# Sports Reference Blog

## Combined Points Added to Team Game Finder

Posted by Mike Lynch on November 4, 2014

Thanks to a recommendation from user Jon P, we have added "combined points" as a search and filter option in the Team Game Finder.

You can now search for games in which the 2 teams combined for 100+ points or the highest scoring games since the merger. Or, if you can stomach it, you can even find games with twice as many turnovers as points.

## Introducing Box Plus/Minus (BPM)

Posted by David Corby on October 29, 2014

We're excited to introduce our newest statistical feature, Box Plus/Minus (BPM), developed and shared by Daniel Myers, well known as an administrator of the APBRmetrics forum and the developer of Advanced Statistical Plus/Minus (ASPM), which has performed very well empirically and to which BPM now serves as successor.

BPM is an advanced stat intended to measure a player's total contribution, calculated using box-score data that's available going back to the 1973-74 season. It follows on work done toward the Statistical Plus/Minus (SPM) concept, and ASPM, and uses long-term data sets developed by Jeremias Engelmann, the creator of ESPN's Real Plus/Minus statistic. Please visit Daniel's write-up to read more about the development of BPM, the background of the concept, and everyone that's contributed to the research.

BPM is presented intuitively, representing points per 100 possessions for which the player was on the court. For example, a player with a +4.3 BPM is said to have contributed 4.3 more points than an average player over 100 possessions, based on measurable statistical output from game box scores. The calculation makes heavy use of context dependent box score stats like USG%, TS%, STL% and others (as well as the statistical interactions between these components). We publish these as Advanced Box Score Stats and you can read about these component calculations in our glossary. Note that there is a separate calculation for the offensive component of a player's BPM, which yields both OBPM (Offensive Box Plus/Minus) and DBPM (Defensive Box Plus/Minus).

Further, BPM is scaled so that -2.0 represents a theoretical "replacement level" - thus, this concept is easily extended to permit calculations of one player's value over that theoretical threshold - that formula is [BPM - (-2.0)] * (% of minutes played), which is VORP, and interpreted as per 100 team possessions.

We'll be presenting BPM, it's offensive and defensive components, as well as VORP in the 'Advanced' table of our player pages, and likewise in our team pages:

For 2014, the leaderboards for BPM, OBPM, and DBPM, respectively, look like the following (> 1,000 MP):

And as the season gets underway, you'll be able to find BPM and its components, as well as VORP, on our actual leaderboard as well as other site features like the Play Index tools and season registers.

## Incompletions Added to Player Game/Season Finders

Posted by Mike Lynch on October 28, 2014

Thanks to a suggestion from user @AdamHarstad, we have added incompletions to the player season and game finders.

With this addition, you can search for things such as most incompletions in a season, game or career.

With Halloween right around the corner, we'll also point out that you can now query monstrosities such as THIS, as well.

If you clicked, we're very sorry you had to see that. Please use this tool responsibly!

## Stat Comparisons Added to Player Game Finder

Posted by Mike Lynch on October 27, 2014

As you may have noticed, we recently added the ability to compare different stat categories in our Team Game Finder. We're happy to announce that we've now brought the same function to our Player Game Finder.

So you can now search for things like:

Obviously, there's plenty of other queries you can run with this. We encourage you to play around with it and we hope you enjoy it!

Just as a reminder, if you do not want to answer survey questions on the Play Index, we offer ad-free subscriptions good on all Sports Reference sites for a little over a nickel per day (\$20/year).

## New Hall of Fame Prediction Model

Posted by David Corby on October 23, 2014

We maintained a Hall of Fame prediction model that, overall, works pretty well. This looks like a good list, yes?

However, keep scrolling and you'll see that a few players are slipping through the cracks. Miller is already in and good money is on both Mutombo's and Nash's eventual inductions.

So we've tweaked our model a little bit. It's no less simple than what we've been doing, but it casts a wider net around those factors that voters and the media prioritize for player inductions. Its criteria include All-Star games, leaderboard appearances, peak dominance, championships, and the player's height (yes, height!). Many of these variables are considered as well by our old model.

A necessary disclaimer is that Hall voters, to varying extents over the years, have used criteria that are difficult to account for in a simple predictive model such as college stardom, FIBA success, foreign careers, and so on. In a nutshell, we're considering plaudits earned while playing in the NBA so adjust as you wish those careers of players like Manu Ginobili and Pau Gasol who have racked up FIBA medals in addition to their pro career.

Anyway, here's the list of the top 25 retired and active players...

RETIRED:                                                             ACTIVE:

Except for Mutombo (6.4% to 86.8%) nothing changes too dramatically. Ben Wallace gets back in the race, and Nash becomes a virtual lock. Chauncey Billups and Shawn Marion move up significantly while Carmelo falls 10+ points.

To gauge your thinking about how many players "should" get in, here are the number of players in every NBA season between 1979-80 and 1999-00, inclusive, that were eventually enshrined:

We used the logistic regression method (same as before) and found that this works fairly well:

```my \$logit = -0.20303
+ -0.14203 * height_inches
+  0.80573  * championships
+  0.01594  * leaderboard_points
+  0.41568  * peak_ws
+  1.02443  * all_star_games
```

I'll get the documentation page up shortly which will explain a few of the nuances, for example how we determine "leaderboard points" and a few other particulars.

## More Game Play Finder Additions

Posted by Mike on October 23, 2014

We've added a text input to the top of the game play finder tool which works similar to our search autocomplete - start entering a player name and select the player you want from the dropdown and it will filter to show only plays in which that player was involved. For instance, if you want to see all of Peyton Manning's passing touchdowns or all of Devin Hester's touchdowns.

The play finder will now show defensive player stats as well, so you can search for something like all of Charles Woodson's interceptions returned for a touchdown and see which quarterbacks were the unlucky victims.

We've added selectors for QB kneels and sacks to the middle column as well, so you can get things like Russell Wilson's yards per rush with kneels excluded.

Finally, we've also added pass/rush distances and directions under at the bottom of the form - clicking that link will expand the form to allow you to pick rush direction or pass location. This will let you fine-tune your searches, so you can find how the Eagles performed on deep passes to DeSean Jackson in 2013 or who's most successful running up the middle in 2014. Note that these are reliant on the play-by-play data of the plays (and so their definition of direction and distance) and a "deep" pass is considered to be one that travels 15+ yards in the air.

## Per Possession Player Stats Added to College Basketball

Posted by David Corby on October 21, 2014

Just  a note that we added per possession statistics to the player pages on our college site. In particular, we express these as per 100 team possessions, which we estimate using component stats* and which we can calculate for every D-I school beginning with the 2010-11 season.

We recently added these to Basketball Reference so that users could easily compare output from different teams or eras, however the disparities in pace of play are even greater in college. Miami (Fl.) had the slowest pace in 2013-14 at 59.1 team possessions per game. On the other hand, Northwestern State and Virginia Military Institute led the NCAA at 78.9 and 77.5, respectively.

For now you can find these tables on the player pages:

And we'll be adding per possession statistics to the school pages, season registers and play-index tools soon enough.

*team_poss = 0.5 x ((fga + 0.475 x fta - orb + tov) + (opp_fga + 0.475 x opp_fta - opp_orb + opp_tov))

Posted in Announcement, Basketball-Reference.com, CBB at Sports Reference | Comments Off on Per Possession Player Stats Added to College Basketball

Posted by Mike on October 20, 2014

Since Peyton Manning has been in the news a bit today due to his new ownership of the all-time passing TD record, we've gotten a few questions on our Twitter account about progressive leaderboards:

We already had single season progressive leaderboards but today we've added career progressive leaderboards so that you can see who was the NFL's career passing touchdown leader in 1966? You can also see on that list where they rank as of the latest games, so #1 Y.A. Tittle in 1966 has the 19th-highest touchdown pass total in the current day.

You can get to a full list of the career or single-season progressive leaders by going to our main leaderboard index and selecting a stat and a year or from the box on the right of any of the career leaders lists.

## The Relationship Between SRS and NFL Ticket Prices

Posted by TiqIQ Staff on October 20, 2014

The home viewing experience for NFL games is the best it’s ever been. With that, the process of going to a game on Sunday can be a bigger hassle when factoring in all parts of the Fan Cost Index. Some struggles to bring fans to the games lead to things like a fantasy football section in the stadium or poolside cabanas in Jacksonville. On the secondary ticket market, there’s a few different factors that set the level of demand for NFL tickets. Market size will always be one of the biggest factors, as is quality of competition on a game by game basis. One thing the team itself can control to typically keep demand high is presenting a good on field product.

Using the past five seasons of TiqIQ secondary market data and the past five years of team SRS, we can see which teams have the strongest correlation of ticket prices to team success. After running the numbers, 12 teams came out with a correlation of .4 or higher. Another seven teams had a negative correlation of -.4 or below.

The strongest positive correlation came from the Denver Broncos at .99. The average price for Broncos tickets has risen every season since 2010, as has their SRS. In 2010 the Broncos had an SRS of -8.9, an average home price of \$120.66 and Kyle Orton at quarterback. This season, the Broncos have an average home price of \$339.71 as they are again one of the top teams in the NFL with Peyton Manning. The \$339.71 average price is the fifth highest in the league this season.

New Orleans, Washington and Seattle were the only three other teams to have a correlation above .7. Seahawks tickets follow a similar trend to the Broncos. In 2010 the Seahawks had an SRS of -9.4 and a season average on the secondary market below \$100. This season, the average price for a home game in Seattle is \$398.51, the most expensive price in the league.

Cowboys tickets had the fifth strongest correlation at .69. While the average price for a home game at AT&T Stadium is usually among the top half of the league, the actual average has risen and dipped with the team’s performance. Unsurprisingly, the \$260.23 average price at AT&T Stadium is the most expensive over the past five years.

The strongest negative correlation came from teams that have not been particularly good over the past five seasons — the Tennessee Titans, Oakland Raiders, Tampa Bay Buccaneers and Jacksonville Jaguars. The most interesting team in the bottom five is the Atlanta Falcons. The price of Falcons tickets on the secondary market have remained consistent — around a \$130 to \$150 average price through the years atop the NFC South and the struggles over the past two seasons.

Posted in Advanced Stats, Announcement, Pro-Football-Reference.com | Comments Off on The Relationship Between SRS and NFL Ticket Prices

## Stat Comparisons Added to Team Game Finder

Posted by Mike Lynch on October 15, 2014

We've recently made a huge enhancement to the search capabilities in the Team Game Finder. You can now search for games in which a team had more or less of one statistic than another. That probably sounds vague, so let's get into some concrete examples:

We hope you enjoy this new feature and please let us know if you discover anything cool with it.