The NBA Rebuild Survival Handbook

The NBA playoffs are upon us! While the Golden State Warriors and Cleveland Cavaliers steamroll their ways to a seemingly-inevitable Finals clash, the majority of NBA fans are left wondering about the prospects of their own squads. Indeed, a few of these fanbases are squirming nervously about in their seats, as they have come to the realization that their teams may embark on the dreaded NBA rebuild.

What is a rebuild, and how does it come about, you say? Just ask followers of the Toronto Raptors and Los Angeles Clippers. After years of fielding decently-competitive outfits, repeated playoff exits have put these teams at risk of losing their top players in free agency or via trade this summer. By the time the 2017-18 NBA season tops off, the Raptors may lose star Point Guard Kyle Lowry and Forward Serge Ibaka, and the Clippers may lose star Forward Blake Griffin and trade other significant pieces of their core. If these events, either spurned internally or by the decisions of these individual players, transpire, what is a team to do?

I attempted to investigate this question by looking at every season dating back to 2007-08. I then looked at each team's total Win Shares for a particular season, and more specifically the percentage of Win Shares that a team lost from the year prior. I used Win Shares because I believe it gives the most accurate and holistic view of a player's value (You can read more about it here).

These losses might have occurred due to a player's retirement, an injury, a trade, or a player signing with another team during free agency. In the time period I covered, NBA teams lost players that accounted for, on average, approximately 28% of their prior season's Win Shares. I decided to characterize any team that was in the 75th percentile or higher in terms of Win Shares lost from the prior season as "rebuilding" teams (this number was around 40%).

I then pared down that list to include only those teams that had Win Share totals above 40 in the season prior to them losing 40% or more of their roster. 40 total Win Shares provides a decent line of demarcation for a good-not-great team, and I wanted to look at teams who had just started to rebuild, not teams that had been mired in seasons-long losing.

In order to determine the keys to a successful rebuild, I didn't want to just focus on championships, as that presents a fairly binary reality (you win the Championship, you're a success; you didn't, you're a failure). Instead, I looked at how effectively teams were building championship-level characteristics. Looking at every Finals team since 2007-08, I noticed the following:

• No team that made the Finals had fewer than 5 average or better NBA players (an average NBA players produced about 2.3 Win Shares in the sample I investigated);
• Each team's best player had at least 7.7 Win Shares (I'll refer to this as the "superstar threshold");
• It is well-documented that teams outside of the top 10 in defense rarely win the championship. So, I identified the lowest total for Defensive Win Shares of teams that made the Finals in the years that I studied. The answer: the 2014-15 Cavaliers, who had 18.2 total Defensive Win Shares.

I then looked at my final list of rebuilding teams, my goal to spot trends - which strategies resulted in teams moving in the right direction? In a sport where luck plays such a massive part in success (just ask the '96-'97 Celtics), are there any concrete moves or philosophies that trend teams upwards? I have a couple of notes on how I compiled my data below (please scroll to the end of the article if you'd like to see them), but here they are - in a world of hoops uncertainty, these are the commonalities; the do's and don'ts of rebuilding in the NBA.

GET TO THE POINT

Everybody knows that today's NBA is dominated by small-ball. The traditional big man who lumbers around the court scarfing down rebounds and bullying guys in the paint has become a fossil. The modern NBA is engaged in a love affair with the three-pointer. But with it has also come the (less-heralded) rise of the point guard. This makes sense - in an era where three-point shooting has become the norm, what's more important than a player who can distribute the ball to said shooters? While the Point Guard position does not necessarily monopolize the facilitating duties of an offense (see teams like Cleveland and Utah, who rely more on wings like LeBron James and Gordon Hayward to run the show), it is also true that the position is about as deep and talented as it has ever been. So if you're a rebuilding team without one - get one. You don't have much of a chance otherwise.

Case in point - the Detroit Pistons, New Orleans Pelicans, and Orlando Magic were all teams I studied that, by-and-large, have led unsuccessful rebuilds. The Pistons lost almost 47% of their team in the 2009 offseason, most notably big men Rasheed Wallace and Antonio McDyess, who alone accounted for over 25% of their team's 40.8 Win Shares in 2008-09. The Pelicans traded Chris Paul in 2011, and the Orlando Magic did likewise with Dwight Howard. Both players are arguably the best in each franchise's history, and Wallace and McDyess represented key cogs of a Pistons team that had been a perennial playoff contender. Since these events, each team has shared one thing in common - the lack of a true floor general.

It hasn't been for lack of trying, mind you - the Pelicans acquired Eric Gordon, Jrue Holiday, Tyreke Evans and drafted Austin Rivers; the Magic drafted Elfrid Payton and Victor Oladipo; and the Pistons drafted Brandon Knight and traded for Brandon Jennings and Reggie Jackson. Unfortunately for these teams, a combination of injuries and ineffectiveness have left this group of players unable to approximate adequate play from this generation's most important position. In the 19 combined seasons since these three teams began rebuilding, the best individual season from a point guard has been Reggie Jackson's '15-'16 season, where he amassed 6.9 Win Shares. Rodney Stuckey's 2010-11 season for the Pistons is next at 5 Win Shares, and no other comes close to eclipsing 5. That won't get it done in today's NBA.

The craziest part about these three teams, and perhaps the strongest evidence for the need for a lead guard or wing in the NBA, has been that each franchise has solved the hardest part of the rebuilding equation - acquiring a superstar. After losing a player the caliber of Dwight Howard or Chris Paul, perhaps the hardest feat is developing or luring a star to a team devoid of one. Nevertheless, Detroit drafted Center Andre Drummond, who's eclipsed the 7.7 "superstar" threshold twice since his rookie season in 2012 and is generally regarded as one of the more talented 5s in the league. The Pelicans drafted Anthony Davis, who has been arguably the best power forward in the game since 2013 (he's surpassed 7.7 Win Shares three times since then). And though the Magic haven't had a star since Howard left, Center Nik Vucevic reached 7 Win Shares in 2014-15, and many expect Aaron Gordon to make a star turn if Orlando can ever unclog space for him at his ideal Power Forward position. The struggles these teams have experienced despite the presence of superstar-caliber talent can be traced back to a lack of identity on offense that goes hand-in-hand with point guard play.

Consider the other side of the coin - the teams that have embarked on successful rebuilds over the past few years. The Celtics traded Rajon Rondo and then acquired Isaiah Thomas during the 2014-15 season, ending the year as a surprising postseason entrant. And Boston illustrates an important point when it comes to point guard - just because you have a good one, doesn't mean you have the right one. Rondo was a serviceable, sometimes spectacular point guard when the Celtics had players around him who could shoot (Ray Allen) and create one-on-one (Paul Pierce). When the Celtics lost those players, they needed a floor general who could get his own shot. Enter Thomas. The next season after the Celtics acquired IT, they experienced an eight-win improvement and reached perennial contender status in the East.

But the Celtics aren't the only beneficiaries. Other rebuilding teams, including the Bucks, the Suns, the Jazz, and the Rockets all experienced significant jumps in quality when they acquired/developed offensive facilitators. However, in three of these cases, those facilitators didn't arrive at the point guard position (which makes this a trickier point to illustrate). The Suns almost doubled their win total from 2012-13 to 2013-14 by employing a dual-point guard lineup with Eric Bledsoe and Goran Dragic. Even though the Bucks signed Matthew Dellavedova this past offseason and have gotten superb play out of Malcolm Brogdon, they truly began to turn the corner last year when they turned their offense over to Giannis Antetokuonmpo. And the Jazz and Rockets both began their respective turnarounds when they handed the keys to super-talented wings Gordon Hayward and James Harden, respectively. The only notable exclusion of a team that has experienced success without a true leader of the offense has been the Dallas Mavericks. However, the Mavericks have a Hall of Fame-level coach in Rick Carlisle and notably employ a read-and-react offense that relies more on quick passing and movement than a typical NBA offense. Case in point - if you don't have a player to guide your offense, you won't start to get better until you do.

NO BIG WORRIES

I fully acknowledge that this next observation might be 10% data, 90% confirmation bias. But here's the thing - while I was reviewing rebuilding teams from the past 10 seasons, I kept noticing a trend - the interchangeability of non-superstar big men.

Make no mistake - a departing superstar at the Power Forward or Center position will still cripple a franchise. It happened when Dwight Howard left Orlando, as Superman amassed 7.7 Win Shares in 2011-12 (which projects to 9.6 Win Shares, as that season was strike-shortened). The next season, four Magic Centers combined for 7.8 Win Shares in total. The same thing happened when Power Forward Kevin Love was traded by Minnesota (14.3 Win Shares by Love in 13-14, 1.9 Win Shares by five total players at the position the next season). But if you need to replace a mid level big man, no need to sweat it.

Take, for example, the 2014-15 Indiana Pacers. The Pacers were arguably one of the better teams in the East for a four-year stretch between 2011 and 2015, reaching the Eastern Conference Finals twice in a row. They ascended to the top primarily on the scoring prowess of dynamo Small Forward Paul George and a fearsome defensive frontline, led by David West and Roy Hibbert. 

However, after the '14-'15 season, the Pacers lost Hibbert (4.2 WS), West (4.3 WS), and Luis Scola (4.4 WS). Though all three were declining, they were all still well above-average contributors to a contender. So losing all three would certainly take its toll, right?

Nope. The following season, three players (Ian Mahinmi, Jordan Hill, and rookie Myles Turner) combined to outperform the departed triumvirate (13.2 combined Win Shares to 12.9 combined Win Shares), and the Pacers ended up winning seven more games than the prior season and returning to the playoffs. The interesting thing is that all three new players significantly outperformed expectations, based on their prior season's performance, or the average performance relative to their draft position in the case of Turner. Mahinmi improved from 2.1 to 6.2 Win Shares, Hill from 2.2 to 3.9. Turner, drafted 11th overall, would've been expected to contribute 0.9 Win Shares based on the average contribution of rookies drafted at that spot, but instead produced 3.1.

Why did this occur? The easiest assumption is playing time. The departures of Hibbert/Scola/West opened up opportunity for the new guys, and they obliged by equaling the production. Despite changes in style, athletic ability, tendencies, etc., the new trio did a fine approximation of the old one.

This makes sense considering the change in playing styles that I mentioned earlier. Things like tempo and the types of shots taken on offense are by-and-large dictated by the point guards and lead guards in the modern NBA, not big men. Therefore, the majority of mid level bigs fall into fairly distinct categories - athletic rim runners, shot-blockers, rebounders, old-school post scorers, and stretch-fours. All five types, at least in the middle tiers, have pretty distinct strengths and weaknesses that usually don't blend. If you rebound, you usually don't shoot threes, and vice-versa. So if your rebuilding team is about to lose this caliber of big, it's likely they can find someone cheaper to fill their shoes.

Interestingly, this trend also seemed to be evident when big men remained with their teams. In other words, Power Forwards and Centers tended to cannibalize each other's production proportionate to each other's playing time. For example, Tyler Zeller of the Boston Celtics produced 6.5 Win Shares as the starting Center on the 2014-15 team. The next offseason, the Celtics signed Amir Johnson to a two-year $24 million deal to play Center. Zeller's playing time dipped in 2015-16, thus resulting in 1.7 Win Shares, a 4.8-share drop from the previous season. And Johnson? He produced 5.9 Win Shares, meaning the transaction resulted in a net-1.1 Win Share increase. This type of Win Share cannibalization also occurred in Detroit, Houston, Utah, Atlanta, Miami, Portland and others throughout their respective rebuilds. In fact, when Portland lost Lamarcus Aldridge and Robin Lopez in the 2015 offseason, it was fair to question if they would recover, considering those two players accounted for a whopping 13.7 of their team's 51.9 Win Shares. But the next season, relatively unheralded big men Mason Plumlee and Ed Davis virtually replicated the total (12.8 combined WS). The reason? Portland already had an All-Star Point Guard in Damian Lillard to coordinate the offense. Having that cornerstone piece of the roster set, the Trailblazers were able to basically plug-and-play the replacements for what were considered core members of their team. 

So if you're a rebuilding team with a non-superstar big about to hit free agency (or if you're considering signing one), it may be best to take a pass. Because chances are, someone else can fill their shoes capably.

NO MIDDLE GROUND

This section is a bit of an extension of the prior one. But, to add some variation, I'll provide an example. Take a look at these two players, both of whom were free agents in the 2016 offseason. Both are Centers, and the numbers below are their per-36 minute statistics in the season prior to their free agency:

Pretty similar, right? Well, the first player is Dewayne Dedmon, who ended up signing a 2-year, $6 million deal with the San Antonio Spurs. The second is Bismack Biyombo, who parlayed his performance into a 4-year, $70 million deal with the Orlando Magic. Now, you may like Biyombo's slightly better rebounding numbers or the fact that he's younger (24 to 27), but it's another thing to like him to the tune of $64 million dollars more. And the final results were stunning - the Spurs (as they often do) found the perfect role for the athletic Dedmon, resulting in a 5.2 Win Share season as their starting Center. Biyombo, buried in a mismatched Orlando depth chart, managed only 2.9.

Given the information above, it's easy to rag on the Magic for what seems to be a ridiculous overpay. But here's the thing - while experts questioned Biyombo's fit on a Magic roster already overstocked with big men, few had quibbles with the actual amount of the contract. It's just that the Spurs did a far better job at identifying value.

Think this is an outlier? I'll give you another example from last offseason, this time with two wing players:

I will preface this by saying - these two are very different players, even though they play the same position. The former signed a two year, $2.2 million deal. The latter - a four year, $70 million pact. The $70 million man is Kent Bazemore, and the other is Quincy Acy.

To be sure - neither player is a game-changer. But I wanted to pick them to illustrate the point that it's not just good organizations with solid management that find the Dewayne Dedmons of the world. Acy was signed by the Dallas Mavericks, but was eventually waived and picked up by the Brooklyn Nets, one of the least successful franchises over the past few seasons. Bazemore was resigned by the Atlanta Hawks, who are generally regarded as a solid organization through-and-through.

But, despite being paid $16 million more in his first season, Bazemore produced roughly equivalently to Acy. In 32 Nets games, Acy delivered 1.1 Win Shares, while Bazemore produced 2.1 in 73 games. 

So the point in all this? A rebuilding team should look to acquire role players by signing lower-tier free agents and keep away from the middle class. Sure, Bazemore's superior passing ability and experience as a higher-usage player might prove marginally more useful than a player of Acy's caliber. But a rebuilding team shouldn't be concerned about margins. Margins come into play in the playoffs, when an inch of space could mean the difference between a two-point victory and a one-point defeat. But for a team trying to rebuild its roster? There's little reason to make a $60 million-plus bet on that mattering.

Besides, a significant portion of what differentiates the middle and lower tiers in the NBA can be chocked up to playing time and notoriety. Biyombo earned much of his $70 million deal on the merits of his Herculean playoff performance against the Cleveland Cavaliers in the 2016 playoffs. Bazemore was one of the hot stories in the early going of the 2015-16 season because of his team's success and his rise from relative obscurity. But each player, though valuable, is limited and therefore not entirely dissimilar to players like Dedmon and Acy who simply hadn't received their opportunity yet. 

The rebuilding teams that I studied who repeatedly tried to sign mid-tier free agents as the bulk of their strategy have by-and-large failed to return to competitiveness. These teams include the Pistons, Magic, Pelicans, Knicks, and to a lesser extent the Atlanta Hawks, Dallas Mavericks and Phoenix Suns. There are success stories in this group - the Hawks and Mavs, specifically - and most of the other clubs have made the playoffs at least once since their rebuilds began. But none have sniffed championship contention, and most are handcuffed by a lack of financial flexibility brought on by a litany of mid-level player acquisitions.

So what does that mean if you're rebuilding this offseason? It means to stay away from the Jamychal Greens (5.8 Win Shares), Patrick Pattersons (3.3), and P.J. Tuckers (3.6) of the world, who have the combination of just enough hype and performance to create an eight-figure bidding war. Instead, try less heralded vets like Dante Cunningham (3.0), Raymond Felton (2.0) or even Thomas Robinson (0.9) and Reggie Bullock (1.0) and you'll probably receive comparable production, all the while saving considerable coin.

MIND YOUR TENDENCIES

A large part of rebuilding strategies is the Draft. That's why teams tank - the hope of drafting a hyper-talented player who's guaranteed to be on the team for the long haul at a (likely) discounted rate. So I looked at the players that these 20 rebuilding teams drafted over the years, 356 rookies in all. I especially looked at their advanced stats. I then looked at correlations between a player's first-year stats and their ultimate career performance (through the 2016-17 season).

The results were intriguing - the statistic with the highest positive correlation to career Win Shares was Offensive Rating, which measures the points a player produces per 100 possessions. The correlation between these two was 0.267 (a moderately positive value), and there were quite a few really good players at the high end of the spectrum - Kenneth Faried, Patrick Patterson, Greg Monroe, Nikola Jokic, Patrick Beverley, and Jimmy Butler among them. The other side of the coin was considerably less favorable. The average Offensive Rating of the group was 96.5 with a standard deviation around 20, so I looked at the list of players that had first-year O-ratings one standard deviation below the mean (which was about 77). Only Julius Randle turned out to be an above average NBA player, and his rating is skewed by the fact that he broke his foot mere minutes into his first season.

If Offensive Rating produced the best positive correlation, it would seem that Defensive Rating would produce the best negative correlation, right? After all, D-rating works in the exact opposite fashion as O-rating: the higher your number, the more points you give up per 100 possessions. So it would stand to reason that defensive turnstiles wouldn't last long, right? Well, there IS a negative correlation between first-year Defensive Rating and career Win Shares (-0.173), but it's not the leader. Coming in at just about 0.02 lower, the largest negative correlation is three-point rate (-0.199).

That seems ludicrous, right? Shooting has taken the league by storm, so why wouldn't players who shoot these shots more often have staying power? The answer appears to be two-fold - first, the relationship appears to have a heavy left-skew, as a lot of very good NBA players didn't shoot threes with great frequency in their rookie season. Above-average-to-great players such as Rudy Gobert, Anthony Davis, Greg Monroe, Andre Drummond, Tristan Thompson and Myles Turner all shot three-pointers with well-below average frequency in their first years. Even in the modern NBA, players can still have value without triples being in their repertoire. In the case of these players, they are all centers, lessening the need for them to have shooting as a part of their repertoire. But just because they didn't shoot doesn't mean they can't. Both Anthony Davis and Myles Turner developed into above-average marksmen for their position - but they also were able to use other skills in their first year to be valuable.

But players who shot threes at a high frequency in their first year should have done so because they were good at making that shot, right? Not necessarily. Just as someone who doesn't shoot threes at all might be considered offensively limited, so might someone who does almost exclusively. Included in the upper tier of the three-point rate group are a few players who are downright dreadful at draining trifectas. Dragan Bender, Justin Anderson, Nikola Mirotic, and Marcus Smart all exceeded one standard deviation above the mean in terms of three point rate, yet failed to clear 34% in their first season. A player who shoots threes too often might mean they have a limited offensive skillset and therefore are forced to take the shots they can most easily get off - the ones farthest from the basket.

While I don't believe that there is enough evidence to conclude that a high three-point rate should be worrisome, I do think there is enough of a correlation to be concerned with low Offensive Ratings. So what do you do if you're a rebuilding team that's drafted such a player? The best strategy might be to cut bait sooner rather than later. To illustrate, here are the players that were first round picks and had O-ratings of 90 or below in my dataset, along with their average Win Shares per season since coming into the league:

• Dragan Bender: -0.13
• Anthony Bennett: 0.16
• Clint Capela: 3.36
• Norris Cole: 0.36
• Kris Dunn: 0.25
• Henry Ellenson: -0.13
• Tyler Ennis: -0.13
• Dante Exum: 0.45
• Christian Eyenga: -0.17
• Gary Harris: 2.32
• Sergey Karasev: 0.3
• Doug McDermott: 1.66
• Fab Melo: 0.1
• Marcus Morris: 3.32
• Emmanuel Mudiay: -1.18
• Adreian Payne: -0.12
• Julius Randle: 1.59
• Austin Rivers: 1.05
• Thomas Robinson: 0.86
• Terry Rozier: 0.67
• Dennis Schroder: 1.9
• Rashad Vaughn: -0.4

Of those players, there might be five that I would regret trading after the first season (Harris, Morris, Randle, Schroder, Capela). Otherwise, their impact has been relatively minimal. Why did I mention only players drafted in the 1st round? Because this is where I believe a rebuilding team can salvage value. Historically, high draft picks (especially in the top 10) have retained their value a lot longer than similarly-performing players without the draft pedigree. Case in point - over the past few years, Emmanuel Mudiay has been mentioned multiple times in trade rumors as the Nuggets attempt to land a star, despite the fact that he's mostly fallen well below expectations (he has the worst average Win Shares of the above group). Why is he discussed as a cornerstone of any deal? Because teams overvalue the talent that made him the 7th overall pick in the draft. The same thing occurred with Anthony Bennett - he was involved in the deal that brought Kevin Love to Cleveland largely because he was a no. 1 overall pick. Teams that drafted players who've struggled in their first seasons can use the residual equity in their status as first-rounders in order to acquire a role player (or as part of a larger deal). Yes, this strategy could result in your team trading away a potentially great player for a merely capable one. But given the recent history of picks with poor first-year offensive performance, you're more likely to be selling high.

AN IMMOVABLE OBJECT

So far, I have (at least tangentially) discussed strategies that will help you with two of the three "championship indicators" I mentioned at the start of my mad ramblings. Using these strategies, you can hopefully get to a point where you have a) a superstar able to produce above 7.7 Win Shares consistently and b) enough average players - 5, to be exact - to be sufficiently deep. But I haven't touched on the defensive end of the floor. How can a team break at least 18 Defensive Win Shares to be in the contender conversation? It's a difficult question to answer. However, in studying Win Shares, one thing stood out - defense is more stable.

Across a number of different segments in the data I studied, Defensive Win Shares had a significantly lower standard deviation than Offensive Win Shares. For all players with at least two seasons of experience between 2007-08 and 2016-17, the average standard deviation for their offensive Win Shares over the course of their careers was 0.905, whereas the average standard deviation for Defensive Win Shares was 0.574. And this was not because of a significant difference in averages - players averaged 0.742 Offensive Win Shares per season in the dataset, and 0.822 Defensive Win Shares. Therefore, teams should be able to trust that their players' defense will stay more consistent throughout their careers.

Interestingly, players that have standout defensive years are more consistent, relatively speaking, than the population in general. When comparing every players' seasons between 2007-08 and 2016-17, the average Defensive Win Shares in a season was 1.244, and the standard deviation was 1.206, meaning that any season above approximately 3.66 would be considered statistically significant (a season exceeding two standard deviations from the mean). However, the players who had at least one season above 3.66 Defensive Win Shares had a career standard deviation of 1.321 and a mean of 2.713 average Defensive Win Shares. Compare this to the 0.592 standard deviation and 0.864 career average Defensive Win Shares of the whole group, and although the raw 1.321 is greater than 0.592, it is actually smaller to their relative averages (1.321 is 48% of 2.713; 0.592 is 68% of 0.864). This trend also holds true with Offensive Win Shares - the players who realized standout offensive seasons were actually more likely to be consistent than the overall group.

So, a team like the Denver Nuggets or Portland Trailblazers, who are in rebuilding mode as well as in dire need of defense (9.1 and 11.7 team Defensive Win Shares this past season, respectively) would be prudent to monitor the situations with the Los Angeles Clippers and the Atlanta Hawks this offseason. If either team loses their top free agents (Chris Paul, Blake Griffin, and/or JJ Redick for the Clippers, Paul Millsap and Tim Hardaway Jr for the Hawks), they may decide that a team wide sell-off is the way to go. In that scenario, the aforementioned defense-needy teams could potentially acquire either LA's DeAndre Jordan or ATL's Dwight Howard, as both are superlative defenders and exceeded 3.66 Defensive Win Shares last year. One caveat, though - both of their current teams are paying in excess of $20 million per year for these two players, and Jordan and Howard are (at the very least) in the midst of their athletic primes, if not declining. Teams like Portland and Denver should only consider such a move if they believe they are close to contending.

***

There is no one right way to conduct a rebuild. If you tear the entire roster apart after losing key players a la the 76ers or Orlando Magic, you might strike gold and draft the next LeBron James. Or, you might get Darko Milicic. If you choose to re-tool and develop, you might become the Atlanta Hawks, who won 60 games four seasons after losing almost half of their team's Win Shares and have the second-longest playoff streak in the NBA. Or, you might end up like Detroit Pistons - a team that has been mediocre and mostly out of the playoffs for the better part of a decade.

If you're going to do it right, though, you better get these elements first:

• An upper-tier point guard or leader of the offense;
• Understanding that big men, especially middle-tier big men, are generally more replaceable in the modern NBA than wings and point guards;
• Avoiding signing mid-market free agents, as the gulf between so-so and average in the NBA is not considerable, especially for a rebuilding team that doesn't expect to contend;
• Taking advantage of a market inefficiency associated with players with high draft pedigrees, and trading those who exhibit particularly worrisome traits (low Offensive Rating);
• Keeping good defensive players, as they're more likely to repeat their performance.

Start with these tenets, and you'll be well on your way to surviving your team's NBA rebuild.

***

(I want to thank everyone who stuck with me to the end of this article. It was a long and winding road, and we lost good people along the way, but we made it! As a token of my gratitude, I'd like to offer all of you stalwarts a BONUS SECTION! [Looks around excitedly, sees eye-rolls, quickly directs gaze back at the screen] Yes, there was ONE MORE strategy I noticed that was more successful than any. It was surefire, I tell you! So I present to my bonus observation...

SIGN LEBRON JAMES

Ok, so this section is a bit tongue-in-cheek. And yes, it's more an opportunity to fawn over the best player in the league than anything else. But seriously - LeBron James is really, really good at basketball. I studied 20 teams. Two of them actually won the NBA championship after their rebuilds began. Can you guess the common thread? Ah, good - I see that knowing smile on your face.

The Cavaliers began their rebuild when LeBron left for Miami, promptly went 97-215 in four seasons without him, and have gone to the Finals both years since he returned, winning once. The Heat rebuilt their roster when they acquired James, and won the championship twice in his four years with the team (LeBron was the MVP in both those years).

Granted, the Cavs and Heat did not win championships SOLELY because of LeBron. They were able to attract other superstars to come to their respective teams (largely because of James' arrival), and had the ownership willing to spend in order to realize championship aspirations (I wonder if LeBron had anything to do with their revenue in those years?). Actually, I've changed my mind. LeBron won those championships.

So, if you ever doubt the magnitude of a star in the NBA, look no further than number 23. Strategies and player development and shrewd management decisions are good, but LeBron is better.)

NOTES

• Here is a link to the spreadsheet I used: https://drive.google.com/open?id=1K58AjIEan1XM1dNca0Nb-1ripoLlM2f6HJLkvoaO1ik
• For the current season (2016-17), I copied over the statistics from basketball-reference.com while the regular season was still ongoing. As such, for all calculations involving this current season, I projected the number of Win Shares a player would produce, assuming that player produced at the same level for the rest of the season. Injuries were not accounted for.
• The 2011-12 season was a strike-shortened season, so I projected these results as well. Because the season was shortened to 65 games, I took all players' Win Shares and projected them to a full 82-game season (again, injuries were not accounted for).
• I calculated a player's expected Win Shares based on their prior year's Win Shares. So, if a player produced 4 Win Shares in the 2012-13 season, then I projected they would produce 4 Win Shares the following season.
• If a player was a rookie, I took the average amount of Win Shares players produced in their first year, relative to their draft position.
• If a player was traded during the prior season, I calculated their expected Win Shares by projecting their performance during whatever period of time they were on their current team. For example, if a player produced 1.5 Win Shares in 20 games after he was traded, I projected that out to an 82-game season.