Fun with NBA Wowy: Is Jaylen Really This Bad?

[PhoSita]

https://prod-nbawowy-frontend.addmorefunds.com/

If you haven't tried this free online tool before, I suggest you check it out.  You can learn some intriguing things about on/off numbers.

You can see, in the minutes that a player (or combo of players) has been on or off the court, how the team has performed.


Celts so far this season: 33 games ... 1548 minutes

111.3 ORTG ... 104.7 DRTG .... +6.6


Kyrie Irving:  +9

999 minutes on the court ... 116 ORTG, 106 DRTG

604 minutes off the court ... 104 ORTG, 103 DRTG

Tatum: +9

1035 minutes on ... 115 ORTG, 105 DRTG
568 minutes off ... 106 ORTG, 105 DRTG


Al Horford ... huh?  -6

709 minutes on ... 110 ORTG, 107 DRTG
894 minutes off ... 112 ORTG, 103 DRTG


Kyrie + Tatum + Horford .... +4

469 minutes on ... 117 ORTG ... 105 DRTG
259 minutes off ... 105 ORTG ... 97 DRTG

Gordon Hayward ... the problem child? ... +1

776 minutes on ... 113 ORTG ... 106 DRTG
827 minutes off ... 110 ORTG ... 104 DRTG


Jaylen Brown ... the real problem child?  -10

746 minutes on ... 105 ORTG ... 104 DRTG
857 minutes off ... 117 ORTG ... 106 DRTG


Gordon Hayward + Jaylen Brown .... -10

385 minutes on ... 102 ORTG ... 101 DRTG
466 minutes off ... 112 ORTG ... 101 DRTG




It's hard to draw any too-strong conclusions from these numbers.  On/off is subject to a lot of noise, plus these aren't terribly huge samples.

The team seems to have performed quite well without Horford on the court, which is surprising.

Kyrie and Tatum are clearly pretty important to the team's success, as the team does very well with them on the court and is just above breaking even without them.

Gordon Hayward isn't as obvious a source of the team's issues as you might think, since the team has done pretty well with him and not super great without him.

Jaylen Brown, on the other hand, looks like a problem -- a big problem -- the team has been great without him and mediocre with him.


But some of this is as you might expect -- the team kills with the starters in and the guys who mostly come off the bench get killed. 

Yet, the Celts are a team that at times has seen its starting lineup struggle and had the  bench bail them out.


So what do you make of these numbers?  Play around with the NBA Wowy page and look for your own nuggets of potentially interesting info.


Is Jaylen Brown the worst player on the Celtics?

[mmmmm]

But some of this is as you might expect -- the team kills with the starters in and the guys who mostly come off the bench get killed. 

Careful with that.  The starting lineup hasn't exactly been static.  Morris and Smart started the season on the bench and now are in the starting lineup whereas Jaylen and Gordon have flipped the opposite way.   This changes the opposing "vs bench" / "vs starters" for those guys.

Also, the opponents we have faced has changed a lot from the beginning weeks of the season, when we faced a lot of tough teams to more recently when we faced a much easier schedule.   And beware of how much these numbers can be skewed by just a few games that were insane blowouts.    You noted your surprise regarding Al Horford's numbers, but you need to keep in mind that his "OFF" numbers include the 133-77 shellac'ing of Chicago that massively padded the numbers for the guys who did play in that game.

There is no simple breakdown that creates good comparative samples without turning into ridiculous cherry picking and small samples, unfortunately.   And collinearity effects can be very strong.

Consider for a moment that there is a bit of a "Mook Effect" so far in this early season due to just how blazing hot Morris has shot the ball (63% TS%!!!!) through these first 33 games.   Not only does that mean that Morris has nice numbers, it also tends to benefit anyone who has shared the floor with him. 

For example, Kyrie has shared the floor with Morris for some 515 (51.6%) of his minutes during which the team has scored at a 121.8 ORtg pace versus giving up just 108.6.  That net of +13.2 is a significant boost over Kyrie's other 485 minutes, which have resulted in 109.9 ORtg & 102.7 DRtg.   Similarly, Smart has played 498 (58.6%) of his minutes with Morris on the floor, resulting in a nice 116.9 ORtg against a 107.8 DRtg, compared to 109.9 & 103.2 without Morris.    Both have positive numbers without Morris on the floor, but it's a significant drop in net rating compared to with him 'on'.

[jambr380]

One thing this does show is that Jaylen and Hayward do not play well together (not surprising at all). I am sure when Hayward went to the bench, this was a relief to Jaylen...but then he also went to the bench!

I realize Hayward is a much better player than what he is showing now, but I am afraid that he is in Jaylen's head (and maybe even vice-versa). With Jaylen's pre-season comments about guys not wanting to step on each other's toes to his latest comments about not taking anything personally after the team meeting, the dude is just not able to find his role on this team.

I honestly believe if Hayward were not here at all, Jaylen would be playing at least as well as he did last year. The two are subconsciously sabotaging each other.

[PhoSita]

Great points, M5.  On / off numbers are eye popping, but easy to distort.

I'm mostly surprised that the team hasn't been terrible without Horford.  There's no doubt in my mind that Horford is really important to the team's success.


The Jaylen numbers concur with what the eye test and the basic stats say this year, which is that Jaylen has been bad, and his bad play has contributed to the team's struggles (to the extent they've struggled).


From what I've seen playing around with these on/off numbers, you can do some weird stuff by making combos of certain players on and/or off, but any combo with Jaylen seems to show the team playing at or around break-even with him on the court.


Generally, though, it's a good sign that it isn't easy to find combinations of players with truly terrible on-court numbers.


Theis and Rozier, for example, are both pretty clearly bench guys.  They've only shared the court for 261 minutes so far this season, but in those minutes, the Celts have a 113 ORTG and a 97 DRTG -- in other words, they're killing opponents.

[PhoSita]

I realize Hayward is a much better player than what he is showing now, but I am afraid that he is in Jaylen's head (and maybe even vice-versa). With Jaylen's pre-season comments about guys not wanting to step on each other's toes to his latest comments about not taking anything personally after the team meeting, the dude is just not able to find his role on this team.

I honestly believe if Hayward were not here at all, Jaylen would be playing at least as well as he did last year. The two are subconsciously sabotaging each other.

FWIW ....

Jaylen Brown ON & Gordon Hayward OFF

361 minutes

108 ORTG
107 DRTG


Gordon Hayward ON & Jaylen Brown OFF

391 minutes

123 ORTG
111 DRTG


The team seems to perform really well offensively in the minutes that Gordon has played without Jaylen (though awful defensively).

It doesn't look like things have been much better when Jaylen has played minus Gordon, though.

[mmmmm]

Great points, M5.  On / off numbers are eye popping, but easy to distort.

I'm mostly surprised that the team hasn't been terrible without Horford.  There's no doubt in my mind that Horford is really important to the team's success.

I think all the factors I noted above come into play there.  We had a couple of blow-out wins against really bad opponents while he was out.  Mook has been hot shooting and doesn't depend much on Horford's play-making like some of our other guys.  And the schedule has been easier for most (not all) of the games Horford has been out. 

Certainly in the last 3 games without Horford, the team has not been very good.

The Jaylen numbers concur with what the eye test and the basic stats say this year, which is that Jaylen has been bad, and his bad play has contributed to the team's struggles (to the extent they've struggled).

Jaylen's been just dying on wide-open threes (shots with no defender within 6 feet).  This is a type of shot that non-primary scorers have to hit at a high rate to make up for the much-lower percentage they hit contested shots at.  Last year Jaylen hit those at a decent 43.8% clip.   This year, he's hit them at a god-awful 25%.

He's also been not-great on merely "open" shots (defender between 4-6 feet) at just 28%, but that's actually not terrible and statistically that sort of number doesn't kill your average if you get a decent share of wide open shots and hit them at a high rate.   But the latter just hasn't happened.   So his whole 3PT% is just dragging.   And that is killing his efficiency and hurting his whole game.

He has literally been far, far more efficient in games where he has ignored the 3PT shot and gone exclusively for shots inside 16 feet.   Overall, he's actually more efficient inside 16 feet this year than last year.

Of course, the numbers are still very tiny.  He has a total of just 107 3PA on the year.  He could hit a small hot streak and in just a handful of games suddenly bring  his shooting percentages right back up to last year.   So it probably makes no sense to read too much into the numbers yet.

EDIT:  Corrected Jaylen's wide-open three rate for 2017-18 to 43.8%.

[playdream]

I'm surprised this numbers match my eye test perfectly
AL has been pretty bad, Gordon is OK(good if you consider the comeback)
Tatum and KI is balling
JB..i'm losing words on him so i will leave it here

[PhoSita]

Jaylen's been just dying on wide-open threes (shots with no defender within 6 feet).  This is a type of shot that non-primary scorers have to hit at a high rate to make up for the much-lower percentage they hit contested shots at.  Last year Jaylen hit those at a decent 41.3% clip.   This year, he's hit them at a god-awful 25%.

He's also been not-great on merely "open" shots (defender between 4-6 feet) at just 28%, but that's actually not terrible and statistically that sort of number doesn't kill your average if you get a decent share of wide open shots and hit them at a high rate.   But the latter just hasn't happened.   So his whole 3PT% is just dragging.   And that is killing his efficiency and hurting his whole game.

He has literally been far, far more efficient in games where he has ignored the 3PT shot and gone exclusively for shots inside 16 feet.   Overall, he's actually more efficient inside 16 feet this year than last year.

Of course, the numbers are still very tiny.  He has a total of just 107 3PA on the year.  He could hit a small hot streak and in just a handful of games suddenly bring  his shooting percentages right back up to last year.   So it probably makes no sense to read too much into the numbers yet.

To build off this, Jaylen's shooting percentages are pretty similar to last year with the sole exception of his 3P%.

Another thing that jumps out in his shooting stats is that the percentage of threes he's taken from the corner is about 20%, whereas his first two seasons he attempted 35-40% of his threes from the corner.

Jaylen's 3P% on attempts from the corner this year, amazingly, is only 19% (!) whereas the last two years he shot well over 40% from the corner.

As you say, it's a small sample, but that's still surprising.

[RockinRyA]

Its hard to quantify the minutes played and how they affect the team. For one, minutes played against the starters of the Sixers are not the same as the minutes played against the bench of the Bulls.

[RodyTur10]

I made a topic with observations based on on-off numbers as well with the same type of observations.

My opinion about Horford is that his numbers aren't that impressive for two reasons:

• Horford playes the majority of his minutes against the starters, thus the best opposition. That deflates the stats that show his overall impact.
• The line-ups with Horford aren't that good defensively, since Horford is undersized for a center. He should play PF besides Baynes or Theis like last year.

[mmmmm]

I made a topic with observations based on on-off numbers as well with the same type of observations.

My opinion about Horford is that his numbers aren't that impressive for two reasons:

• Horford playes the majority of his minutes against the starters, thus the best opposition. That deflates the stats that show his overall impact.
• The line-ups with Horford aren't that good defensively, since Horford is undersized for a center. He should play PF besides Baynes or Theis like last year.

I concur very much with this.  The last point especially.  Not saying that Horford should be used exclusively or even mostly at PF, but the change from last year is too extreme.  Last year, Horford spent about 45% of his minutes at PF and we were extremely successful in that configuration on both ends of the court.   This year, for whatever reason, we have used Horford at PF only about 6.4% of his minutes, playing him as the single big on the floor almost all of the time and this had has distinct negative consequences, in particular in pick & roll defense.  Things have looked better in the last few weeks, but that has largely been against inferior competition and has often been due to our offense wracking up huge points.

So far this season, the DRtg with Al on the floor has been 106.9 points per 100, which is good, but definitely below the team overall DRtg (104.7).

In the minutes with Al at the 4 and Aron at the 5, though, the DRtg has been a fantastic 91.8, which is far more reminiscent of last year.

When Aron is back, I would like to see both of them on the floor a little more often.  I think Al is going to get worn out playing exclusively at the 5 in a one-big defensive scheme.   I get that we want to deploy more small, wing & guard-dominated lineups for the offensive versatility, but what we've done has gone too far, imo.

[PhoSita]

Somehow Baynes - Horford - Smart have only shared the floor for 4 minutes all season.

Over that HUGE sample of 13 possessions, opponents scored 0.769 points per possession.

Clearly those guys should play more together

[RodyTur10]

Somehow Baynes - Horford - Smart have only shared the floor for 4 minutes all season.

Over that HUGE sample of 13 possessions, opponents scored 0.769 points per possession.

Clearly those guys should play more together

What this tells me is that Stevens is more focussed on the offensive gameplan than defense. He values shooting and passing more than rim protection and rebounding. Above all he rates versatility. Stevens wants his players to be able to switch on defense and shoot the 3-point shot.

To his credit Boston is an excellent defensive team (3rd ranked), a good 3-point shooting team (I rank them 7th, adjusted for FG% and volume) and a good passing team (4th AST/TO ratio).

However they're not very efficient overall on offense (only 18th TS%), this is because they're 27th in free throw attempts and even 29th in FTA per FGA. Boston isn't able to generate easy points at the line. This should be a focal point and less emphasis on 3-point shots.

All statistics during losses and wins are relatively the same ranked in comparison with other teams. Which is surprising.
In fact most stats are actually almost exactly the same. Boston doesn't have more turnovers when it loses for instance.

Why does Boston win or lose games? Answer: defensive rebounding.
When you select all teams and only include wins for all teams then Boston ranks 1st in DREB% !!!
But when you do the same for losses then Boston ranks only 27th in DREB% !

So there are two clear issues to fix. Generate free throws (attack the paint) and don't give up offensive rebounds.

As a screen setter and his ability to protect the rim I believe that Baynes can be very important in fixing both.

[PhoSita]

I agree that generating free throws is the heart of the issue for the Celts' offense.  They're always going to be reliant on three point shooting.  They have enough players who, at least in theory, are solid or better shooters.

What they need to supplement the outside shooting is to generate free throws more consistently.  They aren't going to bully teams inside with post-ups or offensive boards.

[SparzWizard]

Somehow Baynes - Horford - Smart have only shared the floor for 4 minutes all season.

Over that HUGE sample of 13 possessions, opponents scored 0.769 points per possession.

Clearly those guys should play more together

What this tells me is that Stevens is more focussed on the offensive gameplan than defense. He values shooting and passing more than rim protection and rebounding. Above all he rates versatility. Stevens wants his players to be able to switch on defense and shoot the 3-point shot.

To his credit Boston is an excellent defensive team (3rd ranked), a good 3-point shooting team (I rank them 7th, adjusted for FG% and volume) and a good passing team (4th AST/TO ratio).

However they're not very efficient overall on offense (only 18th TS%), this is because they're 27th in free throw attempts and even 29th in FTA per FGA. Boston isn't able to generate easy points at the line. This should be a focal point and less emphasis on 3-point shots.

All statistics during losses and wins are relatively the same ranked in comparison with other teams. Which is surprising.
In fact most stats are actually almost exactly the same. Boston doesn't have more turnovers when it loses for instance.

Why does Boston win or lose games? Answer: defensive rebounding.
When you select all teams and only include wins for all teams then Boston ranks 1st in DREB% !!!
But when you do the same for losses then Boston ranks only 27th in DREB% !

So there are two clear issues to fix. Generate free throws (attack the paint) and don't give up offensive rebounds.

As a screen setter and his ability to protect the rim I believe that Baynes can be very important in fixing both.

Funny, this team's problem has always been rebounding even since the Doc Rivers-Big 3 era days. An issue that needs to be addressed.