Preseason Game 3 McHale vs Taiwan

The first most important thing I want to address is the plus/minus.  I hope this won’t become a trend where I start discussing plus/minus after every game.  But, I found it very awkward that Lin ended up with a -5 and Bev a +17.  Was there a stat keeping conspiracy?  If you recall my blog post for the Manila game, Lin ended up with a very conspicuous plus/minus too.  But, that was explained after reviewing the game.  So, I reviewed the Taiwan game and solved this game’s mystery plus/minus as well as sharing a few nuggets I found too.

Let me first start with a discussion on stats.  So, we know that plus/minus is a very misleading statistic and not sure why it’s used by all the major sports sites.  But, it’s there to annoy me.  Let me go on a tangent for a second and state that I hated statistics in school but I love trend analysis, data crunching and cost accounting/margin analysis.

Basketball stats to me is more trend analysis than actual statistics.  It’s very unlikely that you can isolate the variables that the statistician can use to predict future events as you would in statistics because there’s just too many variables in basketball from having 10 players on the court each with unique abilities/skill sets to playing with and against different management styles and strategies.

So, with basketball advanced stats, all we’re really doing is taking historical data and lay them around differently in a cost analysis manner and try to come up with key performance indicators that can be used to easily evaluate a player’s effectiveness and efficiency.  And, today, I will discuss of two these stats, efficiency rating (EFF) and player imagine estimate (PIE) ratios by applying them to the Taipei game. Why? Because the plus/minus of the past few games really annoys me.

Efficiency rating (EFF) essentially  measures how effectively a player can contribute towards statistics that is generally perceived to be favorable as well as unfavorable within a limited number of attempts.  To get a high EFF, you just have to get as many points, rebounds, steals and blocks as possible while taking as little number of shot attempts as possible.  In other words, only act when you know the rate of success is high.  At the same time, you cannot not do anything or you will end up with a low EFF.  In other words, damn if you do, damned if you don’t. Formula is:

((Points + Rebounds + Assists + Steals + Blocks) – ((Field Goals Att. – Field Goals Made) + (Free Throws Att. – Free Throws Made) + Turnovers))

Also note that if you make a free throw you gain one point while you lose one point if you miss a free throw.  This goes against common logic that you want to shoot as many free throws as possible assuming your FT% is higher than your FG% which is everyone not named Dwight Howard.  This goes against common sense because the player gets more points by shooting 3 pters and 2pters but is only deducted a single point for missing them, but these are considered lower percentage shots than free throws.  So, if I were a statistician, I would definitely reduce how a player is punished for missing a free throw.

If players were like investments, as Morey can attest, you want players with high EFF because that player can get you the most returns within a set number of opportunities.  In the finance world, I figure this is equivalent to return on investment or contribution margin percentage.  In other words, for each unit of input or attempt invested or sold, you should get an expected unit of output or return.  Capiche?

Jeremy had the highest EFF in the game Taipei game tied with Danny Granger.  As you can see below, Jeremy’s scoring is more efficient while Danny’s non-scoring stats were more effective such as rebounds and causing less personal fouls and turnovers.

So, are these two players equally as effective and efficient in contributing positively to your team?  If you were the coach/gm, who would you rather have on your team?  Well, there is no right answer haha.  It all depends on how you want to use this stat.

If I were coach, I’d stop Granger from shooting more FGs because he’s shooting 25.0% from the field.  Had he taken 4 less shots in the game, his EFF would’ve increased by 3 points.  Would I do anything different with Jeremy?  Don’t play him when he’s exhausted.  2 of his TOs were committed when he was playing on fumes late in the game.  So, that was more of a coaching error.  And, there were at least a couple of bogus calls against Jeremy which resulted in him having a higher number of PFs.

What else does EFF tell you as a coach/manager?  A player can acquire a high EFF without having to be a high scorer.  But, how can one quickly assess who are the players most capable of contributing to the team without shooting disproportionately high number of FGs?  Simply divide EFF by points.

Omri Casspi all of a sudden is startsto look like a stud!  He was able to produce a very high EFF of 20 by not only shooting very efficiently but also by being efficetive without the ball by grabbing 9 boards and by keeping his PFs and TOs down.  In other words, he’s got the stats of both Lin’s and Granger’s efficiency and effectiveness both with and without the ball in his hands. Sounds like a Scottie Pippen?

So, if you were coach/manager, who would you rather play for you, Granger, Lin or Casspi?  Well, based on the EFF/PT stat, I wouldn’t mind having 5 Casspis on on the floor at the same time as he seems to excel in a more well rounded manner than both Lin and Granger did.  (Please note I removed Brewer and Beverley from the analysis because their points were too low as I removed players who scored less than 10 points from this analysis.)

Let’s confirm my assessment regarding how much impact each of these players had on the Taipei game with the… Player Impact Estimate ratio. This stat provides a data point regarding the percentage of the games total positive net negative contributions a player’s activities contributed that are deemed positive or leading to a team win. However, essentially this is just a weighted average of total player’s contribution of activities acribed by the PIE ratio’s calculation:

[Points + (Field Goals Made+Free Throws Made - Field Goals Attempted - Free Throws Attempted) + Defensive Rebound + Offensive Rebounds / 2 + Assists+Steals + Blocked Shots / 2 - Personal Fouls - Turn Overs] / [Game Total Points + (Game Total Field Goals Made + Game Total Free Throws Made – Game Total Field Goals Attempted – Game Total Free Throws Attempted) + Game Total Defensive Rebound + Game Total Offensive Rebounds / 2 + Game Total Assists + Game Total Steals + Game Total Blocks / 2 – Game Total Person Fouls – Game Total Turn Overs

Again, same as the EFF ratio, the PIE ratio doesn’t seem to value free throws. But, the PIE ratio also doesn’t seem to value Offensive Rebounds as well. At least, according to the calculation, offensive rebounds are half as valuable as steals, assists and defensive rebounds. Isn’t this kind of strange? I find offensive rebounds to be more important as this gives the team more second and third chance scoring opportunities off misses. So, I think players should be awarded more for their offensive rebounding contributions. But, anyways, with that said, lets see how the players compare in the Taipei game through the PIE lens:

pieLin has the third highest PIE% of both teams although Lin has the highest game share in points, assists and steals of the three as indicated in chart below.  Granger and Casspi got higher PIE% for two reasons.  One, they both got a lot of rebounds which Lin only got 2.  And, Lin’s 5 personal fouls and 3 turnovers worked against Lin’s PIE% while Casspi and Granger were not penalized much for these two categories.


OK Stats aside – lets comment on some memes:

First of all, I want to make some comments about what I’ve read this morning after I woke up and got onto the internet.  I listened to Paul on Conservative Media News Network professing his man crush to Jeremy on youtube.  The title is called Linsanity Reborn and Paul enthralled about what a magical night it was and Jeremy along with his fans totally deserved it.  My thought it that was not Linsanity.  Lin got 17 points with his 34.2 minute playing time while shooting extremely efficiently making 6 of 8 shots.  That’s not Linsanity.

Even during the regular season last year, Lin averaged 11 shots a night.  Though his average points per game is down at 13.6, his season average for assists is actually higher at 6.1 compared to last night’s 4 assists.  Now, I think this is an important point because what made Linsanity so special is not just Jeremy but also his crew/bff/teammates who were playing with Lin, hitting their spots, covering their man, playing the epitome of team ball from the jump ball until the final seconds run out.  That didn’t happen last night.  Lin didn’t do anything special.  His teammates didn’t do anything special.

So, why am I making such a big deal out of this?  For awhile now, there’s been articles released by Houston Rockets media et al that has been saying things to the effect that Linsanity is transitory and that Lin is a work in progress as if Linsanity wasn’t something that’s supposed to happen.  So, because of this and other articles of the same ilk, I am extra sensitive to people saying Linsanity is “magical” or “transitory”.  Anyways, that’s my spiel.  End of post/blog.

End meme analysis and beginning of boring +/-analysis.  I actually started the EFF and PIE analysis because of the +/- is so outlandishly NOT explaining how players were doing in the game.  So, after the euphoria of watching Lin’s Houston pulling another W away from the Pacers, I was troubled by why Lin got a -5 plus/minus while Beverley got a +17.  I went to bed thinking HOW IS THIS POSSIBLE?  Didn’t Lin just dominate the entire time he was on the court?  I know the plus/minus stat is not an accurate indicator of performance but that’s a huge variance compared to Lin who out performed everyone and Bev who didn’t get his first FG until the last few minutes of the game!  So, off to bed I went.

And, first thing in the morning, I… did my chores cleaning the house a bit, did some exercise, grocery shopping and got some game snacks to REWATCH the WHOLE GAME again.  And, THIS TIME, I am taking NOTES specifically track how in the world did the plus/minus stat get so illogical.


Q1 Line up – Lin, Harden, Howard, Parsons, Jones

7:50 – Harden is picked by Hibbert leaving George wide open for a 3fg.  Howard should have helped out on the PnR instead of just standing in the paint.

5:48 Hibbert sets another pick this time for Hill but he bumps into Lin and is called for a moving pick.  Again, Howard didn’t help out on the pick but just standing in the paint.  At least he got his arms up this time.  For what, I don’t know.

4:28 Lin’s first TO.  A wide pass to DH near the rim.

3:50 Harden brings the ball up and holds onto the ball running 10 seconds off the shot clock before he passes to Howard.  Howard then holds onto the ball for about 10 seconds before he makes his move.  By now, 3 players have collapsed on him and blocks Howard’s shot.  Howard did get his own rebound and got the assist for passing the ball to Dmo who was under the basket.

3:40 Lin called for charge for running into Hill.  But, Hibbert, who was setting a pick for George, pushed Lin into Hill. So, Lin got blamed for charging when he was the one who should have received the benefit of the call.  And, again, Howard didn’t help out on the PnR defense.

3:31 Lin is subbed out for Beverley.  Lin has a +1 at this time.

0:58 Lin comes back in for Beverely.  Beverley leaves with a +4.

0:52 Lin drives and gets the tear drop to fall but is fouled hard by Hibbert.  Lin didn’t get the call though.

End of Q1 – Lin +4 Bev +4

Q2 Line up – Lin, Beverley, Dmo, Smith and Garcia

6:45 Harden drives and runs into Copeland.  Copeland is called for the foul and Harden gets free throws.  Copeland looks like he was still moving sideways when contact was initiated.  But, after reviewing this sequence, I see Copeland had stopped barely before the contact.  And, Harden lowered his shoulder to go into Copeland’s chest while pushing his knee into Copeland’s thigh.  He then loses the ball, puts his left arm under Copeland’s right arm and starts to do his rain dance.  This sequence of motion causes Copeland to look like he’s still moving at point of contact but he was pushed side ways by Harden’s shoulder, knees and arms.

Harden Dance 1Houston’s offense turns it on during the remaining second quarter with Parsons, Casspi, hitting 3s and Harden, Howard getting some FGs.

End of Q2 Lin +14 Bev +7

Half Time

Q3 line up – Lin, Harden, Howard, Parsons, Jones

This is the very frustrating quarter.

To make a long story short, third quarter was about trying to get Howard involved in the game.  During the first two quarters, passes were not getting to Howard’s sweet spots.  So, Howard was given the opportunity to create his own shots.  Here’s where Howard tried to take on 4 Pacers defenders on his own by forcing up tough lay ups and awkward bankshots that never found the bottom of the net.  The Pacers took advantage of this Howard experimentation and decreased Houston’s lead down to to 2 points.  And, everyone who were on the floor with Howard at the time saw their +/- stat take a huge hit.

  • etane

    Note from 10/13 moved here: I left my game notes at work because I thought I’d have time to finish up this blog post but got really busy and just discovered I left my notes at work now that I am at home ready to finish this blog post up. It’s a blessing really because I think I was getting too long/technical with my blog post. This is an opportunity to write off memory and forget all the details. I will fill in the details in the game chat forum instead of letting it take over the blog post. Boy, I thought I was going to shorten this blog post but it’s getting
    even longer = D. And, I need to get some sleep. TBC’d tomorrow.