prop 2 above. I think the relationship is indirect, and not causal; but they’re related, in that they are both efficiency indicators stemming from some of the same core, causal factors (ie, passing and other attributes, etc).

i think we are all fairly closer to the same page, two things to add -

1) passing definitely impacts turnovers. my take is that during the TSF decision, 'the player who will take the shot if it gets that far' has already been selected, but i do not think the turnover, if it happens, is necessarily his. i think the turnover in the TSF decision can be any offensive player on the court, just weighted more heavily to the guy who has been selected.

roughly, what i think the game does is it calculates a 'passive turnover rate' for a player. this isn't constant, it is probably calculated on every offensive possession, and would include things like the offensive players' bh/pass/iq, the defensive set and quality of the defense, all that jazz. then, there is also an 'active turnover rate' for the player in question. i think basically these are all added together and that is the odds of the T in the TSF decision (turnover). if it happens, a player is then selected to get the turnover (hopefully in a way such that the player's contribution to the team's turnover rate is exactly proportional to the way turnovers are doled out to individuals, or else as close as reasonably possible, which may or may not be the case - if this bit doesn't make sense, just ignore it, its a possibly meaningless side note that relates to my prior comment, possibly in a different thread, about how i pretty much totally trust team wide figures like rebounding, steals, turnovers - but not necessarily individual figures for those same stats - with rebounding as the primary example for suspicion).

2) there is still a little daylight between shoe and myself - this isn't intended to resolve that daylight, but just to clarify the difference. i'm not making the argument about assists in real life - yes, its sort of arbitrary, the real life line of where an assist is counted and where it isn't. but in real life, the players actually pass the ball. assists came into being to try to capture a real basketball concept, where the actions of 1 player substantially contributed to the success of the offense on that possession.

in HD, the opposite is true. players don't actually pass. there is no concept like that. a player just magically has the ball N number of feet from the basket. the entire game could get sim'd, and at the very end, assists could be calculated based on player ratings - and you could get the *exact* same numbers you get today. what im saying is, real life assists, even though somewhat fuzzy, are based (imperfectly) on something that actually happened. in HD, that is not true, there is no passing, no way to do a real assist. the only way to get an assist is that the game creator thought it would look strange not to have assists, so he made some **** up that looks half way decent. the textbook definition of window dressing ('type 2' - my made up term to indicate its window dressing where its not just purely random garbage, but is done decently well, correlated to something real, so it looks half way real. when done right, this actually tricks real people into thinking its not pure window dressing, even though it is - so by my assessment, the HD creator did a good job! but, its still window dressing, its still meaningless - there is no underlying action that resembles passing in this game, whereby one can point to anything resembling an 'action' upon which to assign the assist).

"1) passing definitely impacts turnovers. my take is that during the TSF decision, 'the player who will take the shot if it gets that far' has already been selected, but i do not think the turnover, if it happens, is necessarily his. i think the turnover in the TSF decision can be any offensive player on the court, just weighted more heavily to the guy who has been selected.

roughly, what i think the game does is it calculates a 'passive turnover rate' for a player. this isn't constant, it is probably calculated on every offensive possession, and would include things like the offensive players' bh/pass/iq, the defensive set and quality of the defense, all that jazz. then, there is also an 'active turnover rate' for the player in question. i think basically these are all added together and that is the odds of the T in the TSF decision (turnover). if it happens, a player is then selected to get the turnover (hopefully in a way such that the player's contribution to the team's turnover rate is exactly proportional to the way turnovers are doled out to individuals, or else as close as reasonably possible, which may or may not be the case - if this bit doesn't make sense, just ignore it, its a possibly meaningless side note that relates to my prior comment, possibly in a different thread, about how i pretty much totally trust team wide figures like rebounding, steals, turnovers - but not necessarily individual figures for those same stats - with rebounding as the primary example for suspicion)."


to sum up BillyG, and correct me if I am wrong:

Team passing (weighted by position) effects lots of stuff including:

If the shot goes in
If there is a TO due to the press
If an individual player turns the ball over

yeah, that is pretty fair. i don't know i'd call the 2nd and 3rd thing so different i guess, i think all that happens in that turnover decision.

basically, pass impacted turnovers - end of story - for the first ~10 years of HD. this distinction only matters (about then vs now) because we are having a conversation about how the HD engine work, structurally - and in short, the HD engine is what it was 15+ years ago, with minor tweaks (no structural changes). so back then, when assists were created, there was no passing impact on shooting at all. but yeah, i do think there are multiple impacts of passing on turnovers, if you want to break it down like that - i would think a player's passing factors in to multiple turnover equations, if you will. at minimum, it believe passing factors into a player's rate of turnovers when he has the ball, and when someone else on the team does. maybe press is outside that too - i think some form of fastbreak is supposed to be outside that, but i wasn't thinking press - however, i'm definitely losing grasp on the details about this stuff. one important distinction i'll make - still to this day, the passing rating of one player, does NOT impact the turnover rate of another player. those are supposed to be independent (at least on a direct basis).

supposedly, before seble adding passing + iq of team mates to the scoring decision logic - there was *no* impact of one player's ratings on the stats of another player. now, i think he was giving a direct answer, because there are numerous cases where this is not true, but all of the cases i know about are indirect complexities, and not an 'on its face' type deal - so i mostly believe him. i think he gave an 'on its face' sort of answer, ignoring things like how stamina of players impacts each other, how individual rebounding numbers of players are kinda wonky (a whole different conversation best saved for another time) and there are obvious indirect effects between players there (from the user standpoint), and probably how offenses and defenses adjust to each other in this game, which i assume nobody who plays or work at HD could explain worth a damn. my intuition says there are indirect impacts there but i definitely can't say that with too much confidence. anyway, i think seble was saying, if i go and look at the equations for this stuff, its all about the player in question, not about his team mates - and i pretty much buy that. i just think at times, the system is more complex than that.

I am reasonably sure there is a “does the offense start “ decision for press teams.

Posted by Trentonjoe on 2/1/2020 6:40:00 PM (view original):
I am reasonably sure there is a “does the offense start “ decision for press teams.

very possible, i assume you are correct - i'm sure that article can still be found somewhere, that explains the sim league basketball engine or whatever they call it now. i am often surprised just how extensively the details of this game have melted away after just 3 years without playing, i literally have no idea if what you are talking about exists or not.

3/24/2018 12:04 PM	Trentonjoe
I have a questions about the FCP. 1. In the possession event tree, I am assuming the first decision is "is there a shot, foul, or turnover?" Does FCP increase the chance of a turnover? 2. Or does the FCP add an extra decision? Something like "does the offense get the ball past half court?" 3. Lastly, when calculating if there is a turnover, is it an individual matchup or like pg vs pg? or is a team equation like offenses spd,bh,pass iq(etc) vs defenses speed, defense, iq (etc)?
3/24/2018 4:11 PM	Customer Support
Chris, 1. It's a little more complicated than that. Each possession is broken down into multiple stages. During each stage, the end result can be shot/foul/turnover, or it can just advance to the next stage. The press does increase the turnover rate at each stage. 2. Yes, this is closer to how it works, like how I described the multiple stages of a possession. 3. The calculation depends on the stage of the possession. If it's still in the backcourt, then it's more of a team calculation, but weighted toward the guards, who would be predominantly handling the ball. If it's a later stage, then it becomes more of a man on man calculation, depending on other factors.