Shot creating in offense is seldom talked about yet a very important part of any offense. There aren't any stats for fans to freely access. When a team has players who can create shots for themselves and teammates, team's offense flows and scores on higher % shots. Without good shot creators, you will see stagnant and sluggish offense which will result in bad shot selections or low % of shots. I have found a way to roughly calculate how many shots a player created for himself and for his teammates. Basketball-reference.com tracks player assisted FG % for each season. The FG's weren't assisted were ones he created for himself. For example: Harden's Assisted FG% was .313 last season. His % of un-assisted FG was .687. In other words, 68.7% of his shots were created by himself. His FGA/Game was 17.1, which translates into 11.7 shots he created on his own. Let's tally for his shot-creating for his teammates. His APG (assists per game) was 5.9 last season. Rockets, as a team, shot 46.1%. In actuality, 12.9 shots were helped by Harden's pass but only 5.9 shots went in. In total, Harden created 24.6 shots (11.7+12.9) for himself and teammates. Here are the % of un-assisted FG's: Harden 68.7% Lin 60.6% Parsons 26.3% Beverley 44.3% Delfino 19.9% Asik 31.4% Dwight 30.2% Here is a shot creating chart for Rockets last season: Code: Creating for Harden Lin Parsons Beverley Delfino Asik Self 11.7 6.6 3.2 2 1.9 2.4 Teammates 12.9 13.2 7.6 6.3 4.3 2 Total/Game [B]24.6 19.8 10.8 8.3 6.2 4.4[/B] Per 36 23.1 22.1 10.7 17.2 8.9 5.3 I was quite shocked when I looked into Dwight's shot-creating ability. Here are his numbers last season: Code: Creating for Dwight Self 3.2 Teammates 3.1 Total/Game 6.3 Per 36 6.3 I guess by looking at his FGA/game, which was a measly 10.7, Dwight was never an important part of Lakers offense, nor a good shot creator. Dream's gonna work his miracle on Dwight. Our best shot creators are: Harden>Lin>Beverley>Parsons>rest of the team. It's absolutely imperative that we have at least one of Harden and Lin, plus one of Bev and Parsons on the court at any time during the game. They are our little engines that will make Rockets run next season.
Good try, but unfortunately it makes a very likely off guess that players shoot the same percentage off a pass than he does in iso. 82games(currently offline where I am for whatever reason) has a breakdown of what kind of an assist each player has. The issue even with that is it doesn't given the percentage of shots on makes and misses. I think this type of determination is very difficult unless you're a team, and have access to possession stats beyond what's offered in the public. Overall I'd just go with the eye test. Harden #1. Lin a distant #2. Parsons #3. And Beverley doesn't have enough minutes to really judge. Everyone else basically played off the ball for the most part.
Great analysis. Whenever ts% are brought up, I always feel that is not the best way to judge their offensive prowess. It's all about the shots created and what kind of shots are generated.
I'm not sure what you're trying to prove. Harden and Lin are good "shot creators" because they are primary ball handlers and they also are good at attacking the basket on dribble drives. You don't want or expect your center to do that. I would guess that anyone who brings the ball up court a lot will have a higher % of unassisted scores. Also players who score off offensive rebounds would also boost their %, but I wouldn't exactly call it shot creating.
Does everyone has the skill to score off rebounds? Of course not. Players who can score off rebounds are just as skilled as any PG who scores from coast to coast. It's still a shot they created themself WITHOUT the help of a teammate. One scores off his speed and agility while another off his rebounding skills. I am just providing some raw data which you can't find anywhere on the web. To those who read into them, they might find some useful information. One man's trash is another man's treasure.
This is exactly why we need to start Patrick Beverley and Have Jeremy Lin come off the bench. Not because Beverley is the better player, but because this makes us a better team. Beverley would be a "3 and D" guy in the starting lineup, since he is superior to Lin in these areas. Lin is a superior scorer, passer, and just overall shot creator which is why he would perfectly fit as the Terry, Crawford, Ginobili, Harden (with OKC), 6th man role. I think making Lin our 6th man makes us a better team and allows Lin to be "Linsane" as soon as he steps on the court, rather than starting and deferring to Harden. We would have a great bench with Lin, Asik, and Garcia leading the way.
Interesting, I was thinking in the same way, but was not using this calculation to estimate shot-creating capability, but rather use it to estimate assist efficiency. Basically if one pass was sent to a big man at rim who have 68% shooting efficiency at rim, your chance to generate 1 assist is 0.68, then based on the assist number and assist distribution, one could calculate how many possessions (assist attempt) one would need to generate that assist. Then assist number over estimated assist attempt number would gives assist efficiency, basically gives some idea about the chances of a pass would end up in assist. So here is the breakdown of our 4 major passer on the team, within last season have over 100 assists. One can see that Jeremy tend to feed much higher percentage of at rim shots, and James kick out more 3pt shots. And Beverley gives out much more weighted 3pt assists than at rim passes. Which somehow consistent with my eye test, and also might due to size disadvantage, Beverley sometimes struggle to send out entry pass from outside, whereas Jeremy was pretty good at. However, people would argue that 3pt though lower percentage, but it generate more points than at rim shot. So I extracted the teams' eFG% number with breakdown of the shot type. \ One can see that across the board on the team, even compare eFG% whereas 3pt advantage was considered, at rim shot still a way higher % shot over 3pt. capability of deliver the ball close to the rim definitely would be a plus for a playmaker, especially we now have DH12 on team, an on time entry pass would be sweet. Okay, so plug in the FG% of player who received the assist, Jeremy basically will need 2 assist attempts to generate one assist, which converted at 50.1% assist efficiency. And James has 48%, and Parsons has 47.1%. Beverley on the other hand was the lowest within the 4 here as 43.5%, so basically he will need more possession to generate 1 assist. However, this calculation does contains error, since those hockey assists were not properly credited. Play like this the initial driver/playmaker did not got assist credit, and the last passer got it. My eye test tells me that James and Jeremy initiated a lot more this kind of play, thus their assist efficiency was underrated a little bit, and Parsons had some of these kind of assist credit to him, thus his efficiency is a bit overrated.
Uh, I'd argue that point. Look at the assist distribution of J. Lin vs. P. Beverley here. whereas J. Lin pass ~ 13% of his passes to James Harden, and a total ~23% of passes to our centers. Whereas P. Beverley gives only 7% of his assist to James, and 14% of his assists to our centers. Well, obviously there is effect comes from whom they are playing with, but the tendency is there. P. Beverley tend to pass to 3pt shooters than generate at rim assists. I do have concern that P. Beverley could efficiently deliver the ball to DH12. Last season it might not be that important since we don't have very much post presence, but the coming season, I'd think that would be more important. You could not count on James Harden to create for himself, create for DH12 and other team members, and yet got less assists from another guard on the floor, that will for sure breakdown his fast. I think stacker James and Jeremy's time would do perfectly.
It was a heck of a try. My kudos to the author. He put a significant amount of work into the thread. stats.nba.com and 82games.com are vital resources for posts like this. But again OP - good job.
First, thanks to TTNN for all the detailed data on this subject and fans who did not derail this thread. Repped for your time and effort! Meh is correct on two things: 1. Harden is a lot better at creating shots than number indicated. My raw data did not include his FTA/game. I don't think anyone tracks the % of FT's which was un-assisted. 2. Bev had a small sample size, plus he was still a rookie played his first 40+ games, which do not give a clear indication if he is better than Parsons. Bev played in an 'adrenaline zone' and needed to earn his rotational spot last season. He should settle down and we shall see what kind of shot creator he is. Thanks! Contrary to what you believed, I did not really spend a lot of time on this. If I did look into 82games.com, nba.com or mysynergy.com, I wouldve probably spent days on this. This is something that came up when I looked into some of the players' assisted FG%. I am more fascinated by RAPMs and how I can use Engelmann's numbers to project wins for next season. I'm just waiting for all the teams to finish their FA acquisitions. It's summer w/o bball. Tally up numbers is a lot better than watching highlights from last year.
Code: Paul Bledsoe Griffin Crawford Others Self 9.5 4.6 4.8 5.8 Teamates 20.3 11.3 7.7 5.3 [B]Total/game 29.8 15.9 12.5 11.1 11.3[/B] Edit: % of un-assisted FG's: Paul 77.7 % Bledsoe 60.9% Griffin 35.9% Crawford 43.4% Clippers lost Bledsoe but picked up Collison this offseason. This is the their biggest downgrade this offseason, going from Bledsoe to Collison.
This chart is good but it does'nt really compared the assist ratio of Lin and Bev. Bev played with 2nd unit most of the time whereas Lin played with other starters, that's why Lin got more assists to Harden , Omer, Parsons.
Wow, excellent posts kuku and TTNN! This makes me wonder what kind of stats Morey has that others don't seem to look for?
Hey OP, good stuff. Thanks for starting a thread during the summer that actually has merit. I like where you are going with this. In my opinion, as you alluded to, to get clearer idea of how Harden directly created shots, you have to account for shot attempts that are shooting fouls and do not show up as part of FGA's. While Harden had 5.9 assists per game...he also took a lot of shots that do not go down as FGAs because he was fouled and shot free throws and he also threw some "Free Throw Assists". Those are just as valuable to the offense as an assist. That is to say.. a pass to a teammate .. that causes the opponent to foul the teammate while the teammate is shooting, sending the teammate to the free throw line. Those never hit the assist column or the FGA column but they are very valuable plays ... in a lot of ways more valuable than a pure assist itself..because of foul count and because most players are more efficient shooting free throws than they are shooting field goals. So here's how the free throws he shot and the free throws attempts that his teammates shot as a result of his shot creation could be accounted for. 1. take 44% off Harden's FT attempts and add this number to his overall FGAs. Then take 68.7% of this number. This is effectively the number of shots Harden actually created for himself with some error induced due to non-shooting fouls in the bonus. However a non-shooting foul in the bonus is still viable to me since Harden is such a good free throw shooter. Fouling him in the bonus is desparation by the opponent. This calculation yields a total of 14.845 shots a game that Harden created for himself. Or, a more accurate definition, 14.845 possessions that Harden used to create his own offense, whether it be a shot or an opponent foul leading to free throws. 2. Next, calculate the team's actual FG percentage on all the shots that Harden didn't take. In other words, subtract harden's FGs and FGAs from the team's FGs and FGAs and then get the percentage for the rest of the team minus Harden. (In Harden's case this would make the team shooting percentage slightly higher at 46.6% since he shot 43% himself. A small difference but nonetheless more accurate.) 3. Now approximate how many FGAs Harden created for his teammates using 5.9/.466 = 12.66 FGAs. 4. Now approximate the number of shots created by Harden that went down in the box as free throw attempts by his teammates. The rest of Harden's teammates took approximately 66 shots per game (19%). If Harden created 12.666 of those 66 shots (19%) that they took per game, then it should be fair to approximate that he created approximately 19% of the free throw attempts that they took as well. Harden's teammates shot 15.8 free throws per game. Multiply that by 0.44 and you get right at 7 shots (possessions) per game. Multiple that by .19 and you get 1.32 shots (possessions) that Harden created for his teammates where they went to the foul line. (Note: This calculation does induce some error since I did not account for Harden only playing 78 of the 82 games and it doesn't account for fouls in the bonus but it should be close enough for a rough calculation.) Now add together the possessions Harden created for himself plus the shots he created for his teammates plus the possessions he created for his teammates where they shot free throws: 14.845 + 12.66 + 1.32 ~ 28.82 possessions that Harden either created offense for himself or his teammates. ........................ ........................ ........................ Now to take this a step further, Harden averaged 3.78 turnovers a game. That means that ~88% of the possessions he used turned into a shot or free throws while ~12% of the time he gave up a turnover. As I kept noting during the season....when Harden had 3 turnovers or less per game...our record was phenomenal. When he had 4 or more turnovers per game...we were not very good. Harden only has to improve in one area....cutting the turnovers down while maintaining shot balance and creation. I expect that will happen this year. Kudos to kuku for the thread.
To basketballholic, Thanks for your inputs. As I'd mentioned, the raw data I provided didn't include FTA's. Two reasons: 1. I wasn't willing to spend more time on it. 2. Make it simple. Too many numbers and explanations can become confusing. As for assists per game (APG) loss due to FT's, I didn't think it's neccessary to take them into account. The player's totally/game and team's FGA/game might be affected but the final distribution % will not be skewed by much.
