Author Message
Italian Stallion
Joined: 04 Mar 2009
Posts: 112
PostPosted: Wed Jun 24, 2009 11:50 am Post subject: A different way of thinking about assists Reply with quote
I've looked at various models and it seems like assists are valued at anywhere from 1/2 to 1 point per assist when attempting to value player efficiency.
The recent LA series highlighted something that I sort of knew, but hadn't given a lot of thought.
There's a huge difference between some assists and others!
I'll just give you the extremes and leave it to the numbers crunchers to tell me if anyone considers this sort of thing.
1. Kobe Bryant has the ball, dribbles towards the hoop, draws a double/triple team, makes a good pass to Ariza who is wide open because of the double team and he knocks it down for "3 points".
2. Hedo Turkoglu lobs the ball well out of the reach of most players, but Howard makes an incredible leap and slams it down for two points.
First the differences.
In one case the made shot is worth 2 points and in the other it is worth 3 points.
In one case it was mostly the assist man's exceptional skill that accounted for the wide open shot and in the other it was mostly the scorer's exceptional leaping ability that accounted for the dunk.
Yet both of these scenarios get scored equally statistically.
To me this is preposterous!!!!
Kobe obviously deserves more credit than average for his contribution to the basket and Ariza deserves less than average because most 3 point attempts are not that wide open. (This is especially significant because it lead to 3 points instead of 2 and dividing 3 differently would have a bigger impact than divding 2)
In the latter case, the opposite is true.
THoughts?
Back to top
View user's profile Send private message
HoopStudies
Joined: 30 Dec 2004
Posts: 705
Location: Near Philadelphia, PA
PostPosted: Wed Jun 24, 2009 11:55 am Post subject: Reply with quote
There is a fair amount of thinking on exactly this in Basketball on Paper. Assists are a weird stat and it is useful to account for what you're talking about.
_________________
Dean Oliver
Author, Basketball on Paper
The postings are my own & don't necess represent positions, strategies or opinions of employers.
Back to top
View user's profile Send private message Visit poster's website
Italian Stallion
Joined: 04 Mar 2009
Posts: 112
PostPosted: Wed Jun 24, 2009 11:57 am Post subject: Reply with quote
Thanks Dean. I'll read it. I'm glad someone has actually addressed it. Very Happy
Back to top
View user's profile Send private message
thref23
Joined: 13 Aug 2007
Posts: 90
PostPosted: Wed Jun 24, 2009 2:47 pm Post subject: Reply with quote
The other factor, or another factor, is that there are many times when a good pass helps lead to a basket, but doesn't end up being registered as an assist. I would wager that a player logging 8 assists per game versus 4 assists (all else equal) is likely better than twice the passer compared with the 4 assist guy. For every assist, there are probably X many good passes made that do not end up as an assist. Would anybody argue this?
Of course its ideal to mine in depth data so you don't have to rely on box score data. But in lieu of that, everything is already less than ideal anyway.
Back to top
View user's profile Send private message Send e-mail
jmethven
Joined: 16 May 2005
Posts: 51
PostPosted: Wed Jun 24, 2009 3:03 pm Post subject: Reply with quote
82games.com also tracks a lot of numbers that provide a better idea of how each player's assists function for the team. For example, over half of Steve Nash's assists this year were on dunks or layups. This is not the case for a drive and kick player like LeBron.
Back to top
View user's profile Send private message
jayparadise
Joined: 18 Nov 2007
Posts: 3
PostPosted: Wed Jun 24, 2009 8:45 pm Post subject: Reply with quote
Is there a weighted assist statistic that is used? Something like (2*2PT assists) + (3* 3PT assists)?
Also, has there been any work done to look at assists to certain parts of the floor based on the eFG%? If Nash gets an assist on a dunk, that should be valued less than an assist for a jumper on the baseline, right?
Or how about this? The expected points on a three pointer from the baseline is 1.16 while the expected points on a three pointer from the top of the key is 1.03. So how about crediting an assist on the corner three as 1.84 points as opposed to 1.97 for a three at the top?
The expected points numbers were calculated from the data provided here: http://www.countthebasket.com/blog/2007 ... work-with/
Back to top
View user's profile Send private message Send e-mail
BobboFitos
Joined: 21 Feb 2009
Posts: 201
Location: Cambridge, MA
PostPosted: Wed Jun 24, 2009 9:01 pm Post subject: Reply with quote
Biggest issue with assists is the league doesn't record if missed shots (FGA that do not result in a FGM) were assisted or unassisted. Further, if fouls drawn were assisted or unassisted. Meaning, it would be nice to compare individual and team assisted vs unassisted shots (so a raw assist could actually have meaning, ie. "raises eFG% by 15%") but, the league does not take note of this.
_________________
http://pointsperpossession.com/
@PPPBasketball
Back to top
View user's profile Send private message
Harold Almonte
Joined: 04 Aug 2006
Posts: 616
PostPosted: Wed Jun 24, 2009 9:36 pm Post subject: Reply with quote
The PG's job is to look for the most efficient pass that the flow of the offensive can offer (the scorer's is to look for the most efficient way of scoring), is not into his control where scorers get to be free from their defenders. The same amount of skill he needs to assist for a dunk is probably the same he needs to assist a three, there's a lot of shot creation the scorer is responsible to do, not to mention to score at his average.
Does need a passer to be punished by a receiver's FGMissed?...maybe, if the FGA was closed defended. And of course they need some credits when helping to get FT's, to compensate. I also think an assist diminishes its value(but not the pass's worth) when the scorer needs to make decissions at least a move as simple as a fake (but continuations are a an advantage to offensive).
The question should be, is a third of a 2p score enough to well credit court vision, smart decission making, and ball handling skill, all in one action?.... half a score maybe?
Back to top
View user's profile Send private message
mtamada
Joined: 28 Jan 2005
Posts: 377
PostPosted: Thu Jun 25, 2009 4:02 am Post subject: Reply with quote
jayparadise wrote:
Is there a weighted assist statistic that is used? Something like (2*2PT assists) + (3* 3PT assists)?
Also, has there been any work done to look at assists to certain parts of the floor based on the eFG%? If Nash gets an assist on a dunk, that should be valued less than an assist for a jumper on the baseline, right?
I'd think that if we had an ideal system for valueing assists, assists on dunks should usually get MORE credit, because the passer created for the shooter the easiest shot in basketball (assuming that the player has the ability to dunk). Whereas a pass to a guy 25 feet from the basket is hardly an automatic 3 points; if the shot does go in, much of the credit should go to the shooter, and not so much to the passer. (Though I suppose the passer should get more credit for a pass to an open 3-point shooter than a pass to a guy with an open 19 footer: if they're of average abilities the pass to the open 3-point shooter has a more positive effect on average.)
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3608
Location: Hendersonville, NC
PostPosted: Thu Jun 25, 2009 5:42 am Post subject: Reply with quote
The effectiveness of the pass is converted automatically into its proper credit. Passes into high-FG% territory yield more assists (and more points).
Quote:
Is there a weighted assist statistic that is used? Something like (2*2PT assists) + (3* 3PT assists)?
This may not be a great idea. There are more uncounted 'assists' with 2FGA, because more of them lead to FTA.
Quote:
...a raw assist could actually have meaning, ie. "raises eFG% by 15%"
Well, a counted assist raises FG% to 100% (or 150%) every time. Since we don't count the missed FGA as 'assisted', the fraction that gets credit will have to do.
Just as a player may take a 'good shot' or get in perfect rebounding position, he's only rewarded when it works. You can only work to improve your odds of success, and you end up with the totals you earned.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
gabefarkas
Joined: 31 Dec 2004
Posts: 1313
Location: Durham, NC
PostPosted: Thu Jun 25, 2009 8:15 am Post subject: Reply with quote
BobboFitos wrote:
Biggest issue with assists is the league doesn't record if missed shots (FGA that do not result in a FGM) were assisted or unassisted. Further, if fouls drawn were assisted or unassisted. Meaning, it would be nice to compare individual and team assisted vs unassisted shots (so a raw assist could actually have meaning, ie. "raises eFG% by 15%") but, the league does not take note of this.
82games.com was tracking this a few years ago. I think I participated in a game charting project for this back in like '06 or so. Looks like the results are still on the site.
Back to top
View user's profile Send private message Send e-mail AIM Address
Scott S
Joined: 10 Feb 2008
Posts: 46
Location: East Rutherford, NJ
PostPosted: Fri Jun 26, 2009 1:36 pm Post subject: Reply with quote
I did some research on this subject and I think the initial assertion was actually reversed. By making the pass to Ariza for three, Kobe only marginally improved the team's scoring probability. (He also improves their scoring probability by being a great player and drawing the defense, but that is probably a slightly different subject.) Since the three point shooting percentages in almost all situations are very low, the average three point assist worth might be estimated by the following method:
Approx Scoring expectation of average possession (not including offensive rebounds): 1.00
Approx Scoring expectation of average assisted three point attempt: 1.20 (40%*3)
Approx value of three point assist :0.20 points
This is the value I obtained through regression in the following study:
http://basketball-statistics.com/sereda ... rtone.html
Note that the value of a 2 point jumper assist was similar for slightly different reasons. The value of that alleyoop was much greater in part because it much more greatly increases the teams scoring expectation.
The main problem with assists is the inconsistency of awarding them. I considered awarding different types of assists that we could assess different value to. I posted my thoughts at the following link:
http://sonicscentral.com/apbrmetrics/vi ... c&start=30
This is what I proposed as a method that can be easily and consistently tracked and can help indicated value more accurately.
"Assist1: A pass, with more than 3 seconds remaining on the shot clock, to one of the following three types of field goal scorers:
(1) any open shooter who shoots within 1 foot away from the three point line,
(2) any shooter who is within 5 feet of the basket or
(3) any shooter who has a numerical player advantage on the opponents end.
An Assist1 can only occur when no attempt to avoid a defender has been made by the shooter.
Assist2: Any pass resulting in a score that is not categorized as an Assist1 pass and improves his teams expectation for scoring compared to the expectation prior to the passer obtaining the ball.
These can definately be futher clarified. Assist2 is obvoiusly more subjective and can include "Free Throw Assists" and "Secondary Assists", among others but should still exclude the Ben Wallace assist above (this is the difficult part and I hope the impact would be minimal since the value might be near 0). The value of assist2s should be less than assist1s, but still above 0. I would imagine that the average other pass might even correlate negatively with expected value. Can anyone further clarify what could be included in this definition (not that I would be able to track this in bulk anyway, but just out of curiosity)?
There are an infinite number of ways to define assists, what is the consensus "best" way?"
Back to top
View user's profile Send private message Visit poster's website
Mike G
Joined: 14 Jan 2005
Posts: 3608
Location: Hendersonville, NC
PostPosted: Fri Jun 26, 2009 3:35 pm Post subject: Reply with quote
Scott S wrote:
Approx Scoring expectation of average possession (not including offensive rebounds): 1.00
Approx Scoring expectation of average assisted three point attempt: 1.20 (40%*3)
Approx value of three point assist :0.20 points
Scott, haven't you applied the 'assisted FG-attempt' value (.20) to the actual recorded Assist, rather than to the potential assist? If the receiver of the potential assist (pass leading to shot) only makes .40 of those shots, then only .40 of the potential assists actually become Assists.
Another way to look at it : the average unassisted 3FGA is worth maybe .95, while the Assisted 3FGA is worth 3.00 ; only made FG produce an assist, after all.
So, we may say (trivially) the Assisted attempt was worth 3; or that the potentially-assisted attempt is worth 1.20 . The extra .20 resulting from the pass, divided by the 3FG% (.40) = .50 .
If the passer makes 10 such passes, he gets 4 Assists. The team scored 12 points, rather than 10. There you have +2 points on 10 passes (potential assists); that's .20 per pass. But per Assist, it's .50 extra points.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
Scott S
Joined: 10 Feb 2008
Posts: 46
Location: East Rutherford, NJ
PostPosted: Fri Jun 26, 2009 5:38 pm Post subject: Reply with quote
Mike,
What you are saying is actually true, but I had no measurement of potential assists. So, on average, the best estimate I currently have ospotential assists is an actual assist. I guess if we were crediting plays we would credit .5 to the assister of a made shot and 0 to the assister of a missed shot (or something like that), but, on average, an additional three point assist indicates .2 additional points attributed to the assister. I think this is the best way to look at it until we have further information because all you can ask a player to do is play the probabilities. I think my calculation still applies, but I didn't notice it at the time I did my study and I am sure there are some other factors at play.
Back to top
View user's profile Send private message Visit poster's website
Mike G
Joined: 14 Jan 2005
Posts: 3608
Location: Hendersonville, NC
PostPosted: Fri Jun 26, 2009 6:01 pm Post subject: Reply with quote
Scott S wrote:
... on average, an additional three point assist indicates .2 additional points attributed to the assister. ..
In other words, on a made shot you'd give .2 to the assist man; and on the missed shots he'd get zero?
_________________
`
36% of all statistics are wrong
Author Message
Scott S
Joined: 10 Feb 2008
Posts: 46
Location: East Rutherford, NJ
PostPosted: Sat Jun 27, 2009 1:48 pm Post subject: Reply with quote
Basically, I think you should "reward" the pass itself. That includes both made and missed shots. The value of the average such pass would be 0.2 points if you are lumping them together. If you also have potential assists, then you could more accurately reflect past value by rewarding 0.5 points to actual assists and 0 points to potential assists of missed shots (I am referring only to 3 pointers). However, much of what causes the shots to miss can be attributed to luck or situations beyond the passer's control. Because of this, although including potential assists in addition to actual assists should be much better in reflecting past value (especially in a short time frame), but only moderately better at predicting future value (or with much available data) because I would suspect the large majority of the result of the shot depends on the luck of the shot falling or not falling.
In short, if you only have actual assists, they should be valued as 0.2 points because of the correlation with missed shots of potential assists.
If you have potential and actual assists, 0.5 and 0 (possibly a little more than 0 on average expected impact) would probably be a little more telling.
Back to top
View user's profile Send private message Visit poster's website
Italian Stallion
Joined: 04 Mar 2009
Posts: 112
PostPosted: Sun Jun 28, 2009 4:10 pm Post subject: Reply with quote
The problem I have with some of the analysis I am seeing is that it's missing the very specific things I pointed to and seems to be getting into other good, but more general issues.
It's one thing to throw an ally oop pass to an athletic player who can make the play and another to throw one to Howard where no one can defend it because it's so high it and much easier for the passer to accomplish.
In the first case, it probably takes a great play by both the passer and scorer to both time it, execute the dunk, and make the pass to a good location. So the value needs to be divided in a way to reflect that.
In the second case, Howard is such an incredible athlete, the timing and location of the pass need not be nearly as good to accomplish a successful play. More value should therefore accrue to Howard than the dunker in the other example.
A similar case exists with my example of Kobe.
It's one thing for a player to find a wide open teammate behind the arc and deliver a good pass. It's another thing to be the specific reason he was open to begin with and to then deliver the pass despite being double teamed. The division of the value in the latter case shoud be more towards the passer than in the first.
I am not commenting on which of these or other situations deserves more value. That's for you guys to determine. I'm just saying the division should reflect the contribution of each player to the specific play.
Last edited by Italian Stallion on Sun Jun 28, 2009 4:26 pm; edited 1 time in total
Back to top
View user's profile Send private message
Italian Stallion
Joined: 04 Mar 2009
Posts: 112
PostPosted: Sun Jun 28, 2009 4:15 pm Post subject: Reply with quote
By the way, I also think something has to be done to reflect the value of great passes that lead to another pass that leads to a bucket (like hockey assists).
Great passes that didn't lead to a bucket because the shooter missed is bit less bothersome to me. I sort of hope it evens itself out over time (or close) when comparing players. I'm sure it doesn't because some teams are loaded with better shooters, but I wonder how big that is.
On an entirely different subject, blocks are probably not correct either. If one guy blocks just under 3 shots a game and alters 4 and another blocks just over 2 but alters a 12, the latter may be the more valuable player.
Back to top
View user's profile Send private message
DJE09
Joined: 05 May 2009
Posts: 148
PostPosted: Sun Jun 28, 2009 8:19 pm Post subject: Reply with quote
The problem with much of your analysis, although I agree with much of it, it seems to bring a lot of subjectivity and personal judgement into it.
You say it is easier to throw an Ally-oop to Dwight since he can jump so much higher, so it is a less skilled play than, for example, a Paul-Chandler Oop. But then, perhaps, isn't that a more savvy basketball play, since it has a much greater chance of success, so we should reward tha passer for recognising the higher value play and executing it.
Similarly, you rightly point out there is great value to a team in a player being able to not only draw/demand a double team (Kobe), but to pass out of the double team to the open man. But then we don't penalise Kobe when he decides to take the other, much lower percentage option, of shooting over the double team with the 15ft jumper with a hand in his face - and makes the shot - even though when he chooses this option he usually misses the shot.
There has been much discussion recently about Blocks, and the relative value of Blocks where the defending team doesn't gain posession. It seems here in this discussion that you are really saying that Box Score stats don't truely reflect the whole value of a player with respect to that particular attribute (Blocks = Shot Contesting, Assists = Passing / creating 'open looks'). You are absolutely right. I am absolutely sure teams chart things like double teams encountered (and where on the court and what the player did to get out of it), how many times a player comes to 'help' or double team on defence. How they defence the pick and role (show and recover etc), and none of this is captured by Box Score stats, or even play by play data. Until some of that stuff is routinely (by this I mean Publically available) charted we can't really analyse these things.
I guess I view some of what goes on here as trying to glean what we can about the relative value of players, or combinations of players if we consider team lineups a better measure than individuals, from the standard box score or play-by-play data. If you are going to judge the quality of assists, then we have to have some objective criteria, currently this is "pass that leads to a scored basket". This certainly does not encompass all of a players ability to create shots / set up team mates with shots, but can give us some useful information regarding this - i.e. Chris Paul is very good at this, whilst Emeka Okafor is not.
Personally, I have a low opinion of people who break their team defensive patterns to gamble on steals, I accept that those steals generate high quality offensive posessions but feel that 'failed' attempts often leads to higher quality offensive execution to their opposition. Since 'failed steals' are not charted, I can't properly decide this issue, and am left with my own subjective opinions (i.e. Arenas does this too much, but it is Ok for DWade to gamble like this Smile).
Back to top
View user's profile Send private message
Harold Almonte
Joined: 04 Aug 2006
Posts: 616
PostPosted: Mon Jun 29, 2009 8:15 pm Post subject: Reply with quote
There are some metrics which has the "AssistFigure" adjust, in order to weight assists. I think these approachs aren't so subjective. It's not so relative difficult to define the amount of shot creation of a passer.
In the Kobe example, he's drawing two defenders, and obviously this has a big amount of shot creation. The same if he is letting his defender behind and then drawing the Ariza's defender, creating space for him. (two at a time, or one by one)
In the Alleyoop example, there's less amount of creation. The two players need to create their spaces in order to make their things. (even though one of them is not handling any ball)
In a fast break, the value of an assist could be even lower, because there're more variables in the creation, like the rebound, the rebounder's initial pass, and the lack of defensive recovery. (should these rebounds value between an ordinary rebound and a steal?)
Back to top
View user's profile Send private message
DJE09
Joined: 05 May 2009
Posts: 148
PostPosted: Mon Jun 29, 2009 9:03 pm Post subject: Reply with quote
Ok, so here you are attributing value of an assist as "degree of difficulty in Shot Creation". Whereas in my comment I was valuing an assit based on the efficency of the shot that was created / attempted (open 3 vs relatively uncontested dunk). This is where the subjectivity comes in - what is valuable.
In the Kobe example, it is his ability as a scorer that draws the double team, not his ability as a passer. Also Ariza has to get to the 'open' space for Kobe to pass to him as well (vis-a-vis the Ally-oop) and knock down an open 3 (as opposed to a dunk...). I also note that often it is the player who penetrates that 'creates' the open player on the perimeter, and may pass out, but the ball will be passed on (due to the defender rushing out on the first 'open' perimeter player - the hockey assist) so the creator is not credited with the assist at all in the box score (or PbP data).
Assists record when a pass lead to a score. I do not agree (nor does this thread bear out) that there is consensus as to which types of pass or shot creation is more valuable, hence my claim of subjectivity in the analysis. The fact of the matter is passing ability should be evaluated independently of team mate's ability to score off the pass - that is the potential assists and hockey assists that have already been talked about. Incorporate passing turnovers should be part of this too. This is the sort of things coaches know by 'intuition', ie. they are evaluating their memory of that player's passing/shot creation skills - whereas we are using assists (and possibly a few other things) as some sort of proxy.
It seems to me here, the source of Stallion's original post was a desire to have a meaningful way of disaggregating assist data, so we can comment on the relative merits of players (not all assist are created equal). I reitterate my point, that I think most of the options discussed here are subjective.
Back to top
View user's profile Send private message
Harold Almonte
Joined: 04 Aug 2006
Posts: 616
PostPosted: Tue Jun 30, 2009 2:53 pm Post subject: Reply with quote
-Passing is just a piece of Shot Creation, as it is Ballhandling, and Driving Moves Spacing (with and without the ball).
-Hockey assists should be credited in a boxscore built for metrical issues.
-A passing rating which consider passing turnovers only, hiding the FG% situation of the receiver(or any other decission making situation), is exactly that: a passing rating (even though PGs will be always the best). 82games does have one, but I don't know its metric.
Back to top
View user's profile Send private message
ecumenopolis0
Joined: 15 Jul 2008
Posts: 22
Location: Houston
PostPosted: Tue Jun 30, 2009 7:15 pm Post subject: Reply with quote
Harold Almonte wrote:
-Hockey assists should be credited in a boxscore built for metrical issues.
There are some pretty lame hockey assists that we don't even notice unless we look for them like this.
Also, sometimes there's a kickout and the ball is reversed around the perimeter as the D scrambles -- Duncan out to Finley, Finley to Ginobili, Ginobili to Bowen (Spurs going small Wink ). Should we count only Finley's as a hockey assist? Why not Duncan's? Because to me, any one of these passes is just as valid as the others. It's just one pass leading to the next until someone gets open.
Heh, what if Smith passes to Billups passes to Smith for the three? Does Smith get his own hockey assist? Probably not -- but he may have helped to create his own shot, right?
The hockey assist may have value, but I guess it seems too messy a number to me.
Back to top
View user's profile Send private message AIM Address
Harold Almonte
Joined: 04 Aug 2006
Posts: 616
PostPosted: Tue Jun 30, 2009 8:37 pm Post subject: Reply with quote
Quote:
Because to me, any one of these passes is just as valid as the others. It's just one pass leading to the next until someone gets open.
In a chain of passes (with no stopped balls) wich lead to a final assist, probably all of them count as a "hockey assist", but probably some of them a portioned value of the next. Anyway, was the last pass which seemed to have the best "probablility to become an assist", and it depends finally on the shooter that the potential assist and previous hockey assists get any boxscore value at all (the chicken and the egg).
Yes, it seems a messy.
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3608
Location: Hendersonville, NC
PostPosted: Thu Jul 02, 2009 9:19 am Post subject: Reply with quote
It's not just messy, it's fairly pointless. All these good passes that don't produce a basket are worth something, potentially. But so are picks, decoys, and countless other positive acts. One could add up a myriad of .2's and .02's, and in the end they'd just add up to the actual assists.
There may in fact be players who somehow specialize in 'good passes which are not assists'. But the difference between their assists and their fractional/potential assists is likely dwarfed by the differences in various scorekeepers' liberality in granting assists. That, it seems to me, is where to focus on 'assist justice'.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
DJE09
Joined: 05 May 2009
Posts: 148
PostPosted: Thu Jul 02, 2009 6:34 pm Post subject: Reply with quote
I agree Mike,
Let us not forget the aim is to get the ball into the basket. There is heaps of great stuff that people can do on the basketball court that isn't recognised, Bruce Bowen looses his man and gets open for the corner 3 - if he doesn't get passed the ball it doesn't mean anything, but the Spurs still really really want him to do that.
Players who get assists tend to set their team mates up to score. They do that more than the number of assists they are credited with. I can't think of a player who has an amazing reputation for setting up great looks for team mates who doesn't get assists. It makes sense to assume that the Number of assists (passses that lead to FGM) credited (Yes I am ignoring home court bias Mike) is proportional to then number of potential assists they created (passes that lead to FGA).
When you look across the board at Ast% (both % own shots are assisted, and % they assist team mates), TO%, and Usage you get a reasonable picture of the degree of facilitation a player assumes in their team's offense. When you factor in playing time with this, you can figure the relative strength of the team in this area and how much it is valued by the coach (but obviously there is more to basketball than setting up your team mate to score, something about stopping the other team from doing so too Smile ).
Back to top
View user's profile Send private message
Harold Almonte
Joined: 04 Aug 2006
Posts: 616
PostPosted: Thu Jul 02, 2009 10:45 pm Post subject: Reply with quote
If it were possible, a good excercise would be to watch at how much of the total assists in a season, come from these multi hockey passes. And this added to number of "screen assists", would give an idea wether it's relevant to take ever that on consideration.