Author’s Note: Well, it’s Sixers-Celtics once again! I live near Boston these days and so I end up spending a lot of time talking and thinking about the C’s. This has led to me crafting a piece on the question, why is there such a large difference between the perceptions of Sixers and Celtic fans on the future of the rivalry. In short, my observation is that Sixer fans think that next year, if the rosters are unchanged and everyone is healthy, the two teams are pretty close, and that if the Sixers add a big star like LeBron James, the Sixers will have the vastly superior squad. Whereas in my experience Celtics fans think that with no roster changes the Celts are much, much better than Philly heading into next year, and if the Sixers add leBron the teams are comparable. Since this is a big topic I am proceeding in parts. I’ll have individual pieces exploring important components of the perception gap, and then an essay putting the pieces together. The first piece in the series was my article on Robert Covington. This is piece two.
Brad Stevens is clearly an excellent coach. But there are lots of different ways to be excellent, and understanding the source of Brad’s value added seems to me the most important question for Danny Ainge as he builds the team, for fans of the team, and for GMs thinking of trading with Boston.
For those who haven’t been following Boston closely (I live in New England now so i do), Brad Stevens has been working miracles all season. He has faced solid teams while missing 4 of his 6 best players, or 5 of his top 8, and won or played them close. Recently, in a situation where they were almost surely locked in to the #2 seed and thus had little to play for, and missing many, many key players, they had a stretch where they beat desperate-to-win teams from Portland, OKC and Utah, all of whom had been playing excellent ball.
It’s been an extraordinary performance. At the start of this season I wrote a piece arguing that the Celtics’ roster was merely average. I said I couldn’t predict their record because it was possible that average talent plus great coaching plus great chemistry and fit could lead to a very good record, but as far as the individual players involved went, it was hard to see why they were any better than the teams that had played .500 ball the previous year in Indiana and Chicago and Portland. After all, I argued, good as Hayward and Irving are, those other teams had terrific players too, players like Dame and PG and Butler, but that didn’t stop them from having average records.
I spent a long time debating those issues with angry Bostonians, and of course I was much mocked by them during the Celtic win streak. Fair enough, I’ll take the heat! But months after the piece ran, a thoughtful Celtic-fan commenter here on LB suggested I do what I wasn’t smart enough to do the first time, and just look at the 538 CARMELO projections for the Celts’ players going into this season. And, guess what? CARMELO said the Boston lineup was indeed entirely average — even including Gordon Hayward!
But here we sit, and the C’s finished with 55 wins. Obviously that’s a tribute to the great strides made by players like Tatum, Brown, and Horford, among others, and I don’t want to take anything away from them or their efforts. Still, when basically everyone on a team plays better than their projections, often far better, you have to think the coach is part of the story. I think everyone agrees that Stevens is a great coach who did a great job. But just exactly HOW is he great? Here are some possibilities.
A) He could train players in such a way that they become permanently better than they would have been. Like, maybe Stevens is, or has an assistant who is, a super-skilled shot doctor and he can take so-so shooters and make them very good shooters, that sort of thing. Or we could think of more nefarious stuff; he teaches players how to push off in a way that refs have trouble seeing, something like that. Maybe call this the “Jay-boys,” as both Jaylen Brown (who was not effective as a raw rookie) and Jayson tatum (whom LB readers had, I believe, as their SEVENTH choice in last year’s draft) now appear to be terrific young players whom most observers think will have excellent careers whether they stay in Boston or not. It seems likely that Tatum was just underrated by the average draftnik, especially those with a Sixers focus who’d been burned by a different iso scorer from Duke! But it’s also possible that, had Tatum gone to Sacramento, we’d all be talking about how another Dukie looks primed for mediocrity.
B) He could create a fantastically effective scheme. So for example he could create offensive sets so clever that guys who would shoot 30% from 3 on a different, equally-talented team shoot 37% for Stevens because the scheme gets them open looks so often. We can call this the “Jae Crowder” for short!
C) He could get everyone to train much harder than others do, creating a sacrifice-for-the-team mentality that enables everyone to be better prepared than they would be under another coach, not because he teaches them tricks or uses better strategies, but because they worked harder to get ready. Let’s oversimplify by calling this “getting guys to lift more weights” even though of course it may be about practicing shooting more, or doing cardio, or eating right, or any of a hundred other ways players can sacrifice outside the game itself.
D) He could get guys to sacrifice DURING the game. Sacrifice can mean setting hard picks that are painful to deliver, or passing up stats to let someone else have the glory, or giving defensive help even when it risks making you look foolish if the ball goes back to your man for an easy layup, or standing in and getting posterized because being there makes it 5% more likely the dunk gets missed, or just playing really hard instead of being relaxed sometimes. Maybe Stevens is a great motivator that gets players to leave it all on the floor in all these ways, and more. We’ll call this “in-game sacrifice.”
E) Maybe Stevens has ways of winning close games. The most obvious is really great play calls coming out of timeouts, something Stevens has a reputation for. But there could be other strategies, defensive schemes, whatever, that make the team likely to put up a better record than teams with similar points scored and allowed because they win more than their share of the close ones. Call this “close shaves.”
F) Something I haven’t thought of, please share in comments!
Now, let’s think about each of A)-E). What is the evidence for and against, and what are the implications? Of course these are by no means exclusive, maybe Stevens is great at all these things. But, that isn’t usually the way. If you read Bill James book on baseball managers, he shows that even among great managers, all have strengths and relative weaknesses.
Let’s take A) and B) together. It seems likely that at least one of them is true! But which one it is makes a big difference if you’re thinking of trading with the C’s. As I mentioned, most folks think Brown and Tatum are going to be great even if they got shipped to new Orleans for Anthony Davis. Certainly both have been excellent this year, both counting and on-off stats testify to that. How confident are we that they aren’t just a creation of effects that disappear under another coach? It’s worth reviewing the recent history of players leaving the Celtics.
1) Jae Crowder. Crowder was not viewed as a special player before coming to Boston. When playing under Stevens, he was one of the top on-off players in the NBA for a couple years. Then he went to Cleveland and was awful. Now he’s trying to put it back together in Utah. It’s always possible that he was hurt or something and will return to the terrific player he was under Stevens. But for now, he looks like someone who succeeded only in Brad’s scheme.
2) Avery Bradley. Bradley has been awful this year by on-off statistics. The truth is, on-off statistics say he was awful in Boston too! In my opinion Avery Bradly is not a good NBA player and never really has been; that is to say, he had a year a few years back where he was OK but if your cherry-picked best year is just OK, you’re not really good. So in my opinion he is not an example of Stevens magic. Some people think he was terrific in Boston and still is terrific when not slowed by injury; these folks and I don’t see eye to eye but they do share my view that he is not a “Jae Crowder.” Others think he was excellent under Stevens but no longer is, so there is a faction out there that sees Bradley as another guy in the Crowder group.
3) Amir Johnson. I have written extensively about Amir in the past. My view is, he was a truly excellent player before coming to Boston, was truly excellent under Stevens, and this year, playing out of position at center, has played well, though not as effectively as he did under Stevens and before. There’s a spreadsheet on the web that computes adjusted plus-minus for the full 14-year period 2001-2014, and when you look at the rankings, the top 10 is basically all the guys you’d expect, plus the two most underrated players of the era, Manu Ginobili and Amir Johnson. This year Amir did a very good job, but he’s no longer one of the very top players in the Association.
It has taken Philly fans some time to appreciate the subtle effectiveness of Amir’s game, but it seems as though that is starting to occur. Anyway, I do think he was a lot more effective the past two years than this year, and while it’s possible that’s just the difference between being 29 and 30 or the game changing to force him into the pivot as modern PFs are too fast and shoot too well for him to guard, nevertheless I have to say that while I still adore his game, in my own opinion he fits the category of someone who has been less effective after leaving Brad Stevens. Fun fact for Sixers fans: as of the day I wrote the first draft of this piece, the numbers 97, 98 and 99 players in ESPN’s RPM, all around +1 which is to say average-starter level, were:
RPM is not the end-all and be-all, by no means is anyone saying these three are necessarily equal players. Just thought it was interesting!
4) Kelly Olynyk. KO was pretty good in Boston and has been even better in Miami. So, not a case of a guy falling apart after leaving Stevens. It’s worth noting that Olynyk’s new coach, Eric Spoelstra, is on the short list of coaches seen by many as being comparable in quality to Stevens. But maybe Kelly is just, you know, good!
5) Isaiah Thomas. Fits the no-good-after-Stevens profile, but the injury surely explains part of it and may explain all.
6) Evan Turner. I haven’t followed him closely, but by all accounts he’s never been nearly as productive post-Stevens as he was in Boston.
I’m sure I’m missing some but you get the idea. We can add other guys who came to Boston and have yet to leave — Al Horford dipped when he started with Stevens but is now having probably his best year despite being older than Amir Johnson. Marcus Morris got hot late in the season; overall this season he’s probably been around the level of his recent production.
And then there are the cast-offs and who-dats; this year Stevens has gotten astonishing production levels from guys like Daniel Theis and Shane Larkin. Terry Rozier has on-off stats that are awfully close to those of Kyrie Irving! But those folks don’t tell us much about A) vs. B). Which must be frustrating for opposing GMs. If you get Rozier, are you getting a quality starting PG, as the numbers suggest? Or is he a Crowder whose magic spell will vanish outside the Garden? How about Marcus Smart, attractive free agent signing, or someone who’ll be unplayable in big games because of his inaccurate shooting? The history does not give a definitive answer of course. But certainly there’s reason to be wary of assuming that thriving under Stevens is proof you can thrive elsewhere.
Now let’s pair C) and D). You might ask why it matters whether Stevens’ motivational skills apply primarily to in-game or pre-game effort. But it’s all the difference in the world! Years ago a thoughtful friend told me that he thought Kevin Garnett was overrated. I was surprised and started to quote him all KG’s tremendous stats. He waved me off; he already knew all that. His point was this: KG gave 100% every minute of every game. Which is awesome! And he deserves all the credit for all the points and rebounds and blocks he delivered, he earned them! But the problem, my friend said, is that come the playoffs, all KG could do was what he had done all year. So now imagine another player who gives 80% most of the time and 100% in the big games. If that guy has regular-season performance that is 90% of KG’s, he is the more valuable player, because (in this little made-up calculus) he’ll be 90% as good in the regular season but about 10% SUPERIOR in the playoffs — he is the more talented player and so when everyone gives their all, he can deliver more.
Now, suppose Stevens’ primary superpower is getting everyone to give 100% during the game every night. He’s an exceptional motivator, or plays mind games well, whatever you want to call it. That will win you a lot of regular-season games. But when you get to the playoffs, your opponent, whom you outworked during regular-season games, is now digging just as deep as you are. Being just as unselfish as you are, it’s all about winning. If this is his main source of edge, it really could be that Stevens is “just” a great regular-season coach.
Now, some caveats are in order. First, if a lot of the motivation is for practice, weightlifting, studying game tapes, and the like, well, that stuff is cumulative all season long. A less-well-motivated team can try to catch up in the postseason, hitting the video room for study sessions, staying late for practice shots, but it’s probably too late to make much of a dent in the preparation gap. And even on the in-game stuff, it may not be that easy to flip the switch. If one team has been running hard all year, and the other starts running hard in the playoffs, which team is going to be good at passing/shooting/defending while running hard? I’m guessing the team that did it all year! If one team has been unselfish all year, making the extra pass, and another decides to start being unselfish in the playoffs, which team is going to more often see that extra pass fly into the stands? The team that hasn’t been doing it consistently, right? So, I’m skeptical how powerful the “KG effect” is likely to be in Stevens’ case. My guess is there is a little something to it; something legit to the idea that Stevens is better than most coaches at getting his guys to give their all on a random Tuesday in Detroit, and that that motivational edge will be smaller come the postseason. But I wouldn’t expect Stevens or his teams to turn into pumpkins in big games as a result of this, either.
Oh, and let me mention: I’m not taking a position on KG, other than the obvious position that he was an absolutely tremendous player. I’m not arguing here that he actually lacked an extra playoff gear; I haven’t studied that, it’s just a claim someone made to me once that got me thinking. But I will say this: when people tell me that a player DOES have an extra gear for the playoffs, I’m always surprised that it isn’t seen as a mortal insult. I mean, when you say that about someone, aren’t you fundamentally saying that they’re dogging it the rest of the time? If it’s LeBron, then, sure, he does it and we all understand, it’s just not realistic to expect him to go full-bore every night and then completely take over games in the playoffs as well; even the King may not have THAT much energy.
But if someone tells you that a very-good-but-not-the-King player “saves his best for the playoffs”... um, aren’t you calling him a slacker? If you had a kid who played ball, would you want people to say about him “he’s like Kevin Garnett, he gives it all every night”? Or would you want them to say “He doesn’t give full effort most nights, but he’s incendiary in big games because in those, he really tries!”? Anyway, that’s neither here nor there; just got me thinking, is all.
Finally, we turn to E) — is Stevens good at winning the close ones? Well, based on this year it sure looks that way. Boston has outscored its opponents by 3.6 points per game. There are five teams that have outscored by around that much, the Sixers, Thunder, Jazz, Blazers and Spurs. On average those teams have outscored by a hair more than the Celtics and on average they won just under 49 games. The C’s won 55, so over 6 games more than one would expect based on the comparison group. Six games is a lot! Remember, that’s not saying Stevens added six games of value, this calc gives Brad no credit at all for getting his team to score points or prevent scoring. It’s just about the Leprechaun, the feeling that the C’s always win the close ones. This year it really was true. The method we employ below, of allocating three wins per extra point, suggests the extra wins this season are probably closer to 4 than 6; the Sixers, for example, have won several games less than point differential would suggest, as will surprise no one who followed the squad’s early-season frustrations and late-season blowout wins.
Now, of course the C’s numbers could just be luck, or could be clutch players like Kyrie. So, what about the past, when Stevens had different players and, presumably, different luck? Well, last year was similarly impressive. The Celtics were +2.7 in point differential, normally over a full season a team gets an extra 3 wins or so per point, so that suggest they should have been 8 games over .500. which would be 49 wins. They won 53, so +4.
I’m a lazy, lazy man, but not so lazy I can’t look up another couple years!
2015-16: +0.8, 48-34
OK, this is getting weird, another 4-5 games of outperformance.
2014-15: +2.2, 40-42
2013-14: -2.1, 25-57
Finally, some evidence of imperfection; a meaningful 7-game underperformance relative to point differential in Stevens’ second season with the C’s. And a huge, almost 9 game underperformance in his tanking season... oops, sorry, must have made a typo there, the Celtics are far too noble to tank, I meant to say his rookie season in the NBA. I leave it to the reader’s judgment whether those extra losses his first year were deliberate or a result of his not yet having mastered the close-and-late aspects of the game. Upshot is, across five years in the NBA, Stevens has not performed especially well in terms of wins compared to point differential. But if you focus on the past three years, either because you think he was deliberately losing at the beginning or because he just took a couple years to figure it out, then there is compelling evidence that Stevens is really good at winning more games than the points scored and allowed would suggest. Which, if true, would be extraordinary, an absolutely huge source of value added.
So, what have we learned? Well, a lot is up in the air, Brad Stevens is only 41 years old and may well still be coaching the Celtics in 2050 — yikes! But I’d say there is solid evidence for him getting guys to play better for him than they do elsewhere, solid evidence he’s a great coach to have if you are an NBA rookie trying to learn the game and build a career, and solid evidence you’d like him on the sidelines if you’re in a close match. I think there is a decent story that he may be a great motivator and that the advantage of that come playoff time will be diminished, but it’s too soon to say there’s much evidence one way or another for that story; we’ll know a lot more two or three postseasons from now.
But overall, Stevens is a force to be reckoned with, probably as valuable to the team as a superstar player, and in a way far more valuable, since he doesn’t count under the salary cap and won’t be kept off the floor by injuries. He’s going to make Boston tough to beat for a long, long time.