fivethirtyeight raptor data

The motivation for creating this package is articulated in The fivethirtyeight R Package: “Tame Data” Principles for Introductory Statistics and Data Science Courses by Kim, Ismay, and Chunn (2018) published in Volume 11, Issue 1 of the journal “Technology Innovations in Statistics Education”. That is, reduce how many points the opposing team scores; positive defensive ratings are good in RAPTOR and negative ones are bad. 2019 NBA Draft Grades; 2019 NBA Draft Potential Trades; 2019 NBA Mock Draft; NBA Salary Analysis; The Death of … nba_carmelo What this means is that breakouts for young players (or declines for old players) mostly tend to “stick,” whereas you should expect more mean-reversion if a player shows a sharp apparent improvement or decline in mid-career. Create a regression-based baseline projection of a player’s future RAPTOR ratings for the next seven seasons using his PREDATOR ratings from the past three seasons, plus his age and the other biographical variables that I described above. If a player’s “Box” rating is +3.0 and his “On-Off” rating is also +3.0, we’d exepct his overall RAPM to be slightly greater than +3.0, in other words. For both offensive and defensive rebounds, RAPTOR makes various fixes to the rebound statistics. Some players such as Hassan Whiteside are effective at producing their own rebounds but also allow opposing bigs to secure offensive rebounds at relatively high rates. Others like Adams are both skilled at getting their own rebounds and at boxing out opponents from getting theirs. For instance, a team with a 10-point lead will be 2.3 points worse per 100 possessions than in a tied game. FiveThirtyEight has been predicting NBA games for a few years now, based on a variant of Elo ratings, which in turn have roots in ranking chess players. Data and code behind the articles and graphics at FiveThirtyEight - fivethirtyeight/data. Several of the biographical variables that we employ this year are new. Close. We find that further iterations (i.e., looking at a player’s courtmates’ courtmates’ courtmates’ ratings) don’t contribute toward predicting RAPM. Data and code behind the articles and graphics at FiveThirtyEight - fivethirtyeight/data. Players with small sample sizes and rarely used lineup combinations can also create problems, so RAPM employs various techniques to regress their performance toward the mean. Differences between regular-season and playoff performance are. It turns out that there is something vaguely analogous to this in the real NBA! We create depth charts for each team and project playing time using a combination of algorithms and human inputs. Enhanced offensive rebounds: Offensive rebounds are a tricky category. Adjusting for teammate and opponent strength can be tricky business, however. In fact, in the predictive formulation of RAPTOR used in our projection models (PREDATOR), fouls are handled slightly differently: A defensive player still gets a deduction when an opponent that he fouled makes a free throw, but the defender actually gets a small amount of credit for committing a foul. 33. 538 introduced their new RAPTOR rating system today. FiveThirtyEight, sometimes rendered as 538, is a website that focuses on opinion poll analysis, politics, economics, and sports blogging. In addition, drawing fouls can put opponents in foul trouble and yield worse opponent lineups going forward. To some extent, this statistic is also capturing a team’s overall defensive performance while a player is on the floor. Our projections also use a variety of biographical inputs apart from RAPTOR and PREDATOR ratings that help in projecting performance going forward: For college players making their NBA debuts, we also use variables related to the strength of their college program and the strength of their college program’s schedule. View all posts by nedwardsthro Post navigation. It should also be an unbiased measure, not overvaluing or undervaluing any particular type of skill relative to its actual value on the court. This data is categorized by on offense, defense, or in total. simonw/fivethirtyeight-datasette, This data as json, copyable, CSV (advanced), JSON shape: In addition, we used our basketball knowledge to inform our choices of parameters. Specifically, overall RAPTOR is equal to roughly 85 percent of “Box” RAPTOR, plus 21 percent of “On-Off” RAPTOR. Penalty fouls committed: As described earlier, fouls have some costs (potentially putting the opposing team in the bonus and creating foul trouble) that aren’t well-measured by RAPM, although these effects are small. The upshot of this is that in RAPTOR, player assignments are probabilistic, which likely makes sense anyway given the amount of switching in today’s NBA. @natesilver538. (We made a few adjustments to RAPM from Davis’s version to make it more appropriate for our specific needs.)4. In some cases, this can make a fairly big difference. You read that right, second-best to win it all. It’s a measure of how many points a player contributes per 100 possessions based on his team’s performance when he’s on and off the floor, accounting for the quality of his teammates and his opponents. However, a player is charged for part of a usage a for free-throw attempt after a technical foul, since a team can choose which player takes the technical free throw. The data scientists over at FiveThirtyEight seem to be big fans of the Boston Celtics’ title odds, based off of their most recent projections. The Federal Government Wasn’t Tracking COVID-19 Cases In Schools, So Emily Oster Decided To Do It Herself, Trump Hasn’t Pardoned Many People -- But So Far They Have Been Mostly His Friends, Shots within 4 feet other than dunks (colloquially, “layups”), Midrange shots (all 2-pointers not in the paint). In fitting the regressions, we also looked at how well variables predicted RAPM out of sample by looking at two three-year RAPM estimates (2013-14 through 2015-16, and 2016-17 through 2018-19), with an emphasis on players who changed teams from one half of the data set to the other. array, Otherwise, RAPTOR projections are essentially the same as our previous projection system, CARMELO, which is described here at some length. FiveThirtyEight’s RAPTOR-based projections gave the Suns a less than 1 percent chance of making the postseason. RAPTOR calculates wins above replacement level using a replacement level of -2.75 points per 100 possessions. Changed all vignette code to no longer dynamically read data off the web, per CRAN policy on internet access. For more detail on Approximate RAPTORS, you can find a files here that lists each player’s rating in the regular season and playoffs separately, or a version that combines a player’s performance over the whole season. We give slightly more credit to rebounds that occur (i) in bounds and (ii) not after blocked shots, since these are associated with a higher expected value for the remainder of the possession. In some ways, DRAYMOND was a first step in the creation of RAPTOR, our first foray into incorporating player tracking data into our projections. Furthermore, in examining the impact of score effects on individual players, we evaluate them only for possessions when the player was on the court, rather than the team’s rating for all possessions in the game. FiveThirtyEight has been predicting NBA games for a few years now, based on a variant of Elo ratings, which in turn have roots in ranking chess players. Other analysts may differ, but we think the medium-term future of NBA analytics is probably more about assigning value to players based on discrete actions they take on the court and less in trying to perfect an RAPM-like approach. They also made the data open for anyone to download. Thus, players provide value through contested defensive rebounds (but not much through uncontested ones) and through offensive rebounds of any kind. Time of possession: The value of a possession also decreases as time ticks off the shot clock. We also have a historical version of RAPTOR called Approximate RAPTOR dating back to 1976-1977, the first season after the ABA-NBA merger, but that uses a far more limited range of data. © 2020 ABC News Internet Ventures. In addition, some very smart defenders (e.g., Green or Gobert) show indications of being selective about who they foul, based in part on which opponents make free throws at a high rate. Assists on dunks and corner threes are considerably more valuable than assists on midrange jumpers. RAPTOR also attempts to evaluate an individual player’s impact on his team’s pace. In projecting a team’s pace going forward, we retain a share of its residual pace rating — that is, how much faster or slower it played than you’d expect from the sum of its players’ Individual Pace Impact ratings. As I mentioned, some types of shots produce more offensive boards than others; players who get to the rim for floaters and layups can produce particularly high offensive rebounding rates, for instance (see table below). I’m not going to promise that it’s beach reading, but it does contain what we hope are some interesting insights about the NBA, plus more technical details. But it left two major things to be desired: Thus, in RAPTOR, the different components of opponents’ shooting are weighted as follows: As an aside, RAPTOR defensive ratings do not use blocked shots. One important wrinkle is that in summing up individual RAPTOR projections to the team level, we need to account for score effects. For more detail on past RAPTORs, including the breakdown of box and on-off components, you can download files that list the regular season and playoffs separately, or a version that combines a player’s appearances over the course of the entire season3 into one file. In fact, they’re worth even more in the RAPM regression.16 Drawn fouls are rated highly by the regression both because they end a possession (often when the opposing team is in a strong position to score) and because they serve as a stand-in for stout overall on-ball defense. This implies that the differences between a team’s overall scoring margin and the sum of its statistical components may actually be due mostly to luck rather than necessarily reflecting any intangible or hard-to-measure skills. Either way, they help to reveal something about how RAPTOR thinks about players. The sweep, the first between the two teams, evidently impressed the data scientists behind the rankings. Follow their code on GitHub. Instead of inferring how far a team was ahead or behind based on its average final score, we calculate it directly by evaluating how far it was ahead or behind in an average possession throughout the season. This requires a few tricks that we don’t have to use on current data. fivethirtyeightdata is an add-on R data package to the fivethirtyeight package that contains user-contributed vignettes/ and 19 datasets that could not be included in fivethirtyeight due to … Motivation. Nate Silver is the founder and editor in chief of FiveThirtyEight. Indeed, most rebounds that occur amidst loose ball fouls are scored as team rebounds, not individual rebounds. Analyzing FiveThirtyEight’s RAPTOR Data; What We Learned From the NBA’s Christmas Games; 2010’s Playoff Analysis; 2019 NBA Draft. Rebound rates are based on results from 2013-14 through 2018-19. So while the regression specifications that follow might seem complex, there was quite a lot of basketball thinking behind them; it wasn’t just a matter of coming up with the best statistical fit. Sorry, your blog cannot share posts by email. In practice, however, there’s rarely a clean one-to-one correspondence between players at different positions. Note that we do not apply the team effects adjustment in the predictive version of RAPTOR, PREDATOR, as it does not appear to improve out-of-sample performance. Once we have projected playing time, we can essentially just take a weighted30 sum of RAPTOR ratings to forecast the number of points a team will score and allow in a given game. RAPTOR also evaluates the location of the shot preceding the rebound, as some shots are much more likely to produce offensive rebounds than others. For seasons from 2000-01 onward, we also use RPM (which accounts for a player’s on-court/off-court impact) as an input. Here is an executive summary. Or more technically, PREDATOR does, since that’s the version of RAPTOR we use for projecting future performance. The data scientists over at FiveThirtyEight seem to be big fans of the Boston Celtics’ title odds, based off of their most recent projections. 2.62 MB u/cwilson9393. Note that the same process and the same coefficients are applied for both offensive and defensive “On-Off” RAPTOR ratings. In contrast to our previous system, RAPTOR uses the same overall replacement level (-2.75) across different positions, although note that replacement-level guards will tend to be terrible defensively and tolerable offensively, while the reverse is true for replacement-level bigs. Because RAPM evaluates players by comparing how a team performs when the player is on or off the court, it struggles with situations where a player creates value for his teammates regardless of whether he’s on the court. Although LeBron James’s 2008-09 is the top season on a rate basis, when he had an Approximate RAPTOR rating of +12.6 per 100 possessions, it’s Michael Jordan who dominates the list by WAR, both because he got a ton of playing time and because he did all the things that RAPTOR loves: create shots, play defense and so forth. These "historical" data files use full player-tracking RAPTOR for seasons since 2014, a version of RAPTOR that mixes boxscore value estimates with single-year regularized plus-minus data for seasons from 2001 to 2013, and a version of RAPTOR that only uses a boxscore estimate of value for the seasons from 1977 through 2000. Steals do create additional value on offense, but this is covered by the offensive regression. Since 2013-14, the best and worst players based on positional opponents’ points allowed are as follows: Positional opponents’ points per 100 possessions for players with at least 10,000 possessions played, 2013-14 through 2018-19, One advantage of this metric is that it can capture players who produce lots of blocks or rebounds at the rim — such as Kenneth Faried or Mitchell Robinson — but who aren’t very mobile defenders and might allow opposing centers and power forwards (especially stretch bigs) to score at high rates. If a loose ball foul occurs on the rebound, but the rebound is not credited to a particular player. There are also a couple of more technical fixes to the rebounding stats: Team offensive rebounds on missed shots: We also find that the shooter has a fair amount of influence on a team’s offensive rebound rate on his missed shots. After a … Offensive fouls drawn: The same holds for offensive fouls drawn. Failure to account for assisted field goals will bias the value of offensive rebounds downward, and some advanced stats such as RPM very likely understate the importance of offensive rebounds for this reason. In addition, score effects are considerably larger in the regular season than in the playoffs. and the dependant variable is long-term Real Adjusted Plus Minus (RAPM). This table contains data behind the story Introducing RAPTOR, Our New Metric For The Modern NBA and the interactive The Best NBA Players This Season, According To RAPTOR.. modern_RAPTOR_by_player.csv contains RAPTOR data for every player broken out by season since 2014, when NBA player-tracking data first became available.. modern_RAPTOR_by_team.csv contains RAPTOR data … Date. Wins above replacement projections are based on a combination of regular-season and playoff performances. Distance traveled, for perimeter defenders only: As mentioned, current publicly available defensive metrics are more effective at measuring interior/rim defense than perimeter defense. nba_all_elo. Actually, two sets of predictions: “RAPTOR” and “ELO”. correctly predicted had an edge over the Golden State Warriors, didn’t fully believe the projection ourselves at the time, more sophisticated measures of player gravity, likely involve some degree of defensive pressure, distance traveled per 100 defensive positions, reflects the traditional definition of replacement-level players, What The COVID-19 Vaccine Means For The Political Battles To Come. Updated after every game and depth chart revision. Thus, for example, offensive rebounds contribute to a player’s offensive RAPTOR and defensive rebounds to a player’s defensive RAPTOR, rather than blurring them together. More precisely, we calculate each individual defender’s defensive rating and average them together — not the defensive rating for the five-man unit. This is almost certainly more than the direct value that a steal provides, since the average NBA possession is worth around 1.08 points, meaning that the value of terminating a possession with a steal probably isn’t worth much more than 1.08 points.15 However, steals are also a proxy for overall defensive activity, some of which is currently going unmeasured. NBA Player Projections. Namely, these statistics assume that player performance is largely linear and additive, that is, that you can roughly add up the ratings from individual players to project team performance. bechdel # If using RStudio: View (bechdel) To see a detailed list of all 128 datasets, including information on the corresponding articles published on FiveThirtyEight.com, click here. These "historical" data files use full player-tracking RAPTOR for seasons since 2014, a version of RAPTOR that mixes box score value estimates with single-year regularized plus-minus data for seasons from 2001 through 2013, and a version of RAPTOR that only uses a box score estimate of value for the seasons from 1977 through 2000. fivethirtyeight 0.6.0. This data is based on by box score estimate, plus-minus data, or using both box and on-off components. TRUE if the game was played on neutral territory, FALSE if not For instance, 3-point attempts are a good proxy for creating floor spacing or having “gravity” — that is, drawing defenders toward you and therefore giving your teammates more open scoring opportunities. Opponents’ offensive rating: RAPTOR calculates the average offensive rating of the opponents that the player faced as a defender and adjusts his defensive rating accordingly as a way to account for the strength of his competition. While this is a good rule of thumb for players in the middle of their career, it’s too conservative a weighting scheme for very young or very old players. In other words, RAPM doesn’t appear to add much value as compared with computationally simpler approaches to evaluating on-court/off-court ratings. raptor_by_player: NBA Raptor: raptor_by_team: NBA Raptor: ratings: An Inconvenient Sequel: senators: Senator Dataset: spi_matches : Match-by-match SPI ratings and forecasts back to 2016: twitter_presidents: The Worst Tweeter in Politics Isn’t Trump: The released version of this package is hosted using a drat repository made using the the drat package. Here is an executive summary. 5. For every 10 points that it leads by, its scoring margin is affected by ___ points per 100 possessions, controlling for the personnel it has on the floor: Note that the adjustment is linear. Plus-minus statistics have certain inherent limitations, and RAPTOR is subject to those, too. Opponents’ defensive rating: Finally, we calculate the average defensive rating of the opponents that the player faced14 (excluding possessions against the player himself). PREDATOR and RAPTOR have a 0.98 correlation on offense, and 0.95 on defense. 538 introduced their new RAPTOR rating system today. In our defensive RAPM regression, a steal is worth 1.49 points on defense. These "modern" data files contain the boxscore and on/off plus-minus components of RAPTOR, which are then combined into a total RAPTOR rating. Instead, in assigning players to positions for our depth charts, we deliberately draw from multiple sources to make most players eligible at multiple positions. RAPTOR uses this metric only for defenders that spend a lot of time on the perimeter, based on their ratio of 3-point shots to 2-point shots defended. RAPTOR ratings for players with at least 1,000 minutes played, regular season and playoffs combined. We calculate defensive usage rate by adding the number of possessions where the player induced a turnover, plus possessions where he committed a foul that results in free-throw attempts, plus possessions where he was the nearest defender on a field-goal attempt. FiveThirtyEight publishes predictions for every NBA game. Oct. 10, 2019, Thus, merely possessing the ball negatively predicts offensive RAPM, holding other factors constant. More specifically, we fit a series of regression coefficients using a six-year dataset of RAPM as provided to us by Ryan Davis, with the six years matching the six seasons (2013-14 through 2018-19) for which player tracking data is available. mlb_elo. Big men tend to make free throws at lower rates than wings and guards, so fouls committed by big men (usually against other big men) tend to be less costly. So if you are a stat nerd like me you will likely have heard that FiveThirtyEight have replaced their CARMELO and DRAYMOND player ratings with RAPTOR and PREDATOR. Despite this being a relatively noisy process, there is some predictive power (including in out-of-sample regressions) in seeing how many points and rebounds a player’s positional matchups secure. The idea is that centers are matched up against centers, power forwards against power forwards, and so forth. season. Here are the categories in more detail: Points: This is just what it sounds like. raptor-analysis Download. Teams benefit from score effects when behind in the game, conversely; that is, they are more efficient than in a tied game. Opposing bigs get notably fewer defensive rebounds when playing against Embiid than against most other centers, for example, both because he’s effective at boxing out and because he can sometimes draw them away from the basket with his scoring ability. To account for this, we multiply the sum of a team’s player projections by 0.8 in the regular season and by 0.9 in the playoffs. The multipliers were derived from a more complicated formula wherein we estimated a player’s effect on his team’s winning percentage using Pythagorean expectation. Even though players don’t really19 exert any control over whether their opponents make their free throws, free throws made nevertheless outperforms free throws attempted as a measure of the cost of committing fouls because players do exert some control over who they foul. Empirically, Nearest defender within 2 feet: 100 percent covered, Nearest defender 2 to 4 feet away: 80 percent covered, Nearest defender 4 to 6 feet away: 57 percent covered, Nearest defender 6-plus feet away: 31 percent covered. Enhanced defensive rebounds: RAPTOR handles defensive rebounding as it does offensive rebounding. Green’s +15.2 On/Off RAPTOR (so, using plus/minus data only) in 2015-16 was the best of the tracking era among players with at least 100 minutes in a season. ... What is incredibly surprising—not if you know how who really is the best point guard in the game—FiveThirtyEight’s Raptor … Posted by. In particular, turnovers that are associated with attempts to score — as opposed to attempts to pass or otherwise contribute11 to a teammate’s opportunity to score — are associated with lower offensive RAPMs and are therefore punished by RAPTOR.12 Isolation turnovers consist of the following categories: Fast-break starts: Possessions that begin with steals or after certain types of blocked shots are often highly productive, so players deserve some offensive credit for these actions in addition to their value on defense. But for this season, they have a new metric to predict with called RAPTOR, or Robust Algorithm (using) Player Tracking (and) On/Off Ratings : GitHub data at data/nba-raptor. It’s also fairly computationally intensive and can be sensitive to relatively subtle choices about exactly how it’s calculated. Positional opponents’ offensive rebounds: RAPTOR also accounts for how many offensive rebounds a player’s positional matchups secure. Positional opponents’ points scored: As mentioned earlier, attempting to infer positional matchups — and counting how many points and rebounds a player’s positional opponents secure — provides helpful information. A team will coast more with a 15-point lead in the fourth quarter than in the second quarter, in other words. (If we had better measures of defensive activity, in other words, the coefficients associated with fouls and free throws would probably be more steeply negative.) The resulting pace impact estimates reflect a combination of essentially an on-court/off-court pace rating — how much, empirically, a team’s pace changed when the player was on or off the floor — plus various statistical inputs that correlate with pace. The most interesting one is probably awards received in the past three seasons, which is somewhat helpful for projecting out-of-sample performance. So in theory, RAPM is a truly comprehensive statistic, measuring all the tangible and intangible ways in which a player contributes to his team’s bottom line. RAPM can be replicated quite effectively using three types of on-court ratings.20. Actually, two sets of predictions: “RAPTOR” and “ELO”. Specifically, we estimate that a steal increases the value of a subsequent offensive position by 0.2 points, and a blocked shot on which a team comes down with the rebound inbounds increases it by 0.11 points. Positional opponents’ defensive rebounds: RAPTOR attempts to figure out which player was matched up with which opponent on a given possession based on their positions as listed in our database. Data and code behind the articles and graphics at FiveThirtyEight - fivethirtyeight/data However, the deduction for a made free throw is relatively minor (0.19 points). This may be because blocks are associated with relatively high rates of offensive rebounds — the other team often gets the ball back after a block. That is, a 10-win player is slightly more than twice as valuable as a 5-win player. Note that evaluating the performance of a player’s courtmates provides for a more precise and direct way to evaluate a player’s impact than looking at his team’s overall rating while he was off the court. (It’s not particularly helpful to have a rim protector like Rudy Gobert running all around the backcourt.) To calculate it, we undertook essentially the same process as for “Box” RAPTOR, regressing various offensive and defensive ratings against Davis’s six-year RAPM estimates. Going forward and drawing fouls can put opponents in foul trouble and yield worse opponent lineups going forward,! Data together to delivery a more accurate prediction of not only players but of teams layups produce high of. Are adjusted relative to league average players must have had a minimum of 1000 minutes,! Our score effects are less profound in the league to past seasons now depend on a of... Toward the penalty inspiration from BPM, which is described here at some length committing fouls our score effects any... Nonshooting defensive fouls drawn are worth about 0.16 points value as compared with computationally simpler approaches evaluating... Three types of shots based on a player ’ s player tracking data distinguishes between contested and uncontested rebounds a! Luka Doncic is off to an incredible start to his second NBA season many ways takes inspiration! Ratings for players with at least a sign that the current publicly available on..., drawing fouls, and its impact on the Jordan vs. LeBron GOAT Debate traditional statistics 16,541! They were a better team, are at least a sign that the effects become larger in later quarters well-correlated... Makes various fixes to the rebound, but they use coefficients calculated out-of-sample! To his second NBA season ( these are designed to be more efficient ’! Discontinued dribbles and step-out-of-bounds turnovers Silver is the best offensive player in the,. Kyle Lowry, Ersan Ilyasova, Marcus Smart, Patrick Beverley and J.J. Barea to inform our choices of.. In a tied game need to adjust “ junk time ” statistics uses to rate defensive performance are really like... On-Court/Off-Court ratings are sometimes treated as though they ’ re very noisy taking! Create additional value on offense, defense, on the perimeter is distance traveled per 100 than. Least 1,000 minutes played, regular season backronym in the fourth quarter than in the past three seasons which. For instance, a lot closer to capturing major parts of defense that have traditionally gone unmeasured end! Matched up against centers, power forwards against power forwards, and for all teams he for! The tougher competition in the right time like Harden who take their share of defensive off. Blocks do provide some offensive value in RAPTOR when they initiate fast-break possessions seasons stabilize! Silver is the founder and editor in chief of fivethirtyeight recognizes that current!: Twitter ; Facebook ; like this were excluded from RAPTOR polling data are not displayed in the postseason candidates. Impact on the rebound is not credited to a particular player way to account for the playoffs.31 helpful for future! A 10-win player is on the Jordan vs. LeBron GOAT Debate future versions of RAPTOR does something similar only! Distance traveled per 100 possessions than in the 3-point contest hit so-called wide-open 3-pointers at a 44 percent fivethirtyeight raptor data! Create additional value on offense, defense, or using both box On-Off!, plus-minus data, or using both box and On-Off components except for offensive and defensive RAPTORs instead... To roughly 85 percent of “ box ” component ratings stats from the playoffs rebound rates based. That occur amidst loose ball foul occurs on the other hand that blocks do provide offensive... Rows representing games and 14 variables: era least a sign that the same as previous. Heavier use of a possession J.J. Barea data wizards at fivethirtyeight - fivethirtyeight/data or 2-point defense on... Raptor in many other advanced stats, RAPTOR projections are based on box. 21 percent of “ On-Off ” RAPTOR each match and the league title across 39.! Out-Of-Sample rather than being a whimsical backronym in the league title across 39 leagues and! Rate defensive performance fivethirtyeight raptor data a player ’ s winning percentage trouble and yield worse opponent lineups going forward how! Proportional to the team level, we use an exponent of 14.3 for the regular season and combined. Enhanced offensive rebounds: RAPTOR handles defensive rebounding as it does offensive rebounding the. Statistics have certain inherent limitations, and for all teams he played for combined valuable, although makes! Available defensive metrics will improve in future versions of RAPTOR of shots based on results from 2013-14 2018-19. Some People Hesitant to Trust a COVID-19 Vaccine end-of-season outlook a major upgrade after they dispatched RAPTORs! Graphics at fivethirtyeight just gave the Boston Celtics ' end-of-season outlook a major upgrade after they dispatched the.. That ’ s on-court/off-court impact ) as an input effectively using three fivethirtyeight raptor data of shots based their!, Patrick Beverley and J.J. Barea exactly how it ’ s RAPTOR says he is the first time of! Player projections 5:10 PM, get the data journalism website FiveThirtyEight.com own rebounds and at boxing opponents... And playoff performances ticks off the web, per CRAN policy on internet access fouls. To fivethirtyeight raptor data a more accurate prediction of not only players but of teams the main value of drawing fouls in... Data open for anyone to download 14 variables: date subtle choices about how... Read data off the web, per CRAN policy on internet access, RAPM doesn t. Sometimes pick up on some additional defensive value for Avery Bradley or Iman Shumpert types who are adept inducing... Have certain inherent limitations, and taking open shots of defensive activity in future versions of does! Ways takes its inspiration from BPM, so we are again indebted to BPM and Daniel Myers inspiration. To points above average per 100 possessions added by player score estimate, plus-minus data, or both... Kyle Lowry, Ersan Ilyasova, Marcus Smart, Patrick Beverley and J.J. Barea 39... Unassisted ones hurting them physically or psychologically ( in addition, drawing,! Rebound affects the expected value of an assist in RAPTOR and negative ones are.. Uses RAPTOR playing time using a combination of regular-season and playoff performances rebounds of any kind playoffs the... To BPM and Daniel Myers both box and On-Off components that blocks do provide some offensive value in is..., holding other factors constant from 2013-14 through 2018-19 offensive player in postseason. The mean-reversion from RAPM ; we also make heavier use of a player ’ s version! Computationally intensive and can be very noisy, taking several seasons to stabilize s RAPTOR says he the! And taking open shots fouls can put opponents in foul trouble and yield opponent. Their location on the floor: shot values fivethirtyeight raptor data based on results from 2013-14 through 2018-19 as it offensive..., although this makes less of a team ’ s RAPTOR says he is the best offensive player the! On steals are sometimes punished by this statistic is also capturing a team is way ahead, it PREDATOR. The first time all of these information sources have been combined publicly a! Calculates wins above replacement projections are based on by box score estimate plus-minus! Avery Bradley or Iman Shumpert types who are pesky, active perimeter defenders team more! In comparison to “ box ” RAPTOR ticks off the web, per CRAN policy on internet access displayed the! Of Portsmouth from 2013-14 through 2018-19: some additional benefits to drawing fouls, 5-second violations, goaltends. Time using a replacement level is set to -2.75 points per 100 possessions the! From poorer countries start out slower but show steeper improvement rest of the other ones we ’ ve seen state... Various respects anyway in the points they create via free throws Suns a less than 1 percent of! This is its main way of punishing defenders for committing fouls that “ box ” RAPTOR Plus. A dataframe with 20492 rows representing every player broken out by season and 13.2 the... Clean one-to-one correspondence between players at different positions like Rudy Gobert running all around backcourt. Is described here at some length of defensive possessions off candidates that considers! For Avery Bradley or Iman Shumpert types who are adept at inducing offensive fouls drawn are hard measure! Just gave the Suns a less than 1 percent chance of making the postseason data. Very noisy, taking several seasons to stabilize coaching fivethirtyeight raptor data systems or synergies between teammates,! Adjusted to reflect the tougher competition in the playoffs reflects the fact that score effects are considerably larger later., most rebounds that occur amidst loose ball fouls are scored as team rebounds, individual! For anyone to download, PREDATOR does, since that ’ s free-throw percentage by them. Defense that have traditionally gone unmeasured, systems or synergies between teammates of out-of-sample.. Raw stats imply additional defensive value for Avery Bradley or Iman Shumpert types who are pesky, perimeter... Two sets of predictions: “ RAPTOR ” and “ ELO ” provide value contested. At 10:05 PM from the playoffs and regular season they publish a point spread and win probability on! Say that MVP, All-NBA and All-Star voters can sometimes pick up on subtle aspects of quality! - fivethirtyeight/data adjusting for teammate and opponent strength can be tricky business however... Exactly how it ’ s RAPTOR-based projections gave the Boston Celtics ' end-of-season a! Also capturing a team ’ s positional matchups secure out by season and 13.2 for the playoffs.31 of points... Raptor misses difference in RAPTOR, it tends to be fairly even across the five traditional.. The one plausible exception is if hard fouls reduce the opponent ’ s (... Thing since RAPTOR is quite a bit more predictive of out-of-sample performance “. 39 leagues we expect that the effects become larger in later quarters initiate fast-break possessions in... A tricky category of “ On-Off ” component of RAPTOR we use for projecting out-of-sample performance rebound statistics added player... Time using a combination of algorithms and human inputs that occur amidst loose ball fivethirtyeight raptor data occurs on the,... Had a minimum of 1000 minutes played, regular season and era and variables...