"Playmaker Rating": your thoughts?
So here is something I've been think about lately. I threw together some numbers in a conversation about Jackson's ability to overcome his turnovers and poor shooting with his assists and three point shooting. That got me thinking about what can happen when a guy gets the ball in his hands. He can do a number of things and they will all have a positive outcome, a negative outcome or a neutral outcome. His skill, athleticism and decision making combine to give you a guy's ability to make a play on offense. What I'm going to try to do here is put together a metric that measures a player's ability to create a positive outcome on offense against negative outcomes.
So when you take a look at the possibilities for a positive outcome you basically have a made 2pt FG, a made 3pt FG, an assist, and a made free throw (or preferably two). Obviously a made three should be worth more than a made 2pt FG and a free throw worth less.
As far as negative outcomes go, you have a missed FG, a turnover and a missed FT. A missed FT is probably about half as bad as a missed FG and a turnover is probably worse (no chance for an offensive rebound and a higher chance for easy transition points for the opposition) but I'm not sure how much worse so for the time being I will weight it equally.
Neutral outcomes don't really make a difference here and aren't generally tracked so we will ignore them. In case you were wondering I'm basically talking about dribbling the ball or passing without creating an advantage or losing possession of the ball. These can be important working within a system and running an offense but I'm not going to give a guy a whole bunch of credit for being a minor part of a play without screwing it up.
So putting these things together I came up with a really simple formula to measure these outcomes against one another with the intention of providing a single figure that can show how someone performs on offense with the ball in their hands. I want to be clear that I don't intend this to be one of those catch-all stats that will tell you that Player A is better than Player B. This only takes one very important part of the game into play. It doesn't deal at all with defense or rebounding which of course are critical when evaluating a player.
I'm calling this metric "Playmaker Rating" for the time being. It goes something like this.
Positive outcomes per 36 minutes (you could use any number of minutes, but I do want to control for variation in minutes played so I chose the readily available per 36 stats):
2pt FG made +
Assists +
Free throws made/2 (because they are worth half as much as a FG) +
3pt FG made x1.5 (I think you get it) =
Positive outcomes per 36
Negative outcomes per 36 minutes:
FG missed +
Turnovers +
FT missed /2 =
Negative outcomes per 36
Positive outcomes - Negative outcome = Playmaker Rating
Here are some examples
Chris Paul: PO/36 20.7 - NO/36 11 = Playmaker Rating +9.7 (Playmaker Ratio 1.88:1)
Jose Calderon: PO 15.8 - NO 7.1 = PMR +8.7 (PMRatio 2.23:1)
Deron Williams: PO 19.9 - NO 11.8 = PMR +8.1 (PMRatio 1.69:1)
LeBron James: PO 20.5 - NO 14.8 = PMR +5.7 (PMRatio 1.39:1)
Tony Parker: PO 18.3 - NO 12.7 = PMR +5.6 (PMRatio 1.44:1)
Kobe Bryant: PO 18.9 - NO 14.3 = PMR +4.6 (PMRatio 1.32:1)
Allen Iverson: PO 13.6 - NO 11.5 = PMR +2.1 (PMRatio 1.18:1)
Dwight Howard: PO 11.7 - NO 10.6 = PMR +1.1 (PMRatio 1.10:1)
Here are some Warriors:
Stephen Jackson: PO 15.2 - NO 12.9 = PMR +2.3 (PMRatio 1.18:1)
Jamal Crawford: PO 13.1 - NO 11.2 = PMR +1.9 (PMRatio 1.17:1)
Marco Belinelli: PO 11.4 - NO 10.1 = PMR +1.3 (PMRatio 1.13:1)
CJ Watson: PO 10.3 - NO 7.6 = PMR +2.7 (PMRatio 1.36:1) Surprised? I was.
Here are some other ones I found interesting:
Baron Davis '07-'08: PO 17.1 - NO 13.1 = PMR +4.2 (PMRatio 1.31:1)
Allen Iverson '00-'01 (MVP): PO 17.3 - NO 16.4 = PMR +0.9 (PMRatio 1.05:1)
Allen Iverson '01-'02 (Practice?): PO 17.5 - NO 17.9 = PMR -0.4 (PMRatio 0.98:1) Seriously??
So here is where you come in. I'd love to get some feedback on this idea. Is there a place for this in the sea of existing metrics? Did I overlook something important?
Also I just wanted to single out JAE specifically because I know this kind of stuff is in your wheelhouse. Please let me know what you think. Also, if you find this the least bit interesting I'd love to see what your NBA stat machine doohickey could do with this on a larger scale. (It takes me forever to do the numbers myself, I'm not much of a math guy)
Thanks for taking the time to read what has to be my longest FanPost yet.
This FanPost is a submission from a member of the mighty Golden State of Mind community. While we're all here to throw up that W, these words do not necessarily reflect the views of the GSoM Crew. Still, chances are the preceding post is Unstoppable Baby!
3 recs |
35 comments
Comments
It's interesting, but I have a few suggestions
You should factor in steals & blocks, but blocks less so because they often go back to the opposing team. They are definitely “playmaker” events. Especially steals, which often lead to a fast break and easy points which are big momentum changers. I suppose personal fouls could be involved too on the negative side? But sometimes they’re good fouls, and offensive fouls are already counted in turnovers.
Also, missed FTs are not half as bad as missed FGs, there’s still the foul drawn, which is inherently a good thing. I’d go to a third. Turnovers are much, much worse than a missed FG. The NBA average for rebounding is 70-75% defensive, 25-30% offensive, so you lose the ball 72% of the time you miss a shot and 100% of the time you turn the ball over. Factor in the fast break stuff, and a turnover is probably about twice as bad as a missed FG.
Finally, I think you’re also overvaluing the negative occurances. You’re using the same unit for positive and negative occurrances, but I don’t think that’s fair: What’s the worst thing that can happen from a negative occurance? The opponent scores. That’s the worst possible outcome, which is equal to a made bucket.
I think you should use the below formula:
Positive outcomes:
2pt FG made +
Assists +
Free throws made/2 +
3pt FG made x1.5 +
Steals +
Blocks/2 (because they often go back to the other team) =
Positive outcomes per 36
Negative outcomes
FG missed/2 +
Turnovers +
FT missed/6 =
Negative outcomes per 36
So, to summarize:
add steals & blocks
lower the effect of missed field goals because they sometimes end in offensive rebounds
slightly lower the effect of missed field goals because the player still drew the foul which is a good thing.
"No no Nene!"
http://www.youtube.com/watch?v=zB5DxNl4EB0
by Dubs fan in Boston on Mar 6, 2009 7:21 AM PST reply actions 0 recs
I think he was just trying to factor in those things that happen in an offensive series where turnovers are possible and blocks and steals do not enter into that. They affect the outcome of the game, but the guy tasked with directing the offense isn’t thinking block vs assist. It’s also not really different from “Win Score” if it includes these. Win Score is a good metric.
An easier way to compute it is:
Points – FG*attempts* – 1/2FT*attempts*. It works out identical to the numbers you OM had with less to track.
by jae on Mar 6, 2009 8:08 AM PST up reply actions 0 recs
An easier way to compute it is:
Points – FG*attempts* – 1/2FT*attempts*. It works out identical to the numbers you OM had with less to track.
D’oh. Haha, I told you I’m not much of a numbers guy. That probably would have never occurred to me if you hadn’t pointed it out. Thanks.
Thing 2
by olympicmike on Mar 6, 2009 12:13 PM PST up reply actions 0 recs
JAE...
Help me out. I can’t get it to work that way. What am I doing wrong?
Thing 2
by olympicmike on Mar 6, 2009 2:34 PM PST up reply actions 0 recs
DFiB...
Thanks for the feedback. I was thinking about including some other aspects, but like JAE mentioned that would be too similar to “Win Score”. Also after thinking about it I wasn’t really after getting a players overall impact on the game, just a breakdown of what happens when he tries to make a play on offense. Maybe I should change the name to reflect that better.
It’s aimed more at determining who’s hands you want the ball in on offense. Our current situation (not having a legitimate PG) was kind of the inspiration. CJ scored the best, which I guess shouldn’t have surprised me as much as it did, and Jackson scored the second best (which kind of shows that despite his faults he is basically the best we have to run the offense through). This isn’t so much an argument for Jackson as it is against the construction of our team.
I should go back and do Monta’s numbers from last year to see how he scored.
Thing 2
by olympicmike on Mar 6, 2009 12:06 PM PST up reply actions 0 recs
Yeah...
I get that now.
One thing that should be counted though, is that turnovers are far worse than missed shots. Also, you should get some credit for drawing a foul, but that’s not a tracked statistic, so you’d have to reference JAE’s database.
"No no Nene!"
http://www.youtube.com/watch?v=zB5DxNl4EB0
by Dubs fan in Boston on Mar 10, 2009 6:21 AM PDT up reply actions 0 recs
Its funny. How everyone can agree what the warriros need is
a power forward and quality point guard., Yet Tornoto has one of the best point guards and one the best Power forwards and they are worse than the W’s. What gives? Bad coaching? Overrated players. Distorted stats? After all someone has to score and some on has to get assists? Records don’t lie, but stats do.
by dungeness crabdribble on Mar 6, 2009 7:45 AM PST reply actions 0 recs
Stats very rarely lie. Stats almost always give a very, very good indication of team record and it’s true with Toronto.
Firstly, Toronto has a worse record than the Warriors. It’s close, but saying they’re worse than us is misleading and perhaps just plain false. Toronto has a very good PF and very good playmaker. They are then ‘complemented’ with garbage. Bargnani had a great shooting January, but came back to Earth. On Earth, in every other month, he was a big guy who didn’t shoot well and was perhaps the worst rebounding big in the game. He was more than not productive but actually costs his teams wins vs using a replacement player. O’neal, a shell of what he once was, wasn’t a whole lot better. Moon was adequate and was traded to Miami for Marion, who similarly hasn’t reached the level of production he once did and remains pretty average. Beyond that? Nothing. Little or no bench contributions, little from the off guard spot and a negative at whatever big position Bosh isn’t in equals not very good.
They’ve got a very, very weak team after their top two.
by jae on Mar 6, 2009 8:18 AM PST up reply actions 0 recs
Note: sentence: "Firstly, Toronto has a worse record than the Warriors " should have ended with a question mark.
by jae on Mar 6, 2009 9:22 AM PST up reply actions 0 recs
ehh
its intresting to real fans but most people wont care or understanh how and what it is
trade Jackson!!!!!!
by gswfan13 on Mar 6, 2009 9:29 AM PST reply actions 0 recs
0
and i dont no y your suprised with cj he is obviously the best gaurd on the team along with monta when hes healthy again
trade Jackson!!!!!!
by gswfan13 on Mar 6, 2009 9:30 AM PST reply actions 0 recs
i dont no y your suprised
I dont no y you’re grammar be bad
Thing A
by sam23 on Mar 6, 2009 7:12 PM PST up reply actions 0 recs
by your calculation
lebron and cj r almost equal. That alone should tell you the formula needs to be tweeked
by tafkasam on Mar 6, 2009 11:22 AM PST reply actions 0 recs
Really?
I guess it depends how you read it. LeBron is a +5.7 and CJ is a +2.7 which given the rest of the results is pretty significant. The ratios are similar but if you look at the huge number of positive outcomes for LeBron I think that the stat shows that he is a much bigger part of the game. The higher the raw positive outcomes the harder it would be to keep your ratio up because as you become a central part of your teams offense you also become a central part of the oppositions defensive scheme.
Remember this isn’t designed or intended to measure any of the other ways that LeBron helps his team get wins, but even without factoring in those things I think these numbers show that he is miles better than CJ (as they should).
Thing 2
by olympicmike on Mar 6, 2009 12:39 PM PST up reply actions 0 recs
long tradition among the coaching fraternity
OM, you’ve stumbled upon a method coaches at all levels of the game have been using for decades – particularly in the Bay Area. It may have begun with a local legend by the name of Bud Pressley – manic coach who taught man-to-man defense and finished his career at Menlo College. If so, it was expanded and elaborated by his long time assistant and head coach in his own right at San Mateo HS, Gus Hassapakis. Here’ how they motivated their players:
first keep stats on the following, then award the point-value for each – here’s what they awarded and rationale for why:
personal foul: -1 (allowed other team the ball, and eventually the bonus FT)
FG made 2/3 (obvious)
FG miss -1 (to emphasize good shot selection)
FT made +1
FT miss -1
Off Reb +2 (to motivate going to O-boards)
Def Reb +1
If total Reb > 10, +1 (obvious motivator)
Assist +1
turnover -1
Gain Possession +1 (steal, 5 second call, etc.)
Charge +3 (obviously high to motivate)
Block +1
At the end of the game players would look at their overall score, not the scorebook, and Hasapakis would start the players with the highest plus scores (or lowest negative) in the next game. They were very motivated and he could use this system regardless of total points scored. Obviously Offensive rebounds and charges were most highly valued even in non-shooters. Hassapakis explained that the Off reb was the most important play – it was a second possession, usually close to the basket so scoring would be most likely, and was most likely to force the defenders to foul to stop the put-back which gained fouls on other team’s players and got your team into the bonus. Later Hassapakis added a category of “hustle plays” which was totally subjective and usually involved a player diving for a loose ball, etc. Hence these “stats” weren’t truly objective, but they were truly motivational which is what these and many other coaches wanted.
by hardcore on Mar 6, 2009 10:24 PM PST reply actions 1 recs
Nice post. Thanks for the lesson. As you pointed out, it’s both a way to track and a way to motivate. I think OM is trying to simply find a way to track, but you point out that his thought process is sound.
by jmaaan on Mar 9, 2009 9:48 PM PDT up reply actions 0 recs
thanks, I was impressed that Hassapakis didn’t care as much about using a set lineup as he did keeping his own players motivated and hungry to compete within the team – plus opposing coaches had a more difficult time scouting and preparing for his team because his opening lineups varied. He did start some games with strange lineups as a result occasionally, but it worked for him overall. Stats in and of themselves are not always telling, but using them directly to motivate seems like a perfect application …
by hardcore on Mar 10, 2009 3:21 PM PDT up reply actions 0 recs
Yeah...
I meant to say this earlier but that’s a really interesting story that I hadn’t heard before. I can imagine a whole team of guys crashing the offensive glass and flopping in the lane to try to get those +2’s and +3’s. Haha.
Seriously though it seems like a cool idea. You could adjust the formula if you noticed your team was weak in one area.
Thanks for posting that.
Thing 2
by olympicmike on Mar 10, 2009 3:37 PM PDT up reply actions 0 recs
i voted no but...
i think it can be a semi useful measure for guards, especially point guards but its not for big men whatsoever. also, i guess “playmaker” isn’t the best description of this metric. it’s more like who is the “safest and least risky” player in terms of the fewest mistakes made.
to me its not surprising that CJ scored well on your metric because he usually is very sound in his decision making and doesn’t try to force the issue. he takes mainly open shots he is comfortable with. this doesn’t really make him a playmaker, just a mistake free player. this is the same reason why jose calderon scored so high. he has high assist to turnover, shoots shots he is comfortable with (like CJ) and shoots extremely well from the line.
i agree with olympicmike—you should go back and check monta’s numbers from last years with your metric. i would imagine it would be pretty high considering he shot a very high field goal percentage, attempted a low number of threes, had almost a 2:1 assist:turnover ratio, and had got to the line a solid amount of times (and shot fairly well from the stripe). if i had to guess i would think it would be around or above a ratio of 1.5.
all in all, it is an interesting calculation, but not useful for big men and determining if a player is a “playmaker”. it should be “safest”, “mistake-free” rating. just work on it a little more and i would gladly change my vote. great start though.
by gogoldenbears on Mar 8, 2009 4:49 PM PDT reply actions 0 recs
sorry olympicmike
i didn’t realize you were the the OP and the person who suggested the review of monta’s stats last season
by gogoldenbears on Mar 8, 2009 4:50 PM PDT up reply actions 0 recs
Monta 07/08 = +4.05
By my hasty math, anyway. Same range as Baron, a lot better than this year’s Marco or Crawford … which sounds about right.
Thing 1
by Sleepy Freud on Mar 8, 2009 9:36 PM PDT up reply actions 0 recs
Thanks for the feedback...
I agree with your assessment. I included one big-man in the original list to show that even the best bigs won’t score well (though I should have explained that).
I included all three aspects (PO/NO, Playmaker Rating and Playmaker Ratio) because I think that when you use all three it is more effective. If you just look at the ratio you will see that the most efficient (safe) players are more highly rewarded. If you look at the Rating you get a better idea at a players impact on the game. But even then you really have to see the entire number of outcomes to get the whole picture.
It’s definitely limited as far as it’s usefulness but I think it’s been very successful (to me at least) in judging guards and wing players as far as what they can accomplish when you put the ball in their hands and ask them to create or facilitate on offense. I think that it’s no coincidence that our player with the best Playmaker Rating (Monta, with the only respectable rating pre-injury) being out has really hurt the team. CJ has done a commendable job with this part of his game but he just doesn’t produce the volume of positive outcomes that a better player would. The thing to look at is his raw number of positive outcomes per 36 minutes. He only manages ten which means he isn’t nearly as involved when he is on the floor. I suspect that if he got more involved and forced the issue his ratio would fall a bit.
Thanks to everybody who commented. I have been pretty busy lately so I haven’t really been able to work on this but I might compile some more numbers next week.
Thing 2
by olympicmike on Mar 8, 2009 10:23 PM PDT up reply actions 0 recs
my bad
1. yea i kinda assumed you realized this was the case based on the way you evaluated your metric and the players you chose. i could tell that dwight was just a reference point
2. you’re right, i looked much more at the ratios than at the cumulative number. it is absolutely more useful as a tool when you combine both the cumulative rating along with the ratio. my bad and thanks for the heads up.
haha and i dont think you wasted your time with it at all. it definitely looks great. i apologize i dont have the concentration, motivation and knowledge of sabermetrics of a lot of great GSOMers to actually help rework this, i can only offer qualitative feedback…
by gogoldenbears on Mar 9, 2009 11:37 AM PDT up reply actions 0 recs
So my thoughts:
The only thing that really matters in evaluating if someone is doing something well in the game is whether or not it improves the team’s chance of winning. An effective “playmaker” rating should address this. Now as I understand it, this is a rating that tries to look at how an individual player’s contribution affects a particular aspect of the offensive end of the court, so regression against wins wouldn’t necessarily be the way to do it, but perhaps regression against two factors we know to favor winning would. Specifically, this is a rating to see how a player’s efforts increase his team’s FG efficiency (either by shooting well himself or by setting up teammates) or by controlling the ball and limiting turnovers. How you want to rate those factors is somewhat debatable, but that’s what has to be at the heart of a playmaker rating, else it’s another of the many statistical beauty contests. (And I’m not saying by any stretch that this is what you’ve done)
by jae on Mar 9, 2009 3:15 PM PDT reply actions 0 recs
Thanks...
I appreciate you taking the time to respond. That is the big question. Does it correlate with wins? I’m not sure. It definitely rewards efficient offensive play over “volume” players, but my biggest question is how helpful is an assist?
Obviously something you did contributed to your team earning at least 2pts, but I’m not sure if it’s a really good measure of creating an advantage on offense. I mean if you drive the lane, bring a second defender and dump it off to a big-man at the rim you’ve obviously created an advantage and it should count just as much as if you’ve scored yourself. Of course you could just be the last man rotating the ball to an open three point shooter in the corner. He could be open because of a double team in the post or dribble penetration that caused the defense to collapse. If that’s the case you shouldn’t really get as much credit as the player who really created the advantage.
Interesting stuff, but I’m clearly in over my head at this point. It was a fun exercise and it did just about what I expected it to, but it doesn’t tell me anything that I didn’t know by looking carefully at the stats that I used to create it. If anything it would be more convenient than looking at FG%, FGA/36min, FTA/M, TO’s and the like but I don’t know if anyone would really find it helpful. It’s kind of like a more complicated assist to turnover ratio. For whatever reason some people seem to find that more useful than looking at assists per minute and turnovers per minute which would actually give you a better idea of what is really going on.
Anyway thanks again.
Thing 2
by olympicmike on Mar 9, 2009 3:45 PM PDT up reply actions 0 recs
I liked it. I would love to have a playmaker rating to judge potential guards with. Especially when we are in bad need of a playmaker and CP3 is oddly not available.
I had a couple small concerns- For instance, basketball really suffers when analysing stats from the fact that an assist only comes from the last pass before the basket. As many have said before me, hockey has perhaps a better way of measuring assists. If a players drives and draws the defense, then kicks out to a perimeter shooter who makes the one extra pass to a wide open shooter, the guy who started it all and set it up doesn’t get any assist. This lends credence to jae’s argument about fg. If you have an impact in increasing your teams fg, that can be more telling that assists.
If you’re really looking at “playmaker rating”, it would make sense to look at some of the all time great playmakers- Magic, Kidd, Stockton, even Nash, and see what their rating would have been vs. guys who are just scorers, or who are marginal players but efficient, like CJ. I think you attempted to do this, but perhaps more backtesting can help with tweaking for fit.
by jmaaan on Mar 9, 2009 10:07 PM PDT up reply actions 0 recs
Of course you could just be the last man rotating the ball to an open three point shooter in the corner. He could be open because of a double team in the post or dribble penetration that caused the defense to collapse. If that’s the case you shouldn’t really get as much credit as the player who really created the advantage.
That player still made the decision to pass the ball to the right scoring opportunity rather than trying to force something else. Finding the open man should be rewarded, no matter how that man got open.
"No no Nene!"
http://www.youtube.com/watch?v=zB5DxNl4EB0
by Dubs fan in Boston on Mar 10, 2009 6:18 AM PDT up reply actions 0 recs
That player still made the decision to pass the ball to the right scoring opportunity rather than trying to force something else. Finding the open man should be rewarded, no matter how that man got open.
I do agree. My point was that he did less than the player that received no credit for the play. I’m not saying that the guy shouldn’t get an assist, I was just pointing out that it is an imperfect measure of creating scoring opportunities for other players on your team. Big men for instance can create tons of open shots on the perimeter without actually passing out of double teams directly to the open man. They are in a sense making a play just by drawing the attention of the defense.
I would be interested to see if anyone keeps a “double teams drawn” stat. I’ve never seen it but it could be interesting to look at.
Thing 2
by olympicmike on Mar 10, 2009 11:24 AM PDT up reply actions 0 recs
I would be interested to see if anyone keeps a "double teams drawn" stat. I’ve never seen it but it could be interesting to look at.
I don’t know of anything close to this. It’s not a tracked stat and I cannot see any way to get it from the presently charted data.
There do seem to be players who open up things for other players with their ability to draw double teams, but most players’ FG% seems largely independent of who else is on the court with the exception of the PG. I am not saying there isn’t some of it going on, but it’s a very noisy dataset if there is something going on that’s predictable and has significant impact. The improvement (or decline) in a team’s FG% with most players is simply the isolated effect of that player’s shooting. (e.g. Shaq has long demanded double teams, drawing defenders and would seem to present a great case, but what I’ve mostly noticed [disclaimer: I have not run a comprehensive analysis at this point, but done some preliminary crunching] is that the FG% advantage he brings to a team is more or less identical to the difference between his shooting and the shooting of his backup. He adds made baskets at his position and takes more shots than his backup per minute, but the other 4 guys shoot about as well with or without him. The complex interactions seem to be far less pronounced than the simple ones to the point that the straight analysis “put in a better shot maker and your shooting gets better” gets you the right answer.)
Anecdotally, the effect seems to be even less detectable for those big men who shoot from the perimeter and “draw their man” out of the lane. Might open up things on occasion, but it doesn’t appear to have an impact often enough to notice regularly. (And as far as I can tell, is negated by the effect that the inside shot is almost always more efficient than the outside shot that ‘drawing someone out’ often results in. Couple this with the decline in offense rebounds this seems to result in and a big man’s ‘range’ seems to be pretty low on the list of import assets for a big man.
by jae on Mar 10, 2009 12:15 PM PDT up reply actions 0 recs
Great stuff...
Thanks for the info. From what your saying it sounds like drawing a double team isn’t nearly as important as just being a good enough player to draw a double team.
I wonder if that speaks to the effectiveness of using the double team as a defensive strategy. If getting double teamed doesn’t really help your teammates all that much then as a coach I would probably be a little more quick to double somebody who is hurting me.
Interesting stuff…
Thing 2
by olympicmike on Mar 10, 2009 12:37 PM PDT up reply actions 0 recs
it sounds like drawing a double team isn’t nearly as important as just being a good enough player to draw a double team.
That’s about the size of it as it seems, but of course, players who don’t deserve double teams don’t get them so it’s not really a controlled experiment.
by jae on Mar 10, 2009 6:07 PM PDT up reply actions 0 recs
double teams and double attention
one of the very difficult things to measure quantitatively is how a player forces the defense to adjust hence leaving other teammates greater opportunities – thinking here of Baron, or of Monta when he’s driving to the rim, as the defense cheats to rotate early to stop penetration they they’re a step further away or a step late rotating to help inside or out to the 3pt line. … the guard/wing assist stat would probably be the best single stat, but sometimes the ball rotates through 2-3 players first – difficult to stat that … for a brief time, Nelson seemed to embrace a shift from isolation mismatches to a “Euro style” of drive, kick, drive etc. to force the defense to react and help just to create this exact scenario – unfortunately we didn’t stay healthy enough to see it mature. Maybe next year? <- a new mantra
by hardcore on Mar 10, 2009 9:08 PM PDT up reply actions 0 recs

by 





















