This past Monday, Baseball Prospectus unveiled their brand new leaderboards. There are leaderboards for hitting, pitching, and catcher defense. The intent is to make them more user friendly from a computer or phone. Here at Beyond the Box Score, this is right up our alley, so we couldn’t wait to try them out!
Baseball Prospectus has been the standard bearer for the public sabermetrics movement for more than 20 years. Inarguably, the precedents set by BP for how we develop, interpret, and utilize advanced metrics have dramatically improved our collective understanding of baseball. Beyond the Box Score probably wouldn’t even exist without them, nor would several other analytically inclined sites.
However, it became clear a few years ago that, while the work of BP’s writers and statisticians remained superb, the site itself had been lapped in the field. As documented by Lindsey Adler at Deadspin, the erstwhile owners had nearly run it into the ground. Since then, BP has been repurchased by the staff writers, who appear to be bringing the site back in the right direction.
But standing still means falling behind, and BP stood relatively still for a few years. Other sites like FanGraphs and Baseball Savant kept growing, and it became evident that BP no longer held an advantage over their competitors. Former BP Editor-in-Chief Steven Goldman, speaking on his podcast, The Infinite Inning (extremely highly recommended), recently summarized the current pecking order:
As someone who has spent great deal of his life caring about Baseball Prospectus, trying to make Baseball Prospectus better, and fighting bitterly at times to make Baseball Prospectus better, it bothers me that Baseball Prospectus now seems to serve as a farm team for FanGraphs.
While I don’t mean to drudge up painful recent history, an I should reiterate that the BP staff have always been fantastic, this context shows how badly they needed to overhaul their leaderboards. One facet at which BP has always excelled is developing new proprietary stats. However, their nearly unusable leaderboards nullified the work of their exemplary stats team. Revamped leaderboards are a necessary step towards regaining prominence in the industry.
Given that Beyond the Box Score is a “A Saber-Slanted Community,” I’d like to think we’re the target audience for stat leaderboards. Before we dig into them, I should disclose that I used to contribute to Baseball Prospectus-Bronx. Be that as it may, I don’t have any real connection to BP anymore, and the local sites have now been resorbed into the parent operation anyway.
Let’s see what the new leaderboards have to offer!
We’re excited to introduce the new Baseball Prospectus Leaderboards. The new leaderboards give you a new look at our hitting, pitching, and catcher defense statistics.— Baseball Prospectus (@baseballpro) June 24, 2019
And they work on your phone. https://t.co/6cl9ARWP4T
The new leaderboards are a huge improvement to Baseball Prospectus. This is another leap in the right direction for the new BP management group. Kendall Guillemette and the BP team deserve to take a bow. There are still some flaws I’d like to see resolved, but they are miles ahead of where they were a week ago. The bottom line is the exceptional work of the BP stats team is allowed to shine, which should help ingrain them into modern stathead parlance.
The biggest improvement of the new leaderboards is how clean and readable they are. The home site for the leaderboards provides a brief introduction, then allows you to choose from three prominently displayed selections: hitting leaders, pitching leaders, and catcher defense. After you click one, you are taken to a user-friendly leaderboard with a font that’s easy on the eyes and loads quickly, even on your phone. The old leaderboards were minuscule by comparison, and it’s hard to use stats you can’t see!
Each leaderboard is customizable in several ways: position, lengths of time, season, level, league, team, and playing time (plate appearances for hitters, innings for pitchers, and chances for catchers). You can also easily sort by any stat, ascending or descending, by clicking on it. It took me just 23 seconds from opening my browser to discovering that Eddie Rosario is last among MLB left fielders with -3.1 FRAA over the past two weeks. This speaks well of the load times, sort functions, and overall ease of use.
My favorite feature is the “Download CSV” button. After you have sorted the leaderboard as you see fit, you can download it into a spreadsheet that opens in Excel and Google Sheets. I use this feature all the time with other leaderboards to compare players using different stats or create my own, depending on the project I’m working on. Knowing this is available from BP will significantly increase the chance I use their metrics. It’s also a nice touch that a “Version Date” column is included on the spreadsheet for those of us who need to tidy up our file folders!
They also retain most of the positive features from the old leaderboards. I love the idea of splitting innings pitched into “IP as SP” and “IP as RP.” “Swings” is also a great, simple category that makes a ton of sense. BP’s proprietary stats are toward the front, as well they should be. Furthermore, their evaluation of catcher defense is second-to-none. It’s fantastic that it still has its own separate leaderboard.
While the new leaderboards are a major step forward, there’s plenty of room for improvement. They’re still in beta mode, and there’s a bright red “Feedback” button on the bottom right corner, so undoubtedly they’ll make a few tweaks in the next few weeks.
The first flaw I found is probably also the easiest to fix: when you scroll to the right the player names disappear. The names should be locked so that they are always visible. This will vary depending on your screen, but you can see what I mean here:
Perhaps the most pressing need for improvement has to do with the stats themselves. I assume Baseball Prospectus readers are generally knowledgeable with regards to advanced stats, but not everyone will automatically know what DRC+ means. I’m not even clear on all of the catching stats, such as EPAA. A linked glossary page is essential; mouseover explanations would be better still.
We also need to talk about the playing time minimums. Most users will figure out that they refer to plate appearances for hitters, innings for pitchers, and total chances for catchers. However, it doesn’t actually say so. There will most likely be some folks who think they are sorting by at bats or batters faced.
These minimums can be selected in increments of 50 for batters, 25 for pitchers, and 250 for catchers (you can also select “all” or “qualified”). I assume these increments will increase as the season progresses, but it would be preferable if we could just input our own minimums instead of choosing from a dropdown.
It would also be nice to have more date customization. For 2019, we can choose season, past week, past two weeks, or past month. Therefore, if I want to know who has been the best hitter since September 1, 2017, I’m out of luck.
Lastly, while I do love the downloadable .csv spreadsheets, some of the stat names get a little wonky. For one thing, they’re all lowercase. I know I’m being persnickety, but “DRC+” looks more professional that “drc_plus.” It also took me a few moments to recognize that these weird stats shown below meant singles, doubles, and triples:
For the Future
Once again, these new leaderboards are merely the beta version, meaning the BP folks aren’t done tinkering. In truth, they never are. With relatively frequent updates to various aspects of the site, it’s clear that they will continue improving wherever they can— including the leaderboards.
In addition to the above suggestions, I would love to see a wider variety of advanced stats. One of the best advances propagated by BP recently has been pitch tunneling. We’ve barely scratched the surface of what we can learn about tunneling, largely because of the difficulty of using their old leaderboards. For these reasons, it’s curious and disappointing that their pitch tunneling metrics aren’t included on the new leaderboards (yet).
Even with tunneling set aside, there’s still a lot more BP could include. On the pitching leaderboards, there are 34 statistics listed (not including basics like name, age, team, and league). At FanGraphs, you have the choice of 318, and while some of them are redundant, it illustrates a wide disparity between the two sites. For example, If I want to customize a leaderboard comparing FIP, slider velocity, and swinging strike rate, FanGraphs is the place to go. BP still isn’t close to that level of customization.
Perhaps this is a challenge for BP’s stat developers. They have done a phenomenal job with catch-all stats. I believe DRC+ is the best comprehensive metric for total offense, as is DRA for pitching. In fact, I don’t have to believe it— they’ve provided the proof!
Be that as it may, BP has largley attempted to answer “who” questions, as in “Who are the absolute best players?” whereas the rest of the industry has been asking “why” questions. Recent revelations in baseball analytics involve metrics like launch angle, exit velocity, and spin rate. Just a few weeks ago, Mike Petriello and the Baseball Savant crew introduced us to outfielder jumps. These are granular stats that answer a unique, specific question. DRC+ answers the same question as wRC+, OPS, or oWAR. It’s a better, more accurate answer, but it’s not a new question. By no means is this less important or impressive. BP answers the “who” questions better than anyone else, including providing better data to justify their new stats.
It’s just as imperative for the sabermetric community to revise and improve our answers to existing questions as it is to answer new ones. BP leads the way with the former. Now, perhaps they can make more inroads with the latter. There’s still plenty left to discover! As they do, their leaderboards will grow in size and value.
Once again, the new leaderboards are a significant, necessary leap for Baseball Prospectus. I’m excited by the site’s rapid rate of overall growth, and this is a major contribution. Kudos to the BP leadership for knowing what they needed to fix, and to the statheads and web designers that made it happen.
Daniel R. Epstein is an elementary special education teacher and president of the Somerset County Education Association. In addition to BtBS, he writes at www.OffTheBenchBaseball.com. Tweets @depstein1983