players impacting team mates - passing raising fg% Topic

No, I can't do stuff like that. I just look for patterns by sense of smell.
2/13/2014 8:02 AM
Posted by artie40 on 2/12/2014 5:49:00 PM (view original):
I don't know how to value assists personally, but I have noticed that passing ratings do have an effect on A-to-TO ratio. A big who has bad passing ratings will get almost no assists. A big who has at least decent passing ratings will be closer to a 1-to-1 ratio.

The decision tree format is interesting. Does anyone have an opinion as to passing rating helping in keeping TO's down vs a press?     
My gut says yes. I just looked at two pbps against press teams and found the distribution of turnovers in the backcourt to be pretty even. It was 6 from PG, 7 from SG, 4 from SF and two from the frontcourt.

My SFs on this team have pretty bad BH and Pass. There limited turnovers are either due to their minimized value in the equation or presence of the team passing value. If I find some time and the ice storm doesn't knock power out I will dig a little more.
2/13/2014 8:23 AM
Posted by Trentonjoe on 2/13/2014 8:02:00 AM (view original):
No, I can't do stuff like that. I just look for patterns by sense of smell.
ok .  tommy    ;-)
2/13/2014 8:55 AM
Posted by oldave on 2/13/2014 4:30:00 AM (view original):
Seriously. Struggling with the concept of assists as window dressing. Probably mostly because I WANT it not to be window dressing.

So. Do we feel there is no instructive value in assist numbers?

One thing that jumped out at me with my Whittier wooden team this year is the horribly low number of assists we had in our losses.

We have a pretty good team. And are 22-4 with rpi about 10. But the 4 losses were incredibly ugly blowouts. They were all vs good teams. I happen to think I had more talent in each case. But impartial observers would probably say the talent level was pretty even.

Anyways. In those losses I think I averaged about 5 or 6 assisteses obviously. We had fewer made fg. But the rate of asset per made fg was very low also. Coincidence? Or am I confusing cause and effect or chicken and egg?
oldave, thing is, in our time, assists were 100% window dressing, thats based on my observations and the use of the "window dressing" term by old admin. that doesn't mean they won't have some rough correlation with success - as the formula used to figure how often you got an assist did depend on some ratings, which were valuable for success. today, its still the case, but there is probably a closer correlation - because now passing actually has significantly more value. passing in the old engine was only useful for preventing the turnovers *of the player in question*. in motion it wasn't even as important as bh in that regard. that is pretty low importance, when you think of how valuable passing really is. now, passing ups team fg%, so its more important. the formula for assists is largely based on passing, so now that there is more importance of passing, there is more of a correlation between assists and success. so coincidence, not really, but also, you can't dice individual assist figures.

think of it like this. if you are ranked in the top 25, you are probably doing decently well in the projection report. the projection report doesn't include ranking as a component, but it does include some similar components. if you were to dice rankings to try to find small connections between your actions and your NT seeding, could you get somewhere? maybe, but only because it has some correlation with NT seeding. going against the projection report is much better, and often, at small intervals, small increases in ranking actually result in small decreases in projection report - easy wins are way too valuable in the rankings compared to the projection report. so, you would often be lead astray.

think of assists as the top 25, except its much, much more poorly correlated. think of projection report as the value of passing, which we have no measurement for. direct measurement, that is - you could absolutely do FG% studies to get an idea of how valuable it is. i think we are talking multiple % points between high end and low end passing teams, in terms of effect on team mates' fg%, which would make passing in your pg one of the single most important ratings on the whole team. but i haven't ever studied it nearly as closely as i studied things in the olden days, so im really not sure.

anyway, bottom line is, yeah there is some connection between assists and success, and also between assists and the passing impact on fg%. but its more coincidental, just because there are some common elements, with varying degrees of similarity in their weightings. we have far better ways of measuring the value of our players than assists. so im not sure its still 100% window dressing, meaning there is absolutely no value in it to anyone, but in effect, its still window dressing. someone who knew nothing could get somewhere looking at assists, because they have no other knowledge, and it does somehow correlate to success. but its not strong enough for coaches like you and me to do anything with. thats my opinion, anyway - in the old engine, i was 100% confident of this, but now i just don't pay enough attention to details like that, to say with 100% confidence. maybe there is something there? but id be really surprised...
2/13/2014 9:53 AM
Here is the FAQ on the event tree:

The SimEngine simulates a game one possession at a time. The following is a high-level view of the decision process involved in every normal possession of a simulated basketball game. Normal possessions pertain to almost 95% of all possessions, but do not cover breakaways, put-backs or late-game situations:

We determine who has the ball by looking at every offensive player’s real-life possession rate. This value looks at his real-life field goal attempts, free throw attempts and turnovers. A player’s chance at having the ball for this possession is exactly relative to the real-life possession rates of the other players. If everyone has the same possession rate, everyone will have a 20% of possessing the ball. We use the composition of the team on the floor and the offensive set to determine how much time has come off the clock by the time this decision is made and the possession is terminated.
Now that we have picked a player, we must determine what he will do with the ball. There are three things he can do: turn the ball over (3), shoot (5) or be fouled (4). Every player has a real-life turnover percentage, field goal attempt percentage and times fouled percentage that is relative to his total number of actual possessions. At this point, many factors can adjust those percentages and affect whether we go to step 3, 4 or 5. Generally, we will view the chance that he shoots as the amount remaining after modifications to fouled percentage and turnover percentage. These factors include: the man defending the player, the rest of the defense, the type of offense, the type of defense, a player’s fatigue value, any team under-possession penalty and the player’s over-possession penalty.
If a player has committed a turnover, he may have committed a ball-handling mistake, poor pass or offensive foul. The percentage chance that an offensive foul has occurred is relative his actual personal foul rate. This is figured into the chance in the previous section. Ball-handling and passing turnovers each use a fixed percentage of the remaining chance that is based on historical averages for these turnovers. If it is one of these two types of turnovers there is a chance for a steal. This chance is based on the defenders and is relative the steal weighting used in the decision above. If a steal has occurred, there is a chance for a breakaway. Either way, the player who steals the ball is determined after the steal event is known. Each defensive player’s chance for a steal is relative to his contribution to the steal weighting in #2. If no steal has occurred, the ball simply goes to the other team.
A player has been fouled. This can be a shooting foul or a personal foul on the floor. This decision is determined using historical averages and the player’s real-life field goal attempt rate. If there is a shooting foul, a player has a chance to make the shot. We determine if it is a two or a three in the same way as we will below and the chance of making it is relative to his real-life shooting percentage, but discounted by a historical percentage. Either way, the foul is assigned to one of the defensive players. Each player’s chance of committing the foul is based on his real-life foul rate and is relative to the weighting used to determine if there is a foul in #2. If it is shooting, the shooter will be able to shoot the appropriate number of shots. If the final free-throw shot is missed, we proceed to rebounding (6). If it is not a shooting foul, we restart the possession.
A shot has been attempted. Whether this is a three-point attempt is strictly based on the shooters ratio of three-point attempts to non-three-point field goal attempts. The chance that a player has of making the shot is based on his shooting percentage, the assist rates of his teammates, his defender’s stop percentage (captured by defensive rating) and to a lesser extent the other four player’s stop percentages and the block rates of all five defensive players. If the shot is made, we determine if there was an assist. Assist chances utilize a historical average as a base that is modified by the sum assist rates from the other four offensive players. If there is an assist, every player’s chances are relative to his assist rate compared to the other players. If the shot is missed, we must determine if there was a block. Block chances are relative to the weighting that each player has in the field goal make or miss decision. Either way, if the shot is missed, we must determine who gets the rebound.
Every missed field goal attempt in a normal possession is rebounded. Rebounding looks at the real-life offensive and defensive rebounding percentages of all ten players on the court, though the shooter and his defender generally have less of a chance, especially in the case of a three-point attempt. We look at each one-on-one matchup for this situation (shooting team is still considered offense) to determine if one player has a better than average advantage over his opponent. These players will get a boost to their percentages, while their opponents will see their percentages drop. We then look at all ten modified percentages. The chance that an individual player has of getting the rebound is relative to how his percentage compares. If the ball is rebounded by an offensive player, there is a chance for a put-back. Assuming no put-back, the possession restarts. If the ball is rebounded by a defensive player that team begins its possession on offense

2/13/2014 10:51 AM
The # is 582, this didn't format well.

Also, this is for SIM basketball not HD.

The major difference I see is that SIM Basketball doesn't have a press option (which is why I think there is an early decision that what's listed above) and distro works a little different.

The article above makes it sound like the TO/FOUL/SHOOT  decision happens at the same time but I know the baseball sim does one thing at a time.    It's a series of yes's or no's.
2/13/2014 10:56 AM
Posted by Trentonjoe on 2/13/2014 10:56:00 AM (view original):
The # is 582, this didn't format well.

Also, this is for SIM basketball not HD.

The major difference I see is that SIM Basketball doesn't have a press option (which is why I think there is an early decision that what's listed above) and distro works a little different.

The article above makes it sound like the TO/FOUL/SHOOT  decision happens at the same time but I know the baseball sim does one thing at a time.    It's a series of yes's or no's.
why does it have to be a series of decisions with only 2 outcomes? there is no logical reason for that constraint. setting something up like that, you figure probabilities, then run a RNG - maybe the cutoff for outcome one is .65. so if its under .65, outcome 1 happens, over, outcome 2 happens. this correlates to a 65% chance of outcome 1. there is no additional difficulty in having say, 4 outcomes, with probabilities 10%, 30%, 20%, 40%. you just run a RNG and compare to <.1 for outcome one, >=.1 <.4 for outcome 2, >=.4 <.6 for outcome 3, >= .6 for outcome 4. 

i dont know the baseball sim, so maybe it is that simple. it would seem weird for hitting to be set up that way, but its certainly possible (either way). the HD sim though, im pretty sure, does decide between more than 2 outcomes at some points.
2/13/2014 1:08 PM
that is indeed the sim engine write up - i don't believe one exists for HD. but, old admin explained the HD engine was basically sim basketball, except instead of real life stats, you base on ratings and settings. so like distribution takes the place of real life possession rate. 

now there is significant divergence in HD from the original sim league basketball engine - however - i believe the core state transition logic is still in place.
2/13/2014 1:09 PM
I was wrong, there are a couple results with more than two options. I bumped the baseball decision tree. It's in the main baseball forum if you want to look.
2/13/2014 1:43 PM
Posted by oldave on 2/12/2014 1:58:00 PM (view original):
Posted by a_in_the_b on 2/12/2014 12:50:00 PM (view original):
Language!  There are some words, oldave, that should NOT be used in public!
i did not actually say the name of hewhoshouldnotbenamed  though... i was careful.

reminds me of plumpy... the original HD superstar.  

back then the forums were alot more active (not to mention i dont think there were subcategories like today....  so youd get one thread with discussion of recruiting jucos,  one thread of whether coach K was better than bob knight  and another with thousands of pictures of scantily clad females)  ,  

anyways if you made any mention of plumpy, or pizza, or gorilla suits.... BAMMM,  there he was with a witty reply.   i always suspected he had some kind of  program that alerted him any time certain keywords were uttered.  either that,  or he was god.

not sure why tha treminded me of plumpy....?.   
I've been summoned!!!  lol   oldave!!
10/12/2014 6:03 PM
◂ Prev 1234
players impacting team mates - passing raising fg% Topic

Search Criteria

Terms of Use Customer Support Privacy Statement

© 1999-2018, Inc. All rights reserved. WhatIfSports is a trademark of, Inc. SimLeague, SimMatchup and iSimNow are trademarks or registered trademarks of Electronic Arts, Inc. Used under license. The names of actual companies and products mentioned herein may be the trademarks of their respective owners.