Rondo vs the other elite PGs: by the numbers

[Celtics18]

Adjusted plus/minus and our efficiency differential for the games KG/Rondo missed indicate that losing KG cost us more.

The Celtics have a couple of players that can fill in for Rondo's offensive role. BBD's play as a starter falls so far below KG's that we can't make it up.

I looked at the team's efficiency numbers and found that not to be true.

According to my research: in the games Rondo played where KG didn't, the offensive efficiency was above our season average, but the defensive efficiency was slightly worse.  In the games where KG played with no Rondo, the defensive efficiency was better but the offensive efficiency was worse than our season's average.  The differential stayed about the same, though.

Season points per possession averages:

Offense: 105.8, defense: 97.7, Difference: 8.1

Games in which Rondo played, without Garnett:

Offense: 108.9, Defense: 100.6, Difference: 8.3

Games in which Garnett played, without Rondo:

Offense: 102.9, Defense: 94.8, Difference: 8.1

[Celtics4ever]

Rondo isn't the best only a fool would argue that.   He shooting takes him out of contention, IMHO.  But he is very good and plays on a team with deeper talent than all those other PG.  He would score more on a worse team but would get less assists.   

One Rondo does that a lot of these better PG don't do and that is take over a series.  He was critical in us betting CLE and ORL last year.   He literally took games over.  A lot of these other studs vanish when the chips are down.   I think Rondo getting so banged up vs. ORL hurt us in the finals too.   He played but he was around 70% of himself if that.  I think the kid is money some of these other guys while statistically better are not when games matter.

[Celtics18]

Rondo isn't the best only a fool would argue that.   He shooting takes him out of contention, IMHO.  But he is very good and plays on a team with deeper talent than all those other PG.  He would score more on a worse team but would get less assists.   

One Rondo does that a lot of these better PG don't do and that is take over a series.  He was critical in us betting CLE and ORL last year.   He literally took games over.  A lot of these other studs vanish when the chips are down.   I think Rondo getting so banged up vs. ORL hurt us in the finals too.   He played but he was around 70% of himself if that.  I think the kid is money some of these other guys while statistically better are not when games matter.

Do you realize that Rondo's shooting percentage on mid-range jumpers is among the best in the league?

By the way, how can your opinion be "humble" when you say that only a "fool" would hold an argument that is contrary to yours.   

[Fafnir]

Adjusted plus/minus and our efficiency differential for the games KG/Rondo missed indicate that losing KG cost us more.

The Celtics have a couple of players that can fill in for Rondo's offensive role. BBD's play as a starter falls so far below KG's that we can't make it up.

I looked at the team's efficiency numbers and found that not to be true.

According to my research: in the games Rondo played where KG didn't, the offensive efficiency was above our season average, but the defensive efficiency was slightly worse.  In the games where KG played with no Rondo, the defensive efficiency was better but the offensive efficiency was worse than our season's average.  The differential stayed about the same, though.

Season points per possession averages:

Offense: 105.8, defense: 97.7, Difference: 8.1

Games in which Rondo played, without Garnett:

Offense: 108.9, Defense: 100.6, Difference: 8.3

Games in which Garnett played, without Rondo:

Offense: 102.9, Defense: 94.8, Difference: 8.1

That's interesting I'd thought KG would be slightly more based on the adjusted +/- and what I saw during those games. I looked at the numbers on basketball referrence and I got slightly different numbers than you did (but the same result)

I think schedule is the answer to what my gut said:

Without Rondo with KG
Record 4-1* (you could count the Detroit where KG got hurt in the 1st QTR I didn't)
C's Off Rating 102.1
C's Dif Rating 95.18
+6.92
SOS  .504
Average Opp. Off Rating 105.68
Average Opp. Def Rating 104.5

Without KG with Rondo
6-2
C's Off Rating 113.375
C's Dif Rating 106.6625
+6.71
SOS  .428
Average Off Opp. Rating 106.1
Average Def Opp. Rating 107.625

Basically they had the same impact in the results like you said. But overall the C's were playing worse teams, in particular they were playing worse defensive teams as well.

[Who]

Adjusted plus/minus and our efficiency differential for the games KG/Rondo missed indicate that losing KG cost us more.

The Celtics have a couple of players that can fill in for Rondo's offensive role. BBD's play as a starter falls so far below KG's that we can't make it up.

I looked at the team's efficiency numbers and found that not to be true.

According to my research: in the games Rondo played where KG didn't, the offensive efficiency was above our season average, but the defensive efficiency was slightly worse.  In the games where KG played with no Rondo, the defensive efficiency was better but the offensive efficiency was worse than our season's average.  The differential stayed about the same, though.

Season points per possession averages:

Offense: 105.8, defense: 97.7, Difference: 8.1

Games in which Rondo played, without Garnett:

Offense: 108.9, Defense: 100.6, Difference: 8.3

Games in which Garnett played, without Rondo:

Offense: 102.9, Defense: 94.8, Difference: 8.1

That's interesting I'd thought KG would be slightly more based on the adjusted +/- and what I saw during those games. I looked at the numbers on basketball referrence and I got slightly different numbers than you did (but the same result)

I think schedule is the answer to what my gut said:

Without Rondo with KG
Record 4-1* (you could count the Detroit where KG got hurt in the 1st QTR I didn't)
C's Off Rating 102.1
C's Dif Rating 95.18
+6.92
SOS  .504
Average Opp. Off Rating 105.68
Average Opp. Def Rating 104.5

Without KG with Rondo
6-2
C's Off Rating 113.375
C's Dif Rating 106.6625
+6.71
SOS  .428
Average Off Opp. Rating 106.1
Average Def Opp. Rating 107.625

Basically they had the same impact in the results like you said. But overall the C's were playing worse teams, in particular they were playing worse defensive teams as well.

The strength of the schedule, opposition, greatly alters efficiency marks (both def eff. + off eff) in small sample sizes.

As a result, def eff + off eff numbers are largely meaningless numbers if not weighted against the quality of the opposition (in small sample sizes).

[BballTim]

Adjusted plus/minus and our efficiency differential for the games KG/Rondo missed indicate that losing KG cost us more.

The Celtics have a couple of players that can fill in for Rondo's offensive role. BBD's play as a starter falls so far below KG's that we can't make it up.

I looked at the team's efficiency numbers and found that not to be true.

According to my research: in the games Rondo played where KG didn't, the offensive efficiency was above our season average, but the defensive efficiency was slightly worse.  In the games where KG played with no Rondo, the defensive efficiency was better but the offensive efficiency was worse than our season's average.  The differential stayed about the same, though.

Season points per possession averages:

Offense: 105.8, defense: 97.7, Difference: 8.1

Games in which Rondo played, without Garnett:

Offense: 108.9, Defense: 100.6, Difference: 8.3

Games in which Garnett played, without Rondo:

Offense: 102.9, Defense: 94.8, Difference: 8.1

That's interesting I'd thought KG would be slightly more based on the adjusted +/- and what I saw during those games. I looked at the numbers on basketball referrence and I got slightly different numbers than you did (but the same result)

I think schedule is the answer to what my gut said:

Without Rondo with KG
Record 4-1* (you could count the Detroit where KG got hurt in the 1st QTR I didn't)
C's Off Rating 102.1
C's Dif Rating 95.18
+6.92
SOS  .504
Average Opp. Off Rating 105.68
Average Opp. Def Rating 104.5

Without KG with Rondo
6-2
C's Off Rating 113.375
C's Dif Rating 106.6625
+6.71
SOS  .428
Average Off Opp. Rating 106.1
Average Def Opp. Rating 107.625

Basically they had the same impact in the results like you said. But overall the C's were playing worse teams, in particular they were playing worse defensive teams as well.

  Played worse teams with who? We played the Spurs and the Bulls without KG, right?

[Fafnir]

The strength of the schedule, opposition, greatly alters efficiency marks (both def eff. + off eff) in small sample sizes.

As a result, def eff + off eff numbers are largerly meaningless numbers if not weighted against the quality of the opposition (in small sample sizes).

Yeah that's what I was trying to show with the opponents Offensive and Defensive Rating averages (which were for the whole season)

Basically we were playing worse defensive teams while KG was out (they were slightly better on offense on the flip side) than with Rondo out while KG was in.

[Fafnir]

  Played worse teams with who? We played the Spurs and the Bulls without KG, right?

Worse teams with Rondo:

Games without Rondo and with KG:
Hawks
Pacers
Phily
Magic
Pacers

Games without KG and with Rondo:
Raptors
Wolves
Spurs
Raptors
Bulls
Rockets
Kings
Bobcats

The Raptors/Wolves/Bobcats/Kings really drag down the games that KG missed

[Celtics18]

Adjusted plus/minus and our efficiency differential for the games KG/Rondo missed indicate that losing KG cost us more.

The Celtics have a couple of players that can fill in for Rondo's offensive role. BBD's play as a starter falls so far below KG's that we can't make it up.

I looked at the team's efficiency numbers and found that not to be true.

According to my research: in the games Rondo played where KG didn't, the offensive efficiency was above our season average, but the defensive efficiency was slightly worse.  In the games where KG played with no Rondo, the defensive efficiency was better but the offensive efficiency was worse than our season's average.  The differential stayed about the same, though.

Season points per possession averages:

Offense: 105.8, defense: 97.7, Difference: 8.1

Games in which Rondo played, without Garnett:

Offense: 108.9, Defense: 100.6, Difference: 8.3

Games in which Garnett played, without Rondo:

Offense: 102.9, Defense: 94.8, Difference: 8.1

That's interesting I'd thought KG would be slightly more based on the adjusted +/- and what I saw during those games. I looked at the numbers on basketball referrence and I got slightly different numbers than you did (but the same result)

I think schedule is the answer to what my gut said:

Without Rondo with KG
Record 4-1* (you could count the Detroit where KG got hurt in the 1st QTR I didn't)
C's Off Rating 102.1
C's Dif Rating 95.18
+6.92
SOS  .504
Average Opp. Off Rating 105.68
Average Opp. Def Rating 104.5

Without KG with Rondo
6-2
C's Off Rating 113.375
C's Dif Rating 106.6625
+6.71
SOS  .428
Average Off Opp. Rating 106.1
Average Def Opp. Rating 107.625

Basically they had the same impact in the results like you said. But overall the C's were playing worse teams, in particular they were playing worse defensive teams as well.

You missed a few games.  Garnett played in 9 games without Rondo.  Remember that Rondo missed games during two separate small stretches.  The Cs were 7 and 2 in those games.  

Despite that our efficiency numbers seem to be different. I used the ones on the Hollinger efficiency ranking site, please let me know if there's another site that uses a different formula that you think is better.

Games with Rondo, no KG:

Average offensive opponent rating: 103.3
Average defensive opponent rating: 104.9
Difference: -1.6

Games with KG, no Rondo:

Average offensive opponent rating: 102.8
Average defensive opponent rating: 104.3
Difference: -1.5

So according to the numbers I found the level of opposition was very similar in the games each player missed, and the efficiency margins for the Celtics were pretty similar, as well.

The key difference seems to be that we were better defensively when KG played, and we were better offensively when Rondo played.  

This probably supports what most of us would guess by watching the games.  It also seems to support drza's theories about defensive and offensive impact of point guards.  However, it does not support your theory that KG is more valuable to the team's overall success than Rondo.    

[Celtics18]

  Played worse teams with who? We played the Spurs and the Bulls without KG, right?

Worse teams with Rondo:

Games without Rondo and with KG:
Hawks
Pacers
Phily
Magic
Pacers

Games without KG and with Rondo:
Raptors
Wolves
Spurs
Raptors
Bulls
Rockets
Kings
Bobcats

The Raptors/Wolves/Bobcats/Kings really drag down the games that KG missed

I already pointed this out in another post, but the 4 games that Rondo didn't play in that you missed are pretty important here.  In those games, we played the Raptors twice, the Nets, and Atlanta (being the only decent team).  We lost the first Raptors game.

[BballTim]

  Played worse teams with who? We played the Spurs and the Bulls without KG, right?

Worse teams with Rondo:

Games without Rondo and with KG:
Hawks
Pacers
Phily
Magic
Pacers

Games without KG and with Rondo:
Raptors
Wolves
Spurs
Raptors
Bulls
Rockets
Kings
Bobcats

The Raptors/Wolves/Bobcats/Kings really drag down the games that KG missed

  I think you missed games vs the Hawks, Nets and Raptors without Rondo.

[Fafnir]

You're right I was only considering the most recent games that Rondo missed. Here's the new data, I'm taking this all from basketball referrence game logs (I'm not calculating O-Rating just using advanced boxscores and averaging)

Without Rondo with KG
Record 7-2* (you could count the Detroit where KG got hurt in the 1st QTR I didn't)
C's Off Rating 104.82
C's Dif Rating 94.45
+10.36
SOS  .445
Average Opp. Off Rating 105.144
Average Opp. Def Rating 106.344
-1.2

Without KG with Rondo
6-2
C's Off Rating 113.375
C's Dif Rating 106.6625
+6.71
SOS  .428
Average Off Opp. Rating 106.1
Average Def Opp. Rating 107.625
-1.525

The inclusion of the extra four games actually takes us closer to an apples to apples comparison as the SOS and efficiency get pretty close.

[Fafnir]

The key difference seems to be that we were better defensively when KG played, and we were better offensively when Rondo played.  

This probably supports what most of us would guess by watching the games.  It also seems to support drza's theories about defensive and offensive impact of point guards.  However, it does not support your theory that KG is more valuable to the team's overall success than Rondo.    

Except by adding in those four games you actually get a clear separation in the C's overall play. They're efficiency differential climbs to +10. (They beat the Nets soundly twice, blew out the Hawks and lost by 1 point to the Raptors)

[Fafnir]

I didn't account for home and away games either, that's something that we could do. Typically HCA is worth an expected +3 efficiency differential if I remember that right. (not sure if its +3 raw points or +3 per 100 possessions)

[winsomme]

Do the people arguing that Rondo is overrated really believe that we'd be a better team with Westbrook or Rose as our PG?

Do you guys really need a stat to identify Rondo's ability to run this offense and consistently find the open man and hit him in rhythm with the pass?

Why do we have to get buried in numbers with something so obvious to the eye?