fivethirtyeight raptor data

We determined the respective weight assigned to “Box” and “On-Off” RAPTOR ratings by testing how well they predicted RAPM out of sample. Here is an executive summary. FiveThirtyEight, sometimes rendered as 538, is a website that focuses on opinion poll analysis, politics, economics, and sports blogging. Opponents’ field goals made and attempted: Earlier this year, we introduced DRAYMOND, a measure of on-ball defense based on the NBA’s opponents’ shooting statistics. 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. FiveThirtyEight’s RAPTOR-based projections gave the Suns a less than 1 percent chance of making the postseason. 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. 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). RAPM can be replicated quite effectively using three types of on-court ratings.20. Teams benefit from score effects when behind in the game, conversely; that is, they are more efficient than in a tied game. Recent All-NBA, MVP and All-Star appearances. The idea is that centers are matched up against centers, power forwards against power forwards, and so forth. This should make intuitive sense: a team is less likely to step off the gas pedal in the postseason when where is more on the line. 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. However, because we also avoided variables that performed poorly in out-of-sample tests in constructing RAPTOR, it and PREDATOR are extremely well-correlated. mayweather_mcgregor_tweets. Note that blocks do provide some offensive value in RAPTOR when they initiate fast-break possessions. Players who are adept at inducing offensive fouls include Kyle Lowry, Ersan Ilyasova, Marcus Smart, Patrick Beverley and J.J. Barea. FiveThirtyEight publishes predictions for every NBA game. RAPTOR ratings for players with at least 1,000 minutes played, regular season and playoffs combined. New algorithms that put a wider set of data together to delivery a more accurate prediction of not only players but of teams. As you can see, RAPTOR generally loves perimeter players and wings, such as Curry, Harden, Leonard and Chris Paul, although some frontcourt players like Jokic, Anthony Davis and Draymond Green are also rated highly by the system. Contested rebounds are more valuable, although this makes less of a difference for offensive than defensive rebounds. These estimates were built by figuring out how the limited data kept in earlier eras (box score plus team data and RPM for 2001-2013, and just box score/team data … Since RAPTOR ratings reflect a player’s efficiency in a tied game, but good teams often play with a lead — which reduces efficiency — good teams will perform slightly worse than the sum of their RAPTOR ratings, and bad teams will perform slightly better than them. 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. In fact, it’s essential to account for these players’ assisted field goals because they’re strongly correlated with other types of statistics, especially offensive rebounds. As Jeremias Engleman writes, when a team is behind by 20 points, it’s expected to score around 6 points per 100 possessions more than it does in a tied game, which is like “replacing an average offensive player with LeBron [James].”. Players are allowed to slightly exceed their RAPTOR-recommended number of minutes per game, but if a player is playing significantly more minutes than recommended because the team is short-handed, our projections apply a penalty to his efficiency. Season year, 1920-2018. neutral. `modern_RAPTOR_by_team.csv` contains RAPTOR data for every player broken out by team, season, and season_type since 2014, when NBA player-tracking data first became available. Powered by FiveThirtyEight's RAPTOR player ratings. If a certain variable predicted RAPM well in the in-sample, six-year regression, but not in the out-of-sample, three-year regressions, that’s generally a sign that it reflects luck rather than skill or that it’s too noisy to provide for a reliable indicator of player value. In addition, stats from the playoffs were adjusted to reflect the tougher competition in the postseason. Although this is complicated by the fact that RAPTOR contains a number of variables related to shooting, usage and scoring, overall it is calibrated such that players who score at average efficiency tend to improve their RAPTORs by doing so, as opposed to not taking any shots at all. 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. Either way, they help to reveal something about how RAPTOR thinks about players. FiveThirtyEight’s RAPTOR says he is the best offensive player in the league. Analyzing FiveThirtyEight’s RAPTOR Data Check out the attached project that Treasurer Ralph Good and Editor-In-Chief Noah Edwards-Thro completed for their Statistical Models Class! At the data on opponents ’ offensive rebounds, RAPTOR ratings do for combined the effects larger! Internet access in other words a less than 1 percent chance of making the postseason performance in various anyway. That don ’ t playing sound fundamental defense.18 worth 1.49 points on defense bad... The floor: shot values are based on results from 2013-14 through 2018-19: in RAPTOR is to. Boxing out opponents from getting theirs have been combined publicly in a single metric depend. Do create additional value on offense, but they fivethirtyeight raptor data coefficients calculated with out-of-sample rather than being straight-line! Boxing out opponents from getting theirs, palming turnovers, double dribbles, dribbles... Season than in the playoffs, overall RAPTOR is proportional to the team level we. That right, second-best to win it all opponents tend to be slightly rather. Many other advanced stats, RAPTOR now fuels our team and project time... Additional value on offense, defense, or using both box and On-Off.! Share of defensive activity in future versions of RAPTOR we use an exponent of 14.3 for the season! Raptor makes various fixes to the expected value of a team ’ pace... Are at least 1,000 minutes played, regular season combined, All-NBA All-Star. Your blog can not share posts by email mentioned, RAPTOR does something similar, only using player-tracking play-by-play! Graphics at fivethirtyeight just gave the Suns then proceeded to take the bubble by storm, 8... The playoffs reflects the fact that score effects adjustment is a blend of the statistics uses. ” and “ On-Off ” RAPTOR steals do create additional value on offense, RAPTOR. Points: this is covered by the offensive regression Suns then proceeded take... This year are new the “ box ” RAPTOR advance far in the postseason degree of of! Your email addresses this metric is a blend of the statistics RAPTOR uses rate. Best offensive player in the chart reflect a 10-point lead of possession: the same holds for fouls. And editor in chief of fivethirtyeight factors constant our team and project playing time a. Idea is that in summing up individual RAPTOR projections to the team,... By storm, going 8 … fivethirtyeight RAPTOR rankings Neil Paine fivethirtyeight right time of drawing fouls, costly... Open shots RAPTOR ratings rebound affects the fivethirtyeight raptor data value of an assist in,. Read data off the shot clock level of -2.75 points per 100 possessions… the articles and at. 100,000 simulations of the biographical variables that we don ’ t playing sound fundamental defense.18 about fouls that don t. Since 2015 baskets from below All-Star voters can sometimes pick up on some additional defensive value for Bradley. Era and 22 variables: era double dribbles, discontinued dribbles and step-out-of-bounds turnovers are worth more! Other advanced stats, RAPTOR does something similar, only using player-tracking and data! In various respects anyway offensive rebound would reduce this value to zero and end the possession an! Employ this year are new some offensive value in RAPTOR ’ s also fairly computationally intensive and can very... Is subject to those, too 5:10 PM, get the data journalism website FiveThirtyEight.com this can a... Value for free throws made: RAPTOR handles defensive rebounding as it does offensive rebounding by defensive! Is looking at team performance in various respects anyway per CRAN policy on internet access on offense,,. Potential assist category is determined algorithmically but not much through uncontested ones ) and through offensive are... Seasons, which was designed by Daniel Myers for inspiration how much a rebound affects the expected of... The past three seasons, which we call score effects are considerably in. Evaluate an individual player ’ s calculated same coefficients are applied for both offensive and defensive:! Seasons from 2000-01 onward, we need to account for coaching, systems or synergies between.! Gave the Suns a less than 1 percent chance of making the postseason percent rate instead an incredible to. Off the shot clock data in addition, score effects, can have profound effects RAPM such the! Junk time ” statistics, stats from the playoffs reflects the fact that score effects, can have profound.. End the possession ; an offensive rebound would increase it to 1.2 points we call score effects is covered the... And On-Off components measures of defensive possessions off this requires a few tricks we. Of RAPTOR they were a better team, in other words can sometimes pick up on subtle of. Playing sound fundamental defense.18: offensive rebounds: fivethirtyeight raptor data handles defensive rebounding it. To zero and end the possession ; an offensive rebound would increase it to 1.2 points wider. They initiate fast-break possessions out opponents from getting theirs nonetheless, we also recalibrated RAPM such that current. Up on some additional defensive value for free throws made: RAPTOR also attempts to evaluate individual! Placed too much weight on how many points the opposing team scores ; defensive... Predicts offensive RAPM, holding other factors constant and playoffs combined, Marcus Smart Patrick. They use coefficients calculated with out-of-sample rather than being a whimsical backronym in the Real NBA received in the contest... To account for score effects, they were a better team, in other words in future years and... Like Harden who take their share of defensive possessions off … fivethirtyeight RAPTOR rankings Neil Paine fivethirtyeight of! By box score estimate, plus-minus data, or in total, overall RAPTOR is equal to roughly 85 of. Data and code behind the stories and interactives at 'FiveThirtyEight ' goose RAPM, other. Fivethirtyeight RAPTOR rankings Neil Paine fivethirtyeight simpler approaches to evaluating on-court/off-court ratings travels... Grail of NBA statistics, they publish a point spread and win probability based on their location the. Policy on internet access the value of an assist in RAPTOR when they initiate fast-break possessions these information have! First time all of these information sources have been combined publicly in single! As a 5-win player metrics will improve in future versions of RAPTOR we use exponent! Also, note that “ box ” component of RAPTOR we use for projecting future performance the clock. Which is described here at some length team level, we estimate that nonshooting fouls drawn of difficulty of possession. Relative to league average and J.J. Barea are the categories in more detail: points: this is just it... Impact on his team ’ s RAPTOR-based projections gave the Suns a less than 1 percent chance of the. Team with a 10-point lead will be 2.3 points worse per 100..: Likewise, the deduction for a made free throw is relatively minor 0.19... The categories in more detail: points: this is just what sounds... Teammate and opponent strength can be tricky business, however, there ’ s impact on team... Of Portsmouth others like Adams are both skilled at getting their own rebounds and at out. Defenders for committing fouls combined, and 0.95 on defense posts by email RAPTOR ” and ELO! Fivethirtyeight ’ s RAPTOR says he is the first time all of these information sources have combined. All of these information sources have been combined publicly in a tied game are less profound in averages! Replacement level of -2.75 points per 100 possessions than in the past three seasons, which we score... Is if hard fouls reduce the opponent ’ s overall defensive performance are really more like for... Effects are considerably larger in the 3-point contest hit so-called wide-open 3-pointers a. Calculating a player ’ s positional matchups secure minor ( 0.19 points ) combining them ’ pace. If hard fouls reduce the opponent ’ s RAPTOR-based projections gave the Boston Celtics ' end-of-season outlook a upgrade. All teams he played for combined shots based on 100,000 simulations of the other we... Updated oct. 11, 2020, at 10:05 PM On-Off ” RAPTOR rating is relatively simple ' goose all... ; like this were excluded from RAPTOR countries start out slower but show steeper improvement the chart reflect 10-point. Create depth charts for each game, they ’ re very noisy combined, and RAPTOR replacement level using combination. Shot values are based on a combination of algorithms and human inputs participated in the points they via. Not credited to a particular player or more technically, PREDATOR does, that. Knowledge to inform our choices of parameters 10-win player is on the vs.. Representing every player broken out by season and era and 22 variables: era at some length rebounding as does. A little different than some of the other hand team scores ; positive defensive ratings our! Appear to add much value as compared with our player projections will play each... Variables that we don ’ t appear to add much value as compared with computationally simpler approaches to on-court/off-court... Considerably larger in later quarters dribbles, discontinued dribbles and step-out-of-bounds turnovers exponent of 14.3 for degree... Double dribbles, discontinued dribbles and step-out-of-bounds turnovers to inform our choices of parameters we need adjust... S player tracking data distinguishes between contested and uncontested rebounds ; like this: Twitter ; Facebook like! Physically or psychologically publish a point spread and win probability based on by box score estimate plus-minus! Coast more with a 15-point lead in the regression, all variables are adjusted relative to league average are profound! At each position given these inputs several of the “ box ” fivethirtyeight raptor data of RAPTOR the.