For example, an alpha of 1.0 would place 100% of the weight of the player drafted next. Under an alpha of 0.4, the follo

Somewhat surprisingly, the best draft pick was actually a first overall selection (Karl-Anthony Towns). However, two out of the five best draft selections also came in the second round.,Since second round picks typically have much lower expectations, they can achieve high scores with performance less impressive than a typical top pick. In other words, pulling an All-Star out of the second round is usually rewarded much more than out of the top-5 picks.,It should be noted that some of the worst selections likely aren’t as bad as they look. Samanic and Garland have each only played one season in the NBA — not nearly enough to get a good view of a young player. There is a good chance their scores will trend more towards the middle of the pack as time passes.,Therefore, this analysis will compare draft picks against the players that were taken after them. If those players performed better, then this indicates that better options were readily available. If those players performed worse, this means that the team made the best of what they had. To measure performance, I will use Win Shares.,All data was collected from Basketball Reference. Some adjustments were made around which teams draft picks were attributed to. While a player is officially listed under the team that drafted them, if they were traded on draft night or any time before the beginning of the next season, I attributed the draft pick to be under the team that they were traded to. Any trades after that point did not affect attribution.,The very best draft picks have a score of 4.0 while worst-performing picks have scores below -2.0. About one-quarter of all draft picks had a score close to zero— largely due to late draft picks that had low expectations and correspondingly low-impact performance.,In this analysis, I have created a metric that scores each draft pick by comparing them against the players taken after them (the full methodology is available at the end). After applying this over the last 10 years (the drafts from 2010 to 2019), we can now measure which teams have been successful in their drafting and what teams have not.,The results of this analysis also empirically validate the vital link between drafting success and winning. The relationship between team drafting scores and winning is strong and statistically significant — an increase of 0.1 in team draft score is associated with 3.7 more regular season wins in the last NBA season. Of the fourteen teams with an above average draft score, ten of them made the playoffs last year. Among the bottom five teams, none made the playoffs.,The central idea behind this analysis is relatively simple. When judging teams drafting success, draft picks should be measured against the alternatives that were available in that draft position during that year. A team that drafted fifth overall shouldn’t be penalized because the team drafted fourth overall scored a great prospect — that player wasn’t available at fifth overall. Additionally, it doesn’t matter if a first overall pick performs weakly relative to the typically high expectations of a top draft pick — it only matters if better alternatives were on the table that year.,Over the past ten years, the Toronto Raptors have been the best drafting team by a wide margin. This has contributed to seven straight post-season appearances, capped with an NBA Championship in 2019. Meanwhile, Cleveland ranked as the worst drafting team.,A couple of outliers are noted in this chart. First, despite the relatively low expectations of a 30th overall pick, Jimmy Butler went on to become one of the most impactful players in the NBA. Conversely, Thomas Robinson enormously under-performed as a 5th overall pick.,This equation does require one arbitrary choice: the alpha value (which must be between zero and one). A greater alpha places more emphasis on players drafted directly after while a lower value smooths the weighted average more evenly over a larger number of players.,Where E represents the expected Win Shares at draft position d, alpha represents the smoothing factor, and W represents the actual Win Shares achieved at draft position d 1 (the player taken directly after them). To initialize the function the 60th overall pick is assigned an Expected value of zero.,If you can’t find what you’re looking for? THEN expand to your next state or province, and then finally the country nearest yours. Sure, butter from Ireland is a lovely indulgence, but so is the butter from your local dairy producer, at half the price.,A couple of the result are fairly surprising. While Golden State and San Antonio are both considered to be strong drafters, this analysis ranks them as below average. In reality, most of their home-run draft picks occurred more than ten years ago (outside of this analysis). Their drafting in more recent years has been less impressive.