‌​

Frequently Asked Questions

Accounts

What are the different types of subscriptions you offer, and how much do they cost?

A Season Pass provides non-expiring website access to a single season's worth of content. (For example, 2018 WNBA or 2019 Men's NCAA Basketball.) The price for this package is typically $49.99 for the first half of the season. The price drops to $24.99 after the midway point of the season, and access is available for $14.99 after the season concludes.

A League Pass is lifetime website access for all seasons (past, present and future) to any of our 16 sites, as well as 1,000 requests per hour for that sport's content to the National Statistical API. (For example, Kontinental Hockey League or Women's NCAA Basketball.) It's a one-time purchase of $249.99.

A Full Pass is lifetime website access to all 16 of our sites, plus 5,000 requests per hour to the National Statistical API. The regular price for this package is $1,199.99. A Monthly Full Pass is also available for $24.99 per month (with 2,500 API calls per hour), and you can cancel anytime with one click.

There is an API Plus package for high-volume data users, which includes 100,000 calls per hour. A lifetime pass for the API costs $2,699.99. We offer a Monthly API Plus package for $49.99 per month.

You can sign up for any of our packages at our subscription page.

What is the difference between a Monthly Full Pass and a Full Pass?

A Full Pass is a one-time purchase that offers lifetime access to all of our past, present and future website content, including API access. A Monthly Full Pass is a month-to-month version of the Full Pass (with a lower volume of hourly API calls) at a discount rate that can be cancelled at any time.

Can I get a trial subscription?

All visitors get 10 free pageviews (based on IP address) to sample what National Statistical has to offer. Opening a free account extends that number to 20, and every account receives 100 tokens for our API.

Do you offer discounted subscriptions?

At least once a month, we offer time-sensitive and limited edition coupon codes offering 10 to 50 percent off certain services, which we distribute via our periodic mailing list and our primary Twitter account.

Do you offer group discounts for teams, leagues or organizations?

Yes. Please contact our support desk with a description of what kind of access your group is looking for. In such cases, we often issue a special coupon code with which your group members can sign up for their own accounts at the discounted rate.

I paid for a subscription, but I can't access the site.

Sorry to hear that. Please use our contact form to get in touch with our support desk, and please make sure you have natstat.com and zendesk.com whitelisted in your email client (or check your spam folders for a reply). Otherwise, it might be difficult to get back to you.

I signed up for a monthly subscription. How do I cancel it?

You can easily cancel via one-click unsubscribe on your account page, via Paypal's website, or by contacting NatStat support. Please note that we do not offer partial month refunds.

What personal information do you collect, and how do you use it?

The GDPR requires websites doing business in the European Union to provide a clear, concise summary of personal data usage. National Statistical collects email addresses that are used as login credentials and as a means of communication, including a periodic e-mail list that provides one-click opt-out.

The website uses cookies only to keep customers logged into their accounts, and we track visitor IP addresses and URLs they access. We do not store or utilize third-party cookies in any way. (Customers who use Facebook or Twitter logins may have their National Statistical usage tracked.)

We do not store any personally identifying or financial information in our database. All payment services are performed by third parties which provide no account management.

Data

When is game data available?

Pro league boxscores and play-by-play are usually posted within a half-hour of the final buzzer. College basketball scores for larger conferences are available in the same timeframe, though smaller conferences might take more time. In some cases, small-conference data may be delayed up to one day due to data collection issues on press row. Stat grids and tables are updated daily around 3 AM ET.

When are statistics and rankings compiled?

Advanced stat calculations for teams and players, rankings, situational stats and insights are compiled at approximately 3 a.m. Eastern time every night, after the close of play in all U.S. leagues. Data exports are generated every four hours, and raw data is available through our API.

Is play-by-play data available for all games?

Basketball play-by-play data is available for all U.S. pro leagues, as well as nearly all NCAA women's (Division I) and men's (Divisions I, II, III). Play-by-play for international leagues is not currently available. Baseball play-by-play data is available for MLB, MiLB and KBO, as well as for approximately 80% of college games.

Why do some NCAA basketball play-by-play files contain certain stats, and others don't?

StatCrew, the standard game event spotting software used by the NCAA or decades, does not have consistent versioning across the 1000+ member schools. Our current policy is to post the STATS Inc. on-site spotters' play by play when available -- which does not include the same level of precise detail as StatCrew -- 24 hours after game completion if we do not receive a usable play-by-play file with fully squared stats from press row by that time.

Why do some NCAA basketball games have incomplete, incorrect or no officials data?

Our officials data at the college level (MBB, MB2, MB3, WBB) is dependent on data entry by StatCrew on press row, who type names into the system based on how officials identify themselves verbally before games -- if they record officials' names at all. We make a diligent effort to correct and normalize misspellings, but due to a flawed information-gathering process that is out of our control we can not make any guarantees about the accuracy of officials data on our site and API.

If you are an official with a complaint about how your name and stats are rendered on National Statistical, please instead consider printing your full name on business cards to distribute to StatCrew before games, so as to guarantee that information will be correct on our site and elsewhere.

What do the zones and X/Y coordinates on your pitch charts correlate to?

Pitch locations are converted and simplified from various data sources to a 0-to-100 scale on X and Y axes. Zone quadrants are numbered from 1 to 9 in three rows left to right, from the pitcher's perspective. The outside zones 11 through 14 are also laid out left to right from the pitcher's perspective.

I found some incorrect or incomplete data on the site.

Please check our Database Status page for a list of known boxscore issues first. If the issue you found isn't included there, let us know with the contact form.

What type of usage license do you offer with your data?

Usage of this system and provided services grants a non-exclusive commercial license for any and all data retrieved via our websites and API. We are not responsible for any data or financial loss/damages that arise from use of our systems; please refer to our disclaimer for more information on liability, warranties and exceptions/exclusions. The design of our databases and delivery systems is ©2007-2018 National Statistical, all rights reserved.

What is (Adjusted) Presence Rate?

Presence Rate (PrRt) is a basketball statistic we invented to track overall player activity during a game. It is calculated as the percentage of a team's play-by-play entries that a player initiated, and Adjusted Presence Rate (APrRt) is weighted positively with scoring plays and negatively with turnovers and fouls. It is somewhat more accurate as a measurement of a player's overall in-game contributions than Usage Rate, as it measures activity beyond shot attempts and turnovers.

Website

How often is the website updated?

Scores are updated in real-time as available. Social posts and news items are updated every 15 minutes. Boxscore pages are updated shortly after data becomes available. ELO and Context ratings are regenerated every two hours. Team and player rankings and situational stats are tabulated nightly. Data insights are generated every morning. Schedules and rosters are updated three times daily.

How can I make a watch list and track players and teams?

Click on the symbols in any player list on the site to add them to a custom list for scouting, watch-list or fantasy purposes. You can view, sort, split and export stats for your chosen players and teams in the account section, and opt-in to receive alerts at 6 AM Eastern time when there are relevant games.

API

What is the difference between the National Statistical API and Data Export?

Data Export, included with all Pass subscriptions, allows you to download any stat grid in comma-separated (CSV), XML, JSON, Excel or OpenOffice format. One-click downloads of team, player, game, play-by-play and officials data is available from the top menu on a full-season basis or new data from the past 48 hours.

Customers needing a more dynamic, flexible solution can sign up for our REST-ful API. Our API offers 16 different endpoints and raw, up-to-the-minute data from all 16 of our vertical subsites. Full documentation, including data examples and code samples, can be found here. All National Statistical accounts receive 100 non-expiring API tokens to test the system with before making a purchase.

Can I view samples of the API output before subscribing?

We have over 30 downloadable XML test samples for all 16 of our data endpoints as part of our API documentation.

How do I make a call to the National Statistical API?

Calls are made in URL format, and are required to specify a valid endpoint, competition level (one of 16 available), as well as a valid 11-character API key. For information on how to properly format queries, please visit our API documentation.

Does an API Plus account offer website access?

API Plus accounts are intended for developers with data needs beyond the rate limits of the NSC Full Pass or NSC Monthly Full Pass. Access to website services must be purchased separately, if needed.

Forecasts

How do you calculate your game predictions?

Basketball score predictions are based on teams' offensive and defensive efficiency, with multipliers for the league average, home/away location and fatigue. The underlying basis for the formula is this blog post from The Basketball Distribution from 2009. The average overall performance of this model since April 2018 is 67.6% correct.

Baseball score predictions are built on John White's MLB game predictor model, taking into account both teams' ERA, winning percentage and OPS, as well as the projected starting pitchers. The average overall performance of this model since April 2018 is 57.1% correct.

Hockey score predictions are based on a simplified Corsica model as described here. The average overall performance of this model since April 2018 is 58.6% correct.

What are ELO ratings?

ELO ratings were invented by physics professor Arpad Elo to rate chess players, but the formula is flexible enough to handle any type of competition. Teams gain points for winning games and lose points after losing. The number of points the winner wins is the same as how much the loser loses, so it is a zero-sum system. More points are transferred after upset wins, games won by wider margins, and playoff games.

Because ELO ratings show relative strength of teams, we can generate win probabilities for games based on the ELO rating difference between the teams involved. National Statistical calculates these probabilities before each game, presents them on the website, then highlights in green for correct and red for wrong.

You can read more about the details of ELO ratings and adjustments at Baseball Prospectus, Hockey Analytics or Opisthokonta.

Your ELO ratings seem slightly different from other sites, like FiveThirtyEight. How do you calculate them?

Three important philosophies when calculating ELO ratings are the K Factor, tuning the EH (expected outcome probability), and the offseason adjustment.

K, or change velocity, is an arbitrary number plugged into the formula based on the value of a single game in a season. For basketball leagues, National Statistical sets the K factor at 20. In hockey it is 8, and baseball it is 4. A higher K factor will result in larger swings and a higher number of ELO points being transferred.

Two other factors are added to EH: The weight of home venue advantage for basketball is set at 100; hockey and basketball home advantage is factored as 20. Additionally, an "important-ness" multiplier and a victory margin multiplier are added to the K and EH factors. Regular games are set at 1 (even), college conference games at 1.1, league playoffs at 1.25, playoffs at 1.5 and 1.75 for the finals. The margin of victory is factored in as a power of one-third of the final score difference.

At the end of each season, National Statistical's ELO ratings "reset" by either giving or taking half the distance between the current rating and 1300. This takes into account changes in team rosters and coaching.

How do you calculate your money lines and spreads?

Money lines and spreads are provided by Betsson in Sweden, as they are the only service that provides hockey, baseball and basketball lines from all continents. Money lines are translated from Decimal format to American format. We are a paid customer of Betsson and do not formally endorse the company in any manner.