2015-16 Season Preview: Boston Celtics

BOSTON CELTICS

With a complete break from their 2008 championship squad, the Celtics are marching into the future with a set of role players and young guys with a nice cache of draft picks on the horizon. There’s a question of how good they can be with their core players, but few rebuilding teams are this good – it’s encouraging, and this should be part of an upward trajectory.

2015 in review:

The Celtics were a merry band of overachievers last season, somehow making the playoffs and hitting 40 wins. This is a surprise to anyone who glanced at the roster and knew management’s intent at rebuilding, as most people had them projected with a win total in the mid-20’s. But the catalyst was trading Rajon Rondo a quarter of the way into the season. Boston at the time had 39% win percentage, and after the trade their win percentage shot up to 53%. Once the team traded for Isaiah Thomas using one of the pieces they got for Rondo and a draft pick, the team improved yet again and they finished 20-10 the rest of the season. The team was swept in the playoffs, but the Cavaliers steamrolled the entire conference and Kevin Love was healthy for most of the series.

Rotation players in: Amir Johnson, David Lee.

Rotation players out: Rajon Rondo, Brandon Bass, Marcus Thornton, Jeff Green.

By bringing in Amir Johnson and David Lee, the frontcourt is pretty crowded, but Johnson in particular could inject a lot of value into the roster as long as he’s healthy. Just about every plus/minus model out there has him rated highly, especially the multi-season ones. But there’s a legitimate concern with the health of his ankles. Yet the biggest positive is that the team will have a full season without Rajon Rondo.

2016 Projected

A few months removed from the Celtics shocking everyone with 40 wins with misfit toys, there are still many critics out there who see the previous season as an aberration and see this next season as one Boston will spend diving into the lottery. But this is bogus for multiple reasons all objective, simple, and purely basketball-related. These critics still believe in the old-world value of reputation and awards, hence not recognizing Rondo’s drag on the team. That should be fully proven by now, and if anyone objects just review the Dallas playoff series where he was benched.

The assertion that a team should look a certain way to win a set amount of games is also erroneous. It has been proven, time and again, that a team can win many games using a near infinite number of forms. For a while, the media was obsessed with teams who “won the right way” through unselfishness and passing, and they condemned a flashy scorer who hadn’t proved anything and, through their own theories, would not be able to win a championship – that was Michael Jordan. But human reasoning is based on recent results, and after a while the new normal was established and you apparently needed a superstar scorer to win a title. That idea ignores a few decades of the NBA’s history, like in the parity-laden 70’s when teams like the Celtics (they won in 1974 where the leading scorer was John Havlicek with a moderate 22.6 PPG despite heavy minutes and a high pace), the Bullets, and the Sonics won. In fact, the Blazers exemplified the era more than any team, led by a pass-first center who helped defeat the ultimate superstar of the era, Kareem Abdul-Jabbar, and upset a team with multiple flashy stars in the 76ers. Let us also not forget the 11 Bill Russell titles where a fairly balanced team on defense kept consistently winning over teams with players like Wilt Chamberlain, Jerry West and Elgin Baylor.

You can have a pretty great offense even without a Kobe Bryant-like scorer, and key defenders are often underrated or oft ignored – put those together, and you can field a respectable team that hits the playoffs without obvious names on the roster. We’re in a new era where spacing is an obsession and smallball is a common tactic. Golden State thrashed the league in the regular season, and won a title even while using Draymond Green, who supposedly isn’t even big enough for a power forward because of a simple height measurement, as a center against a hulking Cleveland frontline and the best player of this generation in LeBron James. Then there’s the bonus of a team being led by its point guard grabbing a title, which hasn’t happened in a while and supposedly meant teams where the point guard is the focal point couldn’t win anymore. I would hope that this means people will understand that the means don’t matter in winning games – you just need to figure out a way to outscore your opponent – but history says doubt will never die and we’re going to have a large section of the media skeptical about a new issue that doesn’t really matter either.

With young teams, there’s a huge degree of skepticism because the players don’t have what I call a visible history of value. But that doesn’t mean those young guys will always have losing seasons. At some point, they cross the threshold and have a positive impact; in fact, believing this isn’t the case causes a strange logical paradox because how did those notable contributors, all the stars in the league, gain that credibility? If it matters what clout players have and what awards they have on their mantle, then the 2014 Nets would have accomplished more than the mediocre season they had. Winning basketball games is about basketball, not about how famous the names on the roster sheet are.

If you don’t believe those theoretical underpinnings, then use the example of the 2015 Hawks, who won 60 games with players no one thought were true stars before the season started, or better yet the 2013 Nuggets[1. Paul Millsap was the impetus for the Millsap doctrine, where Kevin Pelton and Ziller showed that productive bench players measured in per minute values translated their value to a starting job very well]. Take Andre Iguodala off the team, imagine Isaiah and Avery Bradley combined to form Ty Lawson, and you can posit 50 wins if things go well and they’re finely coached. Before the 2013 season, Lawson wasn’t highly respected, which is true of Boston’s parts. But winning changes a player’s reputation, and suddenly a huge slew of people understand his value. But something caused that winning – he wasn’t bitten by a radioactive spider; he always had talent and as a young player he naturally improved. We’re going to go through the same cycle of skepticism with the Celtics this season, and I imagine in a year or two people will be talking about Thomas or Marcus Smart differently.

We can start with the concerns about their lack of scoring punch – they have Isaiah Thomas. He’s already proven himself as 20 PPG scorer with the Kings, and in Boston he averaged 19 points a game … in a mere 26 minutes. One may cite his vagabond nature as a red flag, but the Kings have made some strange decisions everywhere and Phoenix had a glut of point guards and played musical chairs. They tossed out Goran Dragic too, and he’s an all-NBA member. Thomas’ teams do very well when he’s on the court[2. His on/off stats are ridiculously and consistently great.]. He’s actually a classic “moneyball” player because he doesn’t have the looks of a true NBA scorer on account of his stature, but results matter, not a measurement off the court. We can find evidence everywhere that his teams are more effective when he plays. He’s an extremely crafty scorer, especially around the rim – the fact that he draws a high number of free throw and hits 65% near the rim is astounding for someone so small. He’s an efficient, capable scorer[3. Strangely, his TS% was 57.4 for three seasons in a row and then hit 57.9 … in both Phoenix and Boston.]. If Boston starts the season well, don’t be surprised to see him as an all-star, as coaches tend to select at least one representative from every good team like they’re staffing the Senate.

Again, we don’t need to veer into heavy analytics here; this is just basketball. Avery Bradley’s exploits as a defender are well-known, and he’s a decent enough shooter too. But Marcus Smart is the guard to keep tabs on because he is a bulldog of a defender who did extremely well in two key indicators: a high amount of steals and offensive fouls drawn. He’s very young, and a breakout is possible given how he played without Rondo.

The frontcourt is unorthodox, but they were proven last year and they brought in a plus defender (when healthy) in Amir Johnson who can either roll to the rim or hit a jump shot from 20 feet and a bit further out as well as David Lee, a former all-star who spent some time with a very good team in Golden State. The frontcourt is deep enough where Brad Stevens can yank an ineffective player too. Kelly Olynyk was an oddly great team defender, and he’s a good enough outside shooter he was a positive on the court. Jared Sullinger was decent when healthy and he experimented with a three-point shot – think of him like a poor imitation of Kevin Love. He’s a good rebounder and passer with range despite the lack of a vertical – he’s no star, but he’s usually a plus on the court. Tyler Zeller is no imposing presence at the rim, but he’s a sneakily effective player who’s a pick-and-roll conundrum thanks to his ability to roll to the rim or pop out for one of the better midrange jump shots for a big man in the NBA. David Lee and Amir Johnson are also great at rim-running too, by the way – Boston will need some easy targets on offense with Smart’s inexperience and Bradley’s shaky point guard skills.

If the team has an imposing identify, it could be with their perimeter defense. Smart and Thomas notwithstanding, the team also has Jae Crowder and the ever enigmatic Perry Jones. With a high pace and an emphasis on outside shooting orchestrated by a great coach, there’s enough talent here to climb in the eastern conference rankings – which sorta says more about the weak conference than anything else.

Quick statistic/graph

Usually (relatively) tiny players are popular figures in the league because of they’re a gimmick and fans can relate to them. Nate Robinson has a shelf full of dunk contest trophies that prove this, and others like Spud Webb and Muggsy Bogues are more well-known than similarly skilled role players. But Isaiah Thomas, even with all his tricks and fancy passes, is still a League Pass player, known well by many NBA diehards but not a real sensation. He may eventually get noticed because he’s on track to be the greatest scorer his size in NBA history. Referencing the table below, Calvin Murphy is generally considered to be the best scorer who’s well under six feet tall, but Thomas scores more often and more efficiently[4. This is measured by points per possession, which cuts off everything before 1974 but only two players before that season qualified, and both guys were nowhere near Thomas’ scoring prowess. These results don’t change if you filter out seasons from when a player is older, by the way, in case Thomas was benefiting from the fact that short scorers don’t age well.]. There are actually few players his size who have even racked up significant minutes in the NBA. And Thomas might not just be the best scorer his size; he could be the best player depending on how his career pans out.

Table: players 5’ 10” or shorter ranked by points per possession, 5000 MP min. (source: b-ref)

Summary

Whatever labyrinth of logic and statistics you want to apply can be circumvented by a fairly simple argument: given that the Celtics won 40 games last, they’re returning most of their key players, they removed one clearly negative piece, and a calendar year’s progess will aid a young team, a win improvement of a few games into the mid-40’s is both reasonable and conservative. Don’t be surprised with 50 wins either if things go well. Plus, if they have a very successful season, the media will jam in some post hoc analysis and decide that some number of their players actually are special – we saw that with Atlanta last season. Just remember that the only thing that actually matters is the score of the game, not how famous their players are, and Boston’s roster has proven they’re capable of winning enough games in a depleted conference to hit the playoffs.

Win predictions:

PBP-Metric[6. This is the initial version of my own metric, which uses a full range of stats from 15 seasons collected from play-by-play logs and tested extensively to avoid overfitting.]: 45.6

PT-PM: 49

Nick‘s[7. For a short description, the predictions use regression models and neural networks to apply various stats like BPM, RAPM, and Win Shares to 10,000 simulations of the season game-by-game to select the “best” result.] : 46

Nathan Walker’s[1. Methodology coming on Friday!]: 52