What happens when we don't start Baynes?

It seems we spot the other teams a big lead when we start the game with Baynes on the bench. Then Baynes comes in and both our O and D play better.
Am I just imagining this, or is it a real pattern?

worcester wrote:It seems we spot the other teams a big lead when we start the game with Baynes on the bench. Then Baynes comes in and both our O and D play better.
Am I just imagining this, or is it a real pattern?

Too early to tell.

anyone know a place for deep analytics ?

https://www.basketball-reference.com/players/b/baynear01.html

https://www.basketball-reference.com/players/m/morrima03.html


comparing those two players and using the defensive rating per 100 and Offense rating per 100 possessions (what I have seen some use a gauge ).

they break down like this

Baynes- 103 ORating, 97 DRating
Morris- 113 ORating, 97 DRating

Using Offense and defense plus/minus they break out this way:

Baynes- Minus 3.4 Offense, PLus 3.4 Defense

Morris- PLus 0.2 Offense, PLus 0.7 Defense

so what does any of that mean?

I have no idea!!

According to the analytics at 82games.com,

http://www.82games.com/1718/1718BOS2.HTM

we do the very best when it is we do 2nd best with Irving-Smart-Brown-Tatum-Baynes

or we do best with -Larkin-Rozier-Smart-Morris-Baynes on the floor


3rd best Irving-Smart-Brown-Tatum-Horford

we do 4th best with -Brown-Tatum-Baynes-Horford-Irving on the floor


Much worse with Brown-Tatum-Irving-Horford-Morris

This is where analytics can be twisted to support the case you are trying to make.

Kind of like taking a player who averages 8 minutes a game and scores 8 points per game- and trying to say he would score 48 points a game if he played 48 minutes.  Of course that is ridiculous.  

How many minutes this season have we seen Larkin-Rozier-Smart-Morris-Baynes on the floor?  With such a small sample size, to suggest that is our most effective 5 is ludicrous.  

Baynes isnt capable of playing big minutes against mobile big men.  If he starts against people like Steven Adams, Aldridge, Jokim Noah, Brooke Lopez - not worries.  If he has to start against Embid, Anthony Davis, Kevin Love - it will be tough going.

It is way too early to make any kind of extrapolation from the data.  In Brad I Trust.

I don't believe in this stuff, because it does not take into who the opposing 5 is on the other team is....wouldn't that make a difference too?

I beg to differ. If you look at the unit with the most time on the floor together it is

1. Irving-Brown-Tatum-Horford-Baynes and it has a win % of 71.4.

Next in time on the floor is
2. Irving-Smart-Tatum-Brown-Horford with a win % of 88.8 - probably our best lineup statistically, even with Smart's poor shooting.

Next in terms of court time is
3. Irving-Brown-Tatum-Morris-Horford with a win % of 40.0!

Next in terms of court time is
4. Irving-Smart-Brown-Ojeleye-Horford with a win % of 66.6.

The lineup with the fifth highest amount of time is
5.Irving-Smart-Brown-Horford-Baynes with a win % of 100.0.

After that the court time for lineups drops off pretty steeply, so the sample size should really be considered from among the top five.

The outlier in these groups is when we have Morris on the floor instead of Baynes. Next question for a statistician with plenty of time on his hands is what lineups are on the floor when the Celts go down by 16 points? I bet it is not when Baynes was in the lineup.

I do like Morris, but Brad should be more judicious in his use at the start of games. Spotting teams 16 point leads regularly is not going to work out in the long run. Surprising how well we do with Smart on the floor.

Certainly interesting and it takes into account how these groups do against whoever is on the court against them.  

Otherwise how can they come up with success rates?

Cowers answer that?

beat

The Celtics have played a total of 17 games this year.

They are being outscored by their opponents during the first qtr 416-407 or around 0.5 points per game.

Worcester,

Baynes has started 10 games and the Celtics have lead in the 1st QTR in 7 of those 10 game.

In the 7 games where Baynes has not started the Celtics lead in the 1st QTR 4 out of 7 times

Based on this rather mundane analysis the Celtics are a better team in the first QTR when Baynes is starting. This does not take into consideration teams/matchups or home vs away games.

The Celtics win the 1st QTR 70% of the time when Baynes starts but only 57% of the time when he does not.

One reason this has occurred may have to do with starting the game with a higher level of physicality. In other words you hit them in the mouth from the opening tip and it may very well take the opponent out of their game early.

This is a very small sample size but as the sample increase it may lend even more credibility to any conclusions drawn.

I think you have made a very interesting observation.

dboss

Thanks for the input Dboss. Yes, it is a small sample size, and we'll see how it shakes out with more games. I can tell you this - if I got smashed in the mouth by Baynes, I'd be forfeiting the entire game, not just the first quarter!

Dboss

How many minutes of those first quarters has he played and what was the score at the time he came out?  I know he didn't play any entire first quarters so that would have some bearing on these numbers perhaps.

I don't expect you to check these but I might when I have a lot more time.


A LOT more time.   

beat

The lineup with the fifth highest amount of time is
5.Irving-Smart-Brown-Horford-Baynes with a win % of 100.0.


I like that % a LOT!!

Irving-Smart-Tatum-Brown-Horford with a win % of 88.8 is from the second largest sample size. Pretty impressive.
The entire team is pretty impressive.

One of the reasons why the sample sizes/5 man unit are so small.  Brad is doing a great job of spreading around the sugar...



Brian Robb @BrianTRobb

about 25 minutes ago

The Celtics have won 15 straight despite no one on the roster averaging more than 32 minutes per game. Just another sign that Brad Stevens working his magic yet again: bostonsportsjournal.com/2017/11/20/bra…
reply retweet like


bob

.

Way too much around the edges that cannot be accounted for to make this kind of assertion 17 games into a season.

The Celtics defense is predicated on adjustments and on the idea that physical play takes a toll - adding up over the course of a game. For my eyes, I see the first quarter as being an anomaly - with the Celtics getting the best shot from a team that is fresh and playing a free form game.

Over time (particularly at halftime) the coaching staff makes adjustments, and slowly as the second half wears on the Celtics turn the defensive screw, at the same time fatigue is beginning to take a toll on the opposition.

Whether he starts or not - Baynes has been a pleasant surprise this year and I think the Celtics would do well to sign an extension ASAP. At some point during every game, his banging, positioning and using his fouls absolutely helps to set the tone.

But when the game is on the line, I want my most versatile players, those that can guard multiple positions and get their own shot if needed. Aron is neither of those things. So unless as I said earlier, you are going up against a strong, post up typel center like Brooke Lopez, Boogie Cousins, Marc Gasol - I dont see Baynes being in the game down the stretch.

Great points Bob and MrKleen.

In the final 5 minutes aren't we usually seeing Kyrie, All, Jalen, Jayson, and Marcus? Whomever we have out there at the end, it sure as hell has been working.

Good statistical points made about various combinations (sometimes too many stats for me to comprehend) and a very solid point made by Cowens about considering exactly who the opponent are having effects on success.

The biggest thing I see missing when Baynes does not start a game is overall toughness of the Cs without Baynes is missing as opposed to when he does start a game. I believe that there is no question that although we have some other tough guys in the starting lineup that Baynes is the guy who other teams are looking at when trying to figure out which team is going to be the toughest on any give night.

Perhaps Baynes should start every game, but not stay in for the usual 8-9 minute periods that most starters play in the first quarter. If he starts and plays 3-5 minutes to set the tone and then comes out, he should be still able to play his regular number of minutes while maintaining his successes.

worcester wrote: In the final 5 minutes aren't we usually seeing Kyrie, All, Jalen, Jayson, and Marcus? Whomever we have out there at the end, it sure as hell has been working.

Exactly and that was my point. If Brad used your numbers as the basis for his decision making, this would not be the 5 on the floor down the stretch.

Au contraire. I i think Brad is using the 88.8% win lineup at th e end of games

worcester wrote:Au contraire. I i think Brad is using the 88.8% win lineup at th e end of games

So why isnt he using the 100% line up?

I'll tell you why. 17 games into a season, his eyes and intuition as a basketball coach supersedes any metric.

beat wrote:Dboss

How many minutes of those first quarters has he played and what was the score at the time he came out?  I know he didn't play any entire first quarters so that would have some bearing on these numbers perhaps.

I don't expect you to check these but I might when I have a lot more time.


A LOT more time.   

beat

Beat

That is a good question. I am not sure what the minutes were.

I assume they were not a lot because Baynes does not play a lot of minutes.

I have a spread sheet that is now tracking Celtics 1st QTR results and if Baynes was a starter. I guess I could add a minutes column but I am so damn lazy since I retired. okay, I can work on that part but it will not be until after Thanksgiving. Is there any other variable you think should be included?

dboss

dboss wrote:

beat wrote:Dboss

How many minutes of those first quarters has he played and what was the score at the time he came out?  I know he didn't play any entire first quarters so that would have some bearing on these numbers perhaps.

I don't expect you to check these but I might when I have a lot more time.


A LOT more time.   

beat

Beat

That is a good question.  I am not sure what the minutes were.

I assume they were not a lot because Baynes does not play a lot of minutes.

I have a spread sheet that is now tracking Celtics 1st QTR results and if Baynes was a starter.  I guess I could add a minutes column but I am so damn lazy since I retired.  okay, I can work on that part but it will not be until after Thanksgiving.  Is there any other variable you think should be included?

dboss

look out you might find yourself a job !!

My son does video analytics for the BC women and he supplies the coaching staff this very stuff...FOR them and the team they are about to play

beat

Beat

Sounds like Marcus is in a good situation. And I have a feeling that one day Marcus will be working for the Celtics.  

Dboss

dboss wrote:Beat

Sounds like Marcus is in a good situation. And I have a feeling that one day Marcus will be working for the Celtics.  

Dboss

Dreams can come true.   

He is building a pretty nice resume. He's said he wants to get into coaching someday.
At least he is in the right city and he's already got a few connections.

beat