<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Lewis Pipkin on Medium]]></title>
        <description><![CDATA[Stories by Lewis Pipkin on Medium]]></description>
        <link>https://medium.com/@pipkinlewis?source=rss-ff1d8e36f3a2------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*vN8KU9kayPL7XUr5MROqhw.png</url>
            <title>Stories by Lewis Pipkin on Medium</title>
            <link>https://medium.com/@pipkinlewis?source=rss-ff1d8e36f3a2------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 16:06:37 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@pipkinlewis/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Every time the Vancouver/Memphis Grizzlies have helped the Los Angeles Lakers]]></title>
            <link>https://medium.com/@pipkinlewis/every-time-the-vancouver-memphis-grizzlies-have-helped-the-los-angeles-lakers-fbf0081d2f12?source=rss-ff1d8e36f3a2------2</link>
            <guid isPermaLink="false">https://medium.com/p/fbf0081d2f12</guid>
            <category><![CDATA[nba]]></category>
            <category><![CDATA[los-angeles-lakers]]></category>
            <category><![CDATA[lakers]]></category>
            <category><![CDATA[memphis-grizzlies]]></category>
            <category><![CDATA[vancouver-grizzlies]]></category>
            <dc:creator><![CDATA[Lewis Pipkin]]></dc:creator>
            <pubDate>Fri, 10 Sep 2021 20:00:08 GMT</pubDate>
            <atom:updated>2021-09-10T20:00:08.937Z</atom:updated>
            <content:encoded><![CDATA[<p>The Grizzlies and Lakers have made four trades with each other in the Grizzlies’ short history, and there have been a handful of times where the Grizzlies waived a player only for them to go straight to the Lakers. I won’t go as far as to say that the Grizzlies are a feeder team for the Lakers, but every move they have made with each other has pretty massively favored the Lakers. Here are all of the trades:</p><h3>July 16, 1996</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mrcdXjQ_TtpGXZ3BQOZyvQ.png" /></figure><p>For the low, low price of two second-round pick swaps, the Vancouver Grizzlies received two viable NBA players to fill out their roster that was still full of players from the expansion draft and the Continental Basketball Association. The Lakers used the resulting cap space to sign Shaquille O’Neal.</p><h3>February 1, 2008</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0X60Ch9bAAlEdw1dP0wXdg.png" /></figure><p>This is the big one. At the time, obviously, this looked extremely one-sided. This was a huge red flag to the rest of the league that something was going on.</p><p>“What they did in Memphis is beyond comprehension,” said Popovich. “There should be a trade committee that can scratch all trades that make no sense. I just wish I had been on a trade committee that oversees NBA trades. I’d like to elect myself to that committee. I would have voted no to the L.A. trade.”</p><p>The Grizzlies bottomed out after Pau broke his foot during a FIBA game, and even after his return, there was no clear path back to the playoffs. Team owner Michael Heisley was allegedly hoping to cut costs to make the franchise look better to a potential buyer (Gasol had a huge salary and the players they received in return were either cheap or on short-term deals), but Grizzlies GM Chris Wallace denied that accusation, saying that this was just the best deal he was offered. The Grizzlies did not get back to the playoffs until 2011, by which time had won two championships with Gasol as a key player.</p><p>In the long-term, of course, the Grizzlies made out just fine with this deal, as Marc Gasol grew into perhaps the greatest player in franchise history, but the Lakers got two rings for a pretty low cost.</p><h3>February 18, 2009</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*clt17VCr8HOP5cD2iVYv_w.png" /></figure><p>A pure salary dump of Mihm’s $2.5 million contract.</p><h3>September 10, 2021</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eioaon3WcTgmzNahcoDrTA.png" /></figure><p>This trade saves the Lakers about $10 million on their luxury tax bill, per Woj. The Grizzlies do get a second-round pick, which is of minor importance because this front office loves to make draft-night deals involving second-round picks. The Lakers also generated a $2.7 million trade exception. Grizzlies fans got to be happy for about five seconds until reading the end of Woj’s tweet where he clarified that the Grizzlies would waive Gasol.</p><p>While the Grizzlies have benefitted to varying degrees from three out of the four trades, the Lakers have gotten five rings as a direct result of these moves, and a sixth one if you count the Grizzlies waiving Dwight Howard and Avery Bradley (who did not play in the bubble) in the summer of 2019, allowing them to sign with the Lakers. (The other two times this happened are Byron Scott in 1996 and, just recently, Rajon Rondo.)</p><p>Are the Grizzlies a feeder team for the Lakers? You decide.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4q-pP3__bcA3UJS1VYOnLA.jpeg" /><figcaption>Photo by Joe Murphy/NBAE via Getty Images</figcaption></figure><p><em>Trade information is courtesy of Basketball Reference. Salary information is courtesy of HoopsHype.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fbf0081d2f12" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Passing Deep Dive: Creating a new metric for evaluating passers in the NBA]]></title>
            <link>https://medium.com/@pipkinlewis/passing-deep-dive-creating-a-new-metric-for-evaluating-passers-in-the-nba-bf16b76fff3d?source=rss-ff1d8e36f3a2------2</link>
            <guid isPermaLink="false">https://medium.com/p/bf16b76fff3d</guid>
            <category><![CDATA[nba]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[basketball]]></category>
            <category><![CDATA[statistics]]></category>
            <dc:creator><![CDATA[Lewis Pipkin]]></dc:creator>
            <pubDate>Mon, 27 May 2019 17:22:32 GMT</pubDate>
            <atom:updated>2019-05-27T17:22:32.036Z</atom:updated>
            <content:encoded><![CDATA[<h3>Introduction</h3><p>I spent a while this past week checking out all of the publicly available passing statistics from NBA.com with the intention of developing a new passer rating. This is intended to be somewhat of an open exhibition on the thought process of developing a new statistic. It’s not perfect, as every statistic is inherently biased toward what can be measured, but I think it’s pretty cool nonetheless. I will not claim to be an expert when it comes to crafting new metrics, but I wanted to try my hand at it. <strong>If you just want to see how my passer rating is calculated</strong> — skip to Part 2, down below.</p><p>So I have these two spreadsheets (available on my <a href="https://github.com/lewispipkin/NBA/tree/master/data">Github</a>) — passesMaster.csv and passing.xlsx. The former is player-level data from NBA.com which includes an estimate of the number of shots taken as a result of passes from one player to each of his teammates. That was aggregated via <a href="https://github.com/abresler/nbastatR">NBAstatR</a>, the ever-useful package by Alex Bresler.</p><p>The latter is hand-pulled data from NBA.com. It took copy-and-pasting raw JSON data from the Google Chrome Developer Tools interface. I’m not going to pretend to know what else you can do with Developer Tools, but that data was acquired thanks to <a href="https://www.reddit.com/r/sportsbook/comments/59hsno/best_website_to_pull_nba_statistics_into_excel/d98o04w/">an old, very helpful Reddit comment</a> by u/cmai3000. There is a third spreadsheet, basestats.xlsx, which comes in later in the <a href="https://github.com/lewispipkin/NBA/blob/master/code/pass.R">code</a> — it’s also transformed JSON data in kind of a strange format, but I end up only using the players’ names and their total turnovers from that one.</p><h3>Part 1: A few passing metrics</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5qAJex8ara28_H0hxc4GGQ.png" /><figcaption>Passes Made per game | Free throw assists per game | Secondary assists per game</figcaption></figure><p>Naturally, ball-dominant players are going to lead in passes made per game. I found it interesting that Jamal Murray lead the league in secondary assists — but 44.2% of his passes are to Nikola Jokic. Lowry, Rubio, and Murray especially are in offenses with more ball movement, especially after the All-Star break, where the Raptors, Jazz, and Nuggets are 5th, 6th, and 7th in passes per game.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fTkqloIO1wvcRWQiuKm8bQ.png" /><figcaption>Potential assists per game | Assist points created per game</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kxLoJvObfgSSY3o_MbQaUg.png" /><figcaption>Productive passes per game | Productive passes/Total passes made | Assist points created/Total passes made</figcaption></figure><p>Troy Caupain, the Cincinnati Bearcats’ all-time leader in assists at 515 assists in 4 years, dominated the entire league in two metrics. While he was a great college point guard, his 16 NBA minutes have been nothing short of legendary, posting a sizzling 30.7 PER and almost doubling <a href="https://socialnewsdaily.com/wp-content/uploads/2013/08/russell-westbrook.png">Mr. Triple Double’s</a> Assist Points Created per Pass figure. However, he and Mr. Alkins are prime examples of why I’m going to filter based on minutes played.</p><h3>Part 2: Passer Rating</h3><p>The formula took a while to develop, but I wanted the following to be true of this metric:</p><p>1. It is roughly normally distributed.</p><p>2. It can be applied to players no matter how many games they played in the season. <em>Note: (I imposed a total minutes played restriction, but the formulae work for any player who recorded at least 1 turnover.)</em></p><p>3. It still directly rewards primary assists slightly more than secondary assists, potential assists, and free throw assists.</p><p>Below are the component formulae for my version of NBA passer rating.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*77b1D8eowyRuuZQTvdM3gA.png" /></figure><p>Then, thanks to <a href="https://stats.stackexchange.com/questions/46429/transform-data-to-desired-mean-and-standard-deviation">this StackExchange thread</a>, I was able to transform the data into a more digestible number — instead of X ~ N(1.42,0.27²), wouldn’t it be nice for it to be something easily interpretable? I decided to go with X ~ N(10,2²) so we could easily say, without a lot of thought, “11.5? That guy’s a pretty good passer.” vs. “7? He needs to improve.” That’s a hell of a lot easier than remembering 0.88 and 1.97 cutoffs for those who’d be statistically significantly great or poor passers according to this metric.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*myO0bW-npNlAbj7A4eHkbQ.png" /></figure><p>So, as I intended, passer rating is roughly normally distributed, with a mean of 9.99, standard deviation of 1.99, and a skewness of only 0.12. <a href="https://public.tableau.com/profile/lewis.pipkin#!/vizhome/PasserRatingPlayerTable/PlayerTable?publish=yes">Here</a> is a link to the passer ratings of all players who qualified based on the minutes restriction I imposed, ≥ 331.8 minutes played (20th percentile).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/606/1*QkXx0u1ftzHGogrOmaNIag.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/598/1*U_hpPU7TOTANr5Bw1FPsYg.png" /></figure><p>Then, we come to the top 10 and bottom 11 passers based on this rating. (I wish it were top and bottom 10, but it’s those below the 2.5th percentile and above the 97.5th percentile.)</p><p>There are very few, if any, surprises in either of these lists, I feel. Russ gets a lot of flack for stat padding, but his passing, along with CP3’s, is truly elite. Both Timberwolves point guards make the top 10 (Derrick Rose was not far outside of this list, with a 13.3 passer rating), and we know how good LeBron, John Wall, and Rajon Rondo’s court vision is. Eric Bledsoe was the only moderate surprise to me, but he’s 3rd in usage among Milwaukee’s high-scoring starting lineup, which would no doubt help him rack up hockey assists.</p><p>When we look at the bottom of the ladder, we see it’s all low-assist big men; guys who aren’t in playmaking roles. The fact that my statistic didn’t rate JaVale McGee (who has a career-high of 3 assists per game and averages 0.4 per game for his career) as a high-quality passer gives me some more confidence in my work.</p><h3>Part 3: Passer Rating leaders by position</h3><p><strong>Average by position</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RA29se2dMOHLPvw5jnZ1fw.png" /></figure><p><strong>Point Guard</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rEUIlEaswU2Ci1RF1MK4WA.png" /></figure><p><strong>Shooting Guard</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ain1ysbUJl7X9PJFRU1kJw.png" /></figure><p><strong>Small Forward</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rYLhsz7ZTLyhFs65l4FCXA.png" /></figure><p><strong>Power Forward</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DQc16Y5V2lv63mO2Z4RNig.png" /></figure><p><strong>Center</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jatmeGfXU9KI8x400VzwPw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/494/1*2I7sFbyHXhCLIyUh5pAmsw.jpeg" /><figcaption>Oct. 19, 2018 — Source: Harry How/Getty Images North America</figcaption></figure><p><em>Stats courtesy of NBA.com and abresler/NBAstatR. Analysis and visualizations done in R and Domo.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bf16b76fff3d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[NBA Win Projections 2018–19]]></title>
            <link>https://medium.com/@pipkinlewis/nba-win-projections-2018-19-92fcb4dbd4c3?source=rss-ff1d8e36f3a2------2</link>
            <guid isPermaLink="false">https://medium.com/p/92fcb4dbd4c3</guid>
            <category><![CDATA[basketball]]></category>
            <category><![CDATA[nba]]></category>
            <category><![CDATA[sports-betting]]></category>
            <category><![CDATA[statistical-analysis]]></category>
            <category><![CDATA[rstats]]></category>
            <dc:creator><![CDATA[Lewis Pipkin]]></dc:creator>
            <pubDate>Thu, 04 Oct 2018 16:46:47 GMT</pubDate>
            <atom:updated>2018-10-04T16:46:47.885Z</atom:updated>
            <content:encoded><![CDATA[<p>After a good bit of data collection/creation yesterday, I’ve created a win projection model that I’m pretty happy about. A few of the variables that go into this model include:</p><p>number of wins for each of the past 4 seasons, <br>number of players added in the lottery each of the past 4 drafts, <br>if there is a new coach at the helm, <br>estimated man games lost due to injury/trade for each of the past 2 seasons*, <br>number of players with a salary of &gt;$15M,<br>a franchise dysfunction index**, and<br>a tank propensity.</p><p>*very, very rough estimate, but I feel pretty good about my method.<br>**highly subjective and not at all scientific.</p><p>As sort of a rough proof-of-concept for estimated man games lost, we see a negative trend with respect to wins in both years:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IS3aWP-QYmYkllW134NA1w.png" /><figcaption>2017–18</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*q29ZpiNUlzjTGMvoYIcbyQ.png" /><figcaption>2016–17</figcaption></figure><p>For teams like the Celtics who lost 81 games of Hayward or the ’17 76ers (heh) who lost 82 games of Simmons, I included the full amount because those injuries changed the way they played for the entire season vs. something like a role player missing 20 games, in which case I’d count that as around 10 because the team would adapt. <br>(Didn’t the Celtics and Sixers adapt to their injuries? Of course, but they became a fundamentally different team playing with Tatum in a star role and TJ McConnell/Sergio Rodriguez starting at point, respectively.)<br>For trades, I included a slight buffer for the adjustments the team would have to make afterward, but nothing too serious. Most of the time that I was collecting this data, I stuck with players who averaged over 20-ish minutes a game or players that I knew played a role on each team, with some exceptions here and there.</p><p>As I mentioned, franchise dysfunction is a subjective measure (as is tanking propensity, really), but I keep my finger on the pulse of all 30 teams and I feel qualified to give a rough subjective estimate of the state of things for each franchise. When projecting something like wins, I feel that it’s alright to inject a small amount of opinion into the model, because there are nuances to the league that a model won’t ever know unless you tell it. We’ll call it an attempt at an instrumental variable.</p><p>Right now, some of the leaders in dysfunction include the Timberwolves (of course), the Knicks, the Kings, and the Wizards (Dwight, Wall, and Beal). As for tanking, we can be pretty sure that the Knicks, Kings, Magic, Hawks, and the Brooklyn Nets (who have their pick for the first time in years!) will all be Vyin’ for Zion™ come lottery time, so their scores are way higher than even the Cavs or Mavericks, each of whom will probably at least try to make the playoffs.</p><p>In addition, I included a win volatility calculation that’s computed similarly to RMSE. It behaves as you’d expect; teams with a relatively stable win total year-to-year have a lower volatility. Sacramento and Cleveland have the lowest from 2014–15 to 2017–18. Philly, Atlanta, Memphis, and Minnesota have some of the highest, which shouldn’t be all that surprising considering each of their respective wild shifts in fortune in the past year.</p><p>So, without further ado, here are my projections:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TXsqkGicScsZJTciccfLGg.png" /></figure><p>While this is probably going to be very wrong, I was grateful these standings didn’t exactly mimic the standings if Bovada’s <a href="https://www.bovada.lv/sports/basketball/nba-season-props">over/unders</a> were exactly true. One thing that concerns me is that most of my projections go under — even if the league does turn out to be a bloodbath this year, which I think it will, there will still be more than 5 teams to eclipse 50 wins. Golden State, Houston, Boston, and maybe Toronto will each flirt with 60 wins, and I can see the Bucks, Jazz, Thunder, and Spurs all cresting 50. Of course, there’s always more data to be collected — one thing I’d like to include in a future version of this is roster continuity or returning minutes, but the latter would be really difficult to project before a season, especially with rosters still in flux vis-à-vis training camps and the impending Jimmy Butler trade. <a href="https://www.basketball-reference.com/friv/continuity.html">Basketball-Reference calculates it</a> after the season, but predicting how many minutes will return before the season is essentially a crapshoot in my eyes.</p><p>Anyway, that’s what I’ve got and I think it’s not too bad. To me, the thought of an OKC vs. LeBron’s Lakers playoff matchup is the best part of this, and I hope we get to see that, just maybe not in the first round.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*u56KU1oQ0fjSnR8LUHXmjg.png" /><figcaption>(Oct. 1, 2018 — Source: Harry How/Getty Images North America)</figcaption></figure><p><em>Many predictors in this model are courtesy of Basketball-Reference.com.</em></p><p><em>Modeling and visualization done in R/ggplot2.</em></p><p><em>The terminology “Man Games Lost” has already been </em><a href="https://www.mangameslost.com/"><em>used</em></a><em> </em><a href="https://twitter.com/mangameslostnba?lang=en"><em>by</em></a><em> </em><a href="http://www.nathancurrier.com/"><em>Dr. Nathan Currier</em></a><em>; it’s hopefully a close estimate of what his measurement is, but it is certainly not as accurate.</em></p><p>Atlanta: 23.5. Over.<br>Boston: 58.5. Under.<br>Brooklyn: 32. Over.<br>Charlotte: 35.5. Under.<br>Chicago: 29.5. Under.<br>Cleveland: 30.5. Over.<br>Dallas: 34.5. Under.<br>Denver: 47.5. Under.<br>Detroit: 38. Under.<br>Golden State: 63.5. Under.<br>Houston: 55.5. Under.<br>Indiana: 47.5. Under.<br>LAC: 36.5. Over.<br>LAL: 48.5. Under.<br>Memphis: 33.5. Over.<br>Milwaukee: 48. Under.<br>New Orleans: 45.5. Under.<br>New York: 29.5. Over.<br>Orlando: 31. Under.<br>Philadelphia: 54. Under.<br>Sacramento: 26. Over.<br>San Antonio: 44.5. Over.<br>Toronto: 55.5. Under.<br>Utah: 49.5. Under.<br>Washington: 44.5. Under.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=92fcb4dbd4c3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Using machine learning to diagnose NBA contract amount]]></title>
            <link>https://medium.com/@pipkinlewis/using-machine-learning-to-diagnose-nba-contract-amount-5b163cca648d?source=rss-ff1d8e36f3a2------2</link>
            <guid isPermaLink="false">https://medium.com/p/5b163cca648d</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[statistics]]></category>
            <category><![CDATA[basketball]]></category>
            <category><![CDATA[nba]]></category>
            <dc:creator><![CDATA[Lewis Pipkin]]></dc:creator>
            <pubDate>Sun, 29 Apr 2018 15:57:49 GMT</pubDate>
            <atom:updated>2018-04-29T15:57:49.120Z</atom:updated>
            <content:encoded><![CDATA[<p>NBA players’ salaries are often hotly contested. Who could forget the wild summer of 2016, where NBA teams spent so much money that the projected salary cap for this season <a href="http://bleacherreport.com/articles/2717210-2017-18-nba-salary-cap-projection-reportedly-reduced-to-99-million">dropped</a> from $108 million to $101 million to $99 million (among other factors like the lower amount of playoff games last postseason)? That offseason gave us plenty of ammo to make fun of teams:</p><ul><li>Knicks (4 years/$72 million for Joakim Noah)</li><li>Lakers (4 years/$64 million for Timofey Mozgov AND 4 years/$72 million for Luol Deng)</li><li>Grizzlies (4 years/$94 million for Chandler Parsons)</li></ul><p>And several other teams with lapses in judgement given the ludicrous projections for future seasons’ salary caps in light of the <a href="https://www.si.com/nba/2014/10/05/new-nba-tv-deal-worth-24-billion">big TV deal</a> that came into effect in 2016. Some of these contracts are included in the forthcoming analysis, but I’ve removed a little over 100 players who fail to meet certain playing time criteria (i.e. bottom 10% in games played or minutes played). Noah and Deng aren’t included, but Mahinmi and Biyombo are, for example. The dataset contains all of the statistics that can be found in the “Per Game” section and “Advanced” section on <a href="https://www.basketball-reference.com/players/s/simmobe01.html">Basketball-Reference.com.</a></p><p>I merged that dataset with a <a href="https://www.kaggle.com/koki25ando/nba-salary-prediction/data">salary information dataset from Kaggle</a>. These are just the 2017–2018 salary numbers, so it’s not a subset that received their contracts in a certain offseason, and it’s the actual amount, not average over the life of the contract. The resulting dataset (available with my R code on <a href="https://github.com/lewispipkin/NBA">Github</a>) has 413 entries, and per standard ML procedure, I split the data into a training set and a test set. I fit two random forest models, one with 500 trees and the other with 1000, and the 500-tree model performed better on the test set, so I moved forward with that model’s predictions.</p><p>(Note: if you need an explanation of the random forest algorithm, go <a href="https://medium.com/@Synced/how-random-forest-algorithm-works-in-machine-learning-3c0fe15b6674">here</a>, <a href="https://medium.com/@williamkoehrsen/random-forest-simple-explanation-377895a60d2d">here</a>, or read the section about it in James, Witten, Hastie, and Tibshirani’s <a href="http://www-bcf.usc.edu/~gareth/ISL/">Introduction to Statistical Learning</a>. But essentially, it’s an extremely powerful model that can make accurate predictions while combatting overfitting.)</p><p>Once I used that model to predict on the entire dataset, I got the following results:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/744/1*VK9MHTp8SFYNlxrby-EGvA.png" /><figcaption>Top 7 largest overpays</figcaption></figure><p>We see some interesting results, some that are backed up by common sense. Mike Conley and Stephen Curry (as a Grizzlies fan, it feels good to use them in the same sentence) failed to live up to their lofty salaries this season due to injuries. Per the variable importance plot for this model, some of the most important variables for both decrease in MSE and node impurity were number of games started (12 and 51, respectively) and minutes played, so it makes sense that it was hard to live up to those price tags compared to other players in the dataset who weren’t injured. The same goes for Paul Millsap.</p><p>Greg Monroe’s salary is a bit inflated because he’s being paid by Phoenix and Boston both, but he’s still underperformed this season. Ryan Anderson and Solomon Hill were part of that Summer ‘16 shopping spree, and Al Horford is way more valuable than the story his stats can tell, but his contract amount is still pretty high (especially considering he’s paid $6 million more than his teammate Kyrie Irving, who’s considered to be underpaid).</p><p>We’ll look at the top 7 underpays as well:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/878/1*6YTVEydQ2ZtE_Pxg_RDyag.png" /><figcaption>Top 7 largest underpays</figcaption></figure><p>No real surprises here, a lot of high-performing role players or players on rookie-scale contracts, along with the perennially underpaid 6th Man of the Year candidate Lou Williams, who locked himself into <a href="http://www.nba.com/article/2018/02/07/report-la-clippers-lou-williams-working-contract-extension">an extremely Clippers-friendly long-term extension</a> last February.</p><p>While I won’t insert more graphs, it’s interesting to note the names that pop up from a similar analysis with the natural logarithm of salary (and, of course, a different random sample of predictors at each node in each tree of the random forest).</p><p>Notable overpays: Otto Porter, Chandler Parsons (duh), Danilo Gallinari, JJ Redick.</p><p>Notable underpays: Will Barton, Michael Beasley, Emeka Okafor, Clint Capela.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*on8cyS-NZ5Iqu-XvZ_3stw.jpeg" /><figcaption>Oct. 8, 2017 — Source: Kevin C. Cox/Getty Images North America</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5b163cca648d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Don’t Forget About Dre]]></title>
            <link>https://medium.com/@pipkinlewis/dont-forget-about-dre-1d7fb1ebae6e?source=rss-ff1d8e36f3a2------2</link>
            <guid isPermaLink="false">https://medium.com/p/1d7fb1ebae6e</guid>
            <category><![CDATA[nba]]></category>
            <category><![CDATA[basketball]]></category>
            <category><![CDATA[detroit]]></category>
            <category><![CDATA[detroit-pistons]]></category>
            <category><![CDATA[lebron-james]]></category>
            <dc:creator><![CDATA[Lewis Pipkin]]></dc:creator>
            <pubDate>Thu, 20 Jul 2017 14:01:40 GMT</pubDate>
            <atom:updated>2018-02-01T17:03:08.519Z</atom:updated>
            <content:encoded><![CDATA[<h4>People continue to overlook Andre Drummond. His skill-set and production will go a long way in bringing the Pistons back to the playoffs.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*qJITFxov1Qp4NZ_DRS5NcA.jpeg" /><figcaption>Gregory Shamus/Getty Images North America</figcaption></figure><p>We’re entering the 4th season of the Stan Van Gundy era in Detroit. So far we have seen two losing seasons for Stan as a head coach and only one winning season that culminated in getting swept in the first round against the Cavaliers (although, a couple of those games could have gone either way). Stan has made some positive changes so far:</p><ul><li>Waived Josh Smith</li><li>Traded a package centered around Kyle Singler for a disgruntled Reggie Jackson (this was the trade that landed Enes Kanter in OKC)</li><li>Acquired Marcus Morris and, earlier this month, flipped him to Boston for a season of Avery Bradley’s excellent defense and 3-point shooting (and at a great price- only $8.8 million)</li><li>Brought in Tobias Harris</li><li>And just this offseason, he drafted chronic bucket-getter Luke Kennard and avoided overpaying for Kentavious Caldwell-Pope.</li></ul><p>These are <em>good</em> moves, but we haven’t seen Detroit make any <em>great</em> moves. When was the last time we heard Detroit come up as a possible destination for a marquee free agent? (I can’t answer that, as extensive searches on Twitter didn’t yield much.) The signings that the Pistons made this offseason (Langston Galloway, Reggie Bullock, and Anthony Tolliver, to name a few) don’t exactly move them into contender territory — although, in this era, that’s becoming more and more difficult.</p><p>But, will this season be the first since 2008 that the Pistons win a playoff series? It depends on a number of things like potential match-ups, but for the most part, the fate of the Detroit Pistons rests squarely on one man’s ridiculously hairy shoulders.</p><p>Coming out of his only season of college basketball, Andre Drummond was a solid player for legendary coach Jim Calhoun’s final season at Connecticut. Drummond averaged 10 points, 7.6 rebounds (3.35 offensive), and 2.67 blocks per game as he started most of the season. Drummond was seen as a project — his struggles at the free throw line were well-documented already as he shot an abysmal 29.5% from the line on 2.6 attempts per game. He had a variety of concerns, including his offensive ability and dedication to the game.</p><p>Drummond, however, is an exceptional talent at rebounding, especially on the offensive glass, and isn’t half-bad at scoring off of those rebounds either. In the 2016–2017 season, Drummond averaged 4.4 second-chance points per game, which accounted for 32% of his total points. Furthermore, 3.5 of his points per game are from putbacks, making him the third best putback scorer in the league last season.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fstreamable.com%2Ft%2Ftmslb&amp;url=https%3A%2F%2Fstreamable.com%2Ftmslb&amp;image=https%3A%2F%2Fcdn-e2.streamable.com%2Fimage%2Ftmslb.jpg%3Ftoken%3D1501554007_33cea9310c0b55b0cfb64e24f5b193fdcb60fc1f&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=streamable" width="1280" height="720" frameborder="0" scrolling="no"><a href="https://medium.com/media/1b3d52cbabf599f93e2aca113af4bd92/href">https://medium.com/media/1b3d52cbabf599f93e2aca113af4bd92/href</a></iframe><p>That isn’t surprising, considering Drummond was the league leader in offensive rebounds and total rebounds (he was third in defensive rebounds to DeAndre Jordan and Hassan Whiteside).</p><p>We’ve seen him develop more offensively and he’s gotten better in the post, with 3 of his 13.6 points per game this season coming from there.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fstreamable.com%2Ft%2F3hx4t&amp;url=https%3A%2F%2Fstreamable.com%2F3hx4t&amp;image=https%3A%2F%2Fcdn-e2.streamable.com%2Fimage%2F3hx4t.jpg%3Ftoken%3D1501554535_9aa768429548931f4dfa38c64ca00dc10cfb36c8&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=streamable" width="1280" height="720" frameborder="0" scrolling="no"><a href="https://medium.com/media/95da74ae78f6244b5d309c414a67d937/href">https://medium.com/media/95da74ae78f6244b5d309c414a67d937/href</a></iframe><p>Drummond’s rebounding and boxing out help his team, as well. Detroit was 4th in rebounds per game last season with 45.7 per game, and 8th in offensive rebounding at 11.1 per game. This leads to more points; the correlation between offensive rebounding and second chance points is around 0.95. As we all know, <a href="https://xkcd.com/552/">correlation does not equal causation</a>, but here we can kind of let that go — how, other than offensive rebounding, can someone get second-chance points? Detroit also allows the fewest second chance points, only 10.1 per game, which is presumably aided by their rebounding in some way (although, at the team level, total and defensive rebounding are essentially uncorrelated with opponent second chance points).</p><p>It’s not all good for the Pistons, however — Drummond has a -6.7 net rating, and I don’t really need to talk about his free throw shooting. And, when I was doing an analysis for <a href="https://94feetreport.com/happy-nba-summer-everyone-c5887dd85bf5">an article I wrote earlier this month</a>, I found that Drummond was <a href="https://docs.google.com/spreadsheets/d/1X0-DOYG4sTo2jmYXHVodoyphT-tatxQo_uHXZmjwKOs/edit?usp=sharing">overpaid by over $2.4 million</a> based on his performance.</p><p>While his game is certainly not impervious to criticism, it’s hard to say that Andre Drummond isn’t a good player. He’s been an All-Star and All-NBA third team before. If you want your team to win, he shouldn’t be your first option, or probably even second, but he’s a rebounding and putback machine; if given the space to operate without such a large load offensively, he can push a team over the edge. (Imagine him on the Cavs in place of Tristan Thompson.)</p><p>Now, of course, Detroit’s success is not solely dependent on Drummond. While Reggie Jackson’s name came up in trade rumors, as of right now it seems he will remain as the starting point guard this season. He needs to return to his 2016 form to give the team a fighting chance in the playoffs, because while Ish Smith is a fair backup, he is not a starter on a team that could win a playoff series. Tobias Harris is a good, fairly efficient scorer (although I’d like to see his 3-point shooting improve a bit). It remains to be seen who will start in place of Marcus Morris at the 3 (if I had to guess, Stanley Johnson), but Avery Bradley is an upgrade over Kentavious Caldwell-Pope.</p><p>Luckily for Detroit, they are not in the stacked Western Conference. Detroit was the 10 seed last season, and a few teams in front of them have gotten much worse, especially Atlanta, Chicago, and Indiana. Philadelphia is gaining on Detroit quickly, though, and Charlotte will be better this year as well. If Miami is the same team they were in the second half of last season, they’re a shoo-in for the playoffs. Cleveland, Boston, Washington, and Toronto will all be good again, and the Bucks will look to take another step forward.</p><p>Detroit is in a good position to compete, but it is hard to imagine they will crack the top 4 and get home court advantage in the playoffs. Home court advantage aside, if Detroit has a match-up with any of Miami, Toronto or Washington in the playoffs, they will have a chance. It’s definitely possible for them to finish above .500 this season. Last season, being just two games over .500 gave the Hawks the 5th seed. This would likely give Detroit a match-up with one of those three teams. Avoiding a bottom seed and an inevitable match-up with LeBron’s Cavaliers or the Celtics is Detroit’s roadmap to success.</p><p>All in all, Drummond and his Pistons have a good shot at a successful season. His offensive game is developing slowly but surely, and his rebounding and ability to pounce on putback opportunities make him a pain for opposing big men. While the Eastern Conference Finals are likely out of reach, a playoff series victory is certainly within reach when you have a capable center like Drummond on your side, free throw shooting be damned.</p><p><em>All statistics courtesy of Basketball Reference and NBA.com. All salary figures courtesy of Spotrac. All video courtesy of NBA.com.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1d7fb1ebae6e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Does Paying More for Players Get You More Wins?]]></title>
            <link>https://medium.com/@pipkinlewis/does-paying-more-for-players-get-you-more-wins-a8ab108a1c8c?source=rss-ff1d8e36f3a2------2</link>
            <guid isPermaLink="false">https://medium.com/p/a8ab108a1c8c</guid>
            <category><![CDATA[basketball]]></category>
            <category><![CDATA[golden-state-warriors]]></category>
            <category><![CDATA[nba]]></category>
            <category><![CDATA[stephen-curry]]></category>
            <category><![CDATA[lebron-james]]></category>
            <dc:creator><![CDATA[Lewis Pipkin]]></dc:creator>
            <pubDate>Wed, 05 Jul 2017 13:46:01 GMT</pubDate>
            <atom:updated>2017-07-05T13:46:01.029Z</atom:updated>
            <content:encoded><![CDATA[<h4>A look at the correlation between a team’s payroll and win percentage</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*vlfwxrqI5YLqVPLiX5LlHA.jpeg" /><figcaption>Ronald Martinez/Getty Images North America</figcaption></figure><p>In this league, you’d assume that the teams that go out and spend the most money in free agency or give their star players extensions are the ones that win the most. It makes sense, right? Teams like the Warriors and the Cavaliers and the Spurs must be paying tons of money to put such a great team out on the court year after year.</p><p>You may be surprised to find out that this year, the teams that won the most weren’t paying all that much.</p><p>As you may have heard by now, the Golden State Warriors were not spending a lot of money on their superstars- this Warriors team is the product of everything that could go right for a franchise going right. Steph Curry was the 4th-highest paid player on his team. Kevin Durant was the only player paid more than $20 million. And it’s not just the Warriors- the Rockets, Jazz, and Celtics were all under the salary cap. The Cavaliers and the Clippers were the only teams paying the luxury tax in 2016–2017 — that got them each 51 wins and, for the Cavaliers, an NBA finals berth.</p><p>In other words, shelling out a lot of money doesn’t always yield on-court success — and on-court success isn’t always dependent on your payroll. The correlation coefficient (<a href="http://www.investopedia.com/terms/c/correlationcoefficient.asp">a number between -1 and 1</a>) for a team’s payroll and win percentage is .33, a weak positive correlation. This is also shown in the regression of win percentage on a team’s payroll, where the coefficient on payroll is positive, but only statistically significant at the p=.1 level (t-value: 1.862/ p-value: .0732). Better than nothing, but not an excellent predictor.</p><p>Furthermore, when regressing win percentage on a dummy variable which takes a value of 1 when a given team is in the luxury tax and 0 if not, the dummy’s coefficient is not statistically significant at all (t-value: 1.325/ p-value: .196). Simply put, there is no noteworthy difference in winning percentage between teams who paid the tax this year and those who didn’t.</p><p>To answer the titular question, not really. While opening up the checkbook for role players like J.R. Smith and Tristan Thompson certainly helped the Cavaliers in the postseason, it was mostly having LeBron James (who, despite his big numbers, was definitely coasting during the regular season), Kyrie Irving and Kevin Love that got them to a 12–1 record before the Finals. Meanwhile, the Cavs’ number one adversary was somehow still paying the reigning two-time MVP $12.1 million in 2017 (<a href="https://94feetreport.com/steph-curry-signs-supermax-deal-with-golden-state-ac24de7aae02">that egregious underpay has since been rectified</a>).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*pK87PpC3fHl9KNm6JPNZLw.jpeg" /><figcaption>Ronald Martinez/Getty Images North America</figcaption></figure><p><em>All salary figures courtesy of Spotrac.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a8ab108a1c8c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Is it time for Memphis to move on from Chandler Parsons?]]></title>
            <link>https://medium.com/@pipkinlewis/is-it-time-for-memphis-to-move-on-from-chandler-parsons-76abf8eb51c1?source=rss-ff1d8e36f3a2------2</link>
            <guid isPermaLink="false">https://medium.com/p/76abf8eb51c1</guid>
            <category><![CDATA[memphis]]></category>
            <category><![CDATA[basketball]]></category>
            <category><![CDATA[nba]]></category>
            <category><![CDATA[memphis-grizzlies]]></category>
            <category><![CDATA[sports]]></category>
            <dc:creator><![CDATA[Lewis Pipkin]]></dc:creator>
            <pubDate>Sun, 02 Jul 2017 17:46:00 GMT</pubDate>
            <atom:updated>2017-07-02T17:46:00.794Z</atom:updated>
            <content:encoded><![CDATA[<h4>The Grizzlies paid a hefty price for the forward, what should they do with his contract from here?</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/0*fdF-Z_jfWzq9o9IT.jpg" /><figcaption>Rob Carr/Getty Images North America</figcaption></figure><p>“Say ‘I ruined the Grizz season, baby!’”</p><p>“…Say what?”</p><p>And then the <a href="https://twitter.com/JohnMartin929/status/840407596565700608">video</a> cut off. This is all we saw of Chandler Parsons’ interaction with this fan, an interaction which almost certainly ended very abruptly afterward. This video is, unfortunately, symbolic of the 2016–17 season for Memphis’ newest max contract player.</p><p>To say the least, Parsons’ first season with the Grizzlies was rough. Parsons played 34 games for the Grizzlies this year. He was, in a word, unimpressive. His 6.2 points per game on 34% shooting (26% from deep) didn’t help the team at all in the big picture. He’s never quite been accepted in Memphis; he and his knees have become the scapegoats for all of the fans’ strife. Before him, it was Jeff Green. Before him, it was Dave Joerger, or Lionel Hollins, or many others before them. Everything wrong with the Grizzlies can, apparently, be traced back to Parsons. And the sentiment does make sense, in a way. His money could be going to someone else.</p><p>But who else? The Grizzlies weren’t in the running for Kevin Durant. Al Horford wouldn’t fit into their scheme. The only free agent Memphis reasonably could have gotten that would have been a better fit on paper is Harrison Barnes. Parsons is the biggest free agent signing Memphis has ever had. Even better, he’s a shooter, exactly what Memphis has needed since the so-called death of Grit ’n’ Grind (note: they’re still grittin’ and grindin’, baby). And, despite his injury, Parsons would have commanded max money anywhere, especially Portland (who “hit the lottery” by not signing him, per C.J. McCollum). As many people seem to conveniently forget, Parsons averaged 14.8 points per game on 39.5% 3-point shooting during his two years in Dallas, with a career-high 41.4% coming in the 2015–16 season.</p><p>The Grizzlies needed to sign him to a huge contract to signal to future free agents that Memphis can be a place that free agents can come and win. Plus, with Parsons’ reputation as a recruiter, he may be who Memphis needs to bring in more personnel to take them to the next level.</p><p><a href="https://94feetreport.com/happy-nba-summer-everyone-c5887dd85bf5">Free Agency 2016 Roundup: The Best and Worst Contracts of 2016</a></p><p>To say that Parsons ruined the Grizzlies’ season, though, is flat-out stupid. Did Parsons make the Grizzlies lose two games each to the Lakers, Mavericks, Pelicans and Kings? Did Parsons curse his future team, years ago, to meet San Antonio in the playoffs year after year? Most importantly, did Parsons keep Troy Daniels from getting consistent minutes?</p><p>And really, all things considered, this season wasn’t a failure at all. This Grizzlies team faces the same problem as 28 other teams, which is, how on Earth are they supposed to compete with the Warriors? While the ultimate goal is to win a championship, in an era with a seemingly unstoppable team, having something the fans and the community can rally around is a good consolation prize. And, in a way, Parsons provided that for Memphis fans this season.</p><p>But, should the Grizzlies cut ties with Parsons, even at the risk of having to package him with draft picks to try to unload his contract? I think not.</p><p>Parsons was ruled out for the remainder of the season on March 13th due to a partial tear in his left meniscus. This was under a year after he was ruled out for the season with the Mavericks due to a full tear in his right meniscus. He’s had a myriad of other knee injuries in his short career. His knees, simply put, aren’t great. But, every injury he’s had, he’s recovered from nicely. This season is the only season he’s played under 60 games, and every year outside of this one and his rookie season, he’s averaged double digit points on above-league-average 3-point shooting. Chandler’s no bum.</p><p>But how will he fare in the future? There’s no way to tell for sure, but there’s a good way to take a look at it, drawing from statistician Nate Silver’s methodology. Silver used a method called PECOTA (<em>Player Empirical Comparison and Optimization Test Algorithm) </em>to find “doppelgangers” for baseball players in order to forecast their performance. Looking at how similar players fared every season can help to predict how a given player will perform. On Silver’s website FiveThirtyEight, there’s a basketball version as well, called<a href="https://projects.fivethirtyeight.com/carmelo/"> CARMELO</a>. CARMELO determined that Parsons’ closest player comparison is Jared Dudley. Not a bad comparison! I wouldn’t max Jared Dudley, but he’s a career 40% 3-point shooter. Dudley’s never had a bad knee injury like Parsons has, though.</p><p>So, let’s look at some players who share that with him. Russell Westbrook tore his right meniscus in the 2013 playoffs, and has since come back and averaged a triple-double. Westbrook is a bit smaller than Parsons, and plays a much different role, of course, but that’s encouraging. Doug McDermott tore his right meniscus his rookie season, and returned to average 9.4 points per game on 42.5% 3-point shooting. McBuckets is also much closer to Parsons’ size and play style than Westbrook is.</p><p>However, the best comparison I could come up with is Dorell Wright. Wright tore his left meniscus in 2007, and later went on to start all 82 games for the Warriors in 2011 (after a few more injuries), averaging 16.4 points per game on 37.6% 3-point shooting (with 2.4 makes per game). Danilo Gallinari has recovered from meniscus tears, as have Al Harrington, Josh McRoberts, and plenty of others. However, there are also plenty of players who can’t say the same thing.</p><p>The point is, there’s no surefire way to tell how good or bad Parsons will be in the years to come. His paycheck is, frankly, way too big, but if Memphis had let Parsons slip away, future free agents may have never even considered Memphis. If you keep up with his Instagram stories like I have, you know that almost every day he is putting in work to get back on the court. Memphis, I’m telling you, Chandler Parsons wants to get back on the court and show out for Grizzlies fans. He came to Memphis for a reason other than money.</p><p>There’s no way to be sure that Parsons will ever be the same player again. There’s no way to be sure that he will ever hit 10 threes in a half again like he did for the Rockets. As a fanbase, though, be sure to give him the support he needs — the kind of love that Memphis is known for giving its players — so that the gap between him and the fanbase is bridged.</p><p><em>All statistics courtesy of Basketball Reference</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=76abf8eb51c1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Free Agency 2016 Roundup: The Best and Worst Contracts of 2016]]></title>
            <link>https://medium.com/@pipkinlewis/happy-nba-summer-everyone-c5887dd85bf5?source=rss-ff1d8e36f3a2------2</link>
            <guid isPermaLink="false">https://medium.com/p/c5887dd85bf5</guid>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[nba]]></category>
            <category><![CDATA[basketball]]></category>
            <category><![CDATA[statistics]]></category>
            <dc:creator><![CDATA[Lewis Pipkin]]></dc:creator>
            <pubDate>Fri, 30 Jun 2017 01:54:41 GMT</pubDate>
            <atom:updated>2017-06-30T17:49:14.306Z</atom:updated>
            <content:encoded><![CDATA[<h3>2016 NBA Free Agency Review</h3><h4>With the 2017 free agency period underway, let’s take a closer look at the contracts signed last summer to see if players were overpaid or underpaid</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hGnaGQd1YC4mBbCS9KIimg.jpeg" /></figure><p>Happy NBA Summer, everyone!</p><p>July 1st is one of the most exciting times for NBA fans: the first day that free agents can verbally agree to contracts. A huge amount of rumors and agreements are going on right now, but what I want to do is take a look back at last year’s free agency and tell you, through a mix of personal opinion and data, the best and worst contracts from last year’s free agency.</p><p>The way I determined whether or not a player was overpaid is fairly simple. After playing with regression models based on player performance, I found one that I feel is a good mix of offense and defense. Among the variables that this model uses to predict a player’s salary are defensive rating, PER, true shooting%, games started, points per game, and age. I only considered a player’s performance for the team they signed their contract with in free agency; for example, Langston Galloway’s projection is for his time in New Orleans, not Sacramento. Just for ease, I removed players who played fewer than 30 games this season. For those interested, some more statistical information will be listed at the end of this article.</p><p>So, let’s get on with the fun!</p><h3><strong>Best contracts</strong></h3><h3><strong>LeBron James (3 years, $100m)</strong></h3><p>Actual salary: $30,963,450</p><p>Projected salary: $29,538,275</p><p>Overpay of $1,425,175</p><p>Let me make myself clear up front: despite this being a model that I chose, I am not going to agree with every one of its results. This is not an overpay; this is a steal of a contract for the best player in the league. If it weren’t for that pesky salary cap, you can be sure that LeBron would have been paid much, much more than $31m this season.</p><h3><strong>Kevin Durant (2 years, $54m)</strong></h3><p>Actual salary: $26,540,100</p><p>Projected salary: $25,580,599</p><p>Overpay of $959,501</p><p>One of the limits of statistics is that sometimes, the numbers can’t adequately capture a player’s impact on the court or what they mean to a team. Much like LeBron James, Kevin Durant is the kind of player that you pay as much as they want to be paid.</p><h3><strong>Dion Waiters (1 year, $2.9m)</strong></h3><p>Actual salary: $2,898,000</p><p>Projected salary: $15,039,691</p><p>Underpay of $12,141,691</p><p>Dion had an excellent season as Miami rallied from an 11–30 record to go 41–41 for the season. He struggled through some injuries (he only played 46 games due to a thigh injury and an ankle injury), but ultimately outplayed his small salary. He has opted out of his player option and will certainly be paid 8 figures.</p><h3><strong>Ty Lawson</strong></h3><p>Actual salary: $1,315,448</p><p>Projected salary: $7,555,098</p><p>Underpay of $6,239,650</p><p>It may surprise you to know that Ty Lawson is still in the league. This year was essentially Ty’s “prove-it” year: he was coming off a poor showing in Houston and a slightly better stint with the Pacers. He greatly over-performed his minimum salary, but his play is still a far cry from his early career in Denver. He was a passable backup/spot-starter for the Kings (although they did not exactly have realistic playoff aspirations), and should get a slightly larger paycheck this offseason.</p><h3><strong>Worst contracts</strong></h3><h3><strong>Jeff Green (1 year, $15m)</strong></h3><p>Actual salary: $15,000,000</p><p>Projected salary: $9,153,023</p><p>Overpay of $5,846,977</p><p>Jeff Green has been a polarizing player for years. He has a good amount of skill and athleticism, but he’s never been able to put it all together. Green was coming off portion of a season as a Clipper after being traded by the Grizzlies for Lance Stephenson. The Magic took on Green for a year, presumably on a “prove-your-worth” type of contract- but it’s hard to imagine that another team would have paid Uncle Jeff as much as Orlando did. Regardless, Green did not impress, posting absolute or near career lows in most shooting percentages and dropping below 10ppg scoring for the first time in his career.</p><p>This season was not all bad on the court for Green, though- he had a career-best 1.1 turnovers per game with a 20% usage rate, and shot 86.3% from the free throw line. I still believe Jeff Green has a place in this league, but not at that high of a price.</p><h3><strong>Bismack Biyombo (4 years, $72m)</strong></h3><p>Actual salary: $17,000,000</p><p>Projected salary: $15,430,036</p><p>Overpay of $1,569,964</p><p>With such a huge jump in the salary cap, we were bound to see some massive reactionary overpays. Biyombo had an excellent 2016 playoffs, especially for a player who does not typically score much. His 26-rebound performance against Cleveland in the Eastern Conference Finals certainly helped his stock as he entered free agency. Biyombo was paid $17m this season, and while his counting stats (6ppg, 8rpg, an assist and a block) don’t jump off the page, he was a fair defender. He had the 3rd best defensive rating on the Magic, although the Magic were a bad defensive team by just about every metric.</p><h3><strong>Solomon Hill (4 years, $48m)</strong></h3><p>Actual salary: $11,241,218</p><p>Predicted salary: $10,189,410</p><p>Overpay of $1,051,808</p><p>This looked like a bad signing when it happened. Hill had not proven in his time with the Pacers that he was worth $12m per year on average on a contract longer than he had even been in the league. Hill was the 4th highest paid player on the Pelicans this year (about a half-million less than Jrue Holiday), yet he shot poorly from all over the floor. Hill shot below league average from every shooting zone, including from 0–3 feet- effectively, dunks and layups. New Orleans’ front office took a big chance on Hill, and he was a royal letdown for the first season of his contract.</p><h3><strong>Festus Ezeli</strong></h3><p>Actual salary: $7,400,000</p><p>Projected salary: Negative</p><p>Overpay of a lot</p><p>My model gave Ezeli’s “performance” this season a negative salary, since Ezeli didn’t play a game for the Trail Blazers, and thus did not accumulate any statistics. Ezeli was out for the entire season with various left knee injuries, and his future is certainly not in Portland, per GM Neil Olshey. There’s not much else to say on Ezeli - he certainly has a place in the league if he’s healthy, but that is certainly not guaranteed at this point.</p><h3><strong>Chandler Parsons (4 years, $94m)</strong></h3><p>Actual salary: $22,116,750</p><p>Projected salary: $9,269,817</p><p>Overpay of $12,846,933</p><p>You knew this was coming. In fact, I’ve got an entire article devoted to Parsons’ situation in Memphis coming soon.</p><h3>More of the best</h3><p><strong>Nenê</strong></p><p>Actual salary: $2,898,000</p><p>Projected salary: $6,405,853</p><p>Underpay of $3,507,853</p><p><strong>Seth Curry</strong></p><p>Actual salary: $2,898,000</p><p>Projected salary: $7,805,293</p><p>Underpay of $4,907,293</p><p><strong>Matt Barnes</strong></p><p>Actual salary: $6,125,000</p><p>Projected salary: $12,824,264</p><p>Underpay of $6,699,264</p><p><strong>Zaza Pachulia</strong></p><p>Actual salary: $2,898,000</p><p>Projected salary: $10,292,705</p><p>Underpay of $7,394,705</p><h3>More of the worst</h3><p><strong>Allen Crabbe</strong></p><p>Actual salary: $18,500,000</p><p>Projected salary: $8,065,817</p><p>Overpay of $10,434,183</p><p><strong>Manu Ginobili</strong></p><p>Actual salary: $14,000,000</p><p>Projected salary: $5,688,163</p><p>Overpay of $8,311,837</p><p><strong>Timofey Mozgov</strong></p><p>Actual salary: $16,000,000</p><p>Projected salary: $11,380,490</p><p>Overpay of $4,619,510</p><p>Did I miss anyone? Do I have a contract in the wrong place? Is my take just plain wrong? Let me know!</p><p><a href="https://docs.google.com/document/d/1-khsy9cDOphjFN8lkuh-Rbpd1VZGtl7HVC8JmfLUmuY/edit?usp=sharing">Link to model summary: https://docs.google.com/document/d/1-khsy9cDOphjFN8lkuh-Rbpd1VZGtl7HVC8JmfLUmuY/edit?usp=sharing</a></p><p><em>All statistics courtesy of Basketball Reference</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c5887dd85bf5" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>