Introducing The FanGraphs Library
While we’ve always been very proud of the quality and quantity of the statistics offered here on FanGraphs, we have never had a proper place to explain exactly what they all are, what they mean, how they’re calculated, and how they are best used. That’s why we’re extremely excited to announce the newest section of FanGraphs – The FanGraphs Library.
Steve Slowinski, creator of the excellent SABR Library, has agreed to come on staff and maintain the Library here on the site going forward. He’s created an in-depth resource that will serve as a place for explanation and education on all of the metrics we have here on the site, as well as many of the principles that we often refer to. He’s added some extremely fun tools to the pages, including calculators which will let you reproduce the metrics on your own, charts that show not only league averages for each metric but also the spread of those metrics from best to worst, and links to various explanations of how these metrics were conceived.
It’s really a fantastic resource, and one we’re thrilled to have on the site. Steve will be blogging both over at the Library and here on FanGraphs, and we’ll continue to update the Library going forward. We hope you enjoy using this resource as much as we’ve enjoyed bringing it to the site.
Hooray!
I’ve been waiting for this moment! (For part of my life.)
Thanks for the intro, Dave. If anyone’s flipping through the Library and finds things to be changed or has any questions, let me know. Contact info is on the Library sidebar. I want this to be as accurate as possible for everyone, but obviously I don’t know everything….probably the opposite of that, really.
Fantastic addition to the site.
The number of baseball fans who have limited exposure to sabermetrics still far outnumbers those who do. When I discovered fangraphs, about a year ago, I had to piece most of it together by searching older posts, something the blog format is pretty terrible for. I stuck around, and eventually picked up most of the obscure jargon, but I’m sure there were plenty of other casual fans who didn’t think it was worth the trouble.
Very cool stuff.
This is really, really awesome! I can’t wait to read it all.
Something I’ve been really wanting to know since Fangraphs started a series on historical WAR:
How accurate is it, especially for historical players? Does anyone know?
I read via your Library that it is calculated using Total Zone, but I’m not sure what level of reliability we can have in them. Can you explain that?
With what level of confidence should we use it in comparing historical players to today’s players?
I see it bandied about as though it’s near gospel, but how reliable is, say, Tris Speaker’s WAR? I’d be willing to be convinced it’s reliable within +/-5% over the course of his entire career versus today’s players (that is, his WAR has a +/- variation of an additional 5% over today’s players.) Then again, I could be convinced it’s reliable within +/-30% over the course of his career, too. Point being, I don’t know.
I have never seen anybody take the time to explain it, though, and/or put up confidence intervals for historical WAR. Maybe that could be a good article for this site? I’m sure I’m not the only one who wonders about this stuff.
Thanks for considering the thought, and for all your hard work.
This is worthy of a whole blog post, but I’ll try and give you a quick answer for now. Obviously there’s uncertainty and error in all the iterations of WAR, so when using whatever version you choose, you should just be aware of those limitations.
In this instance, yeah, TZ may not be as “accurate” as UZR or some of the other defensive stats. With that in mind, you shouldn’t compare Ted Williams’ 1947 rWAR to Albert Pujols’s 2010 fWAR. They’re two different calculations, two different scales, and it doesn’t make sense that way.
When you use historical rWAR figures, though, normally it’s to compare players with other historical figures. All these players rWARs are calculated with the same inherent biases, so they’re perfectly fine to use in this context. Who knows if their values are 100% accurate, but since they’re all calculated the same way, they provide a reference point and a way to compare between historical players.
Anyway, that’s just some thoughts off the top of my head.
Great idea! Here’s another one: How about a link to it on your home page? I don’t see one anywhere.
Top right corner, labeled glossary. The drop down has links to the library, the old (inferior) glossary, and MGL’s primer on UZR. We’ll try to make it more prominent, however.
Kudos to your baseball passion. All y’all.
sweet
This was literally the only thing that I thought needed a little improvement on Fangraphs, and now it’s fixed. You guys continue to rock.
Very, very cool.
A couple of years ago, I reacted to Sabermetrics like it was Mahmoud Ahmadinejad coming over for dinner.
Since then, I have made at least made an attempt to read, educate and observe the teachings. Most importantly, I have developed a certain, reasonable amount of trust of the Sabrepeople, if you will.
I am happy to report, I don’t believe you are pinky communist fags bent on destroying America’s past-time. In fact, I have developed respect, for (in general) what is a group of dedicated, hard-working, enthusiastic baseball lovers. The FanGraphs Library is just another example of what I’m talking about, and it is very useful for even non-Sabergeeks.
I want to thank Mike, Dave (who caught my attention with his fine work at USS Mariner) the rest of the people on this site, and other sites like it. Keep up the good work, and remember to have tolerance, even in the face of the Secularists.
This is awesome. In future updates, will we be able to run the spread charts (best, 90th, 75th, 50th, 25th, 10th, worst) by position? For example, I’d love to see wOBA spreads for catchers only, or say right fielders.
The library looks outstanding. Steve did an awesome job putting it together. It’s going to be a great resource for people just getting into sabermetrics,
One word: Awesome.
Thanks, guys.
Serious win. The lack of an in-depth glossary was Fangraphs’s one major weak point, and there couldn’t be a better fit for Saber Library’s content. Great merger, can’t wait to see it develop.
Now you guys just need to take over ESPN and we’ll be all set.
…Although I did just notice that some of the old links to Saber Library articles seem to be broken. I know I’ve seen a bunch of links to the library in the wild recently — is there a way to integrate that content into the new site so that those links don’t die?
Enjoy the sabermetric tab, and the park factor consideration, if not full exploration. I think the historical impact of foul territory as a park factor affecting SO by increasing or decreasing foul balls caught on the field (and opportunities to continue an at-bat, and hit a groundball, or a flyout, or a hit, or a BB) has been underestimated. Certainly underquantified. My post on facebook quantifies the effect from 1954-2000 and makes a strong point that it might have contributed to the SO peak of the 90?s. Foul territory and not steroids, or both?!
http://www.facebook.com/photo.php?fbid=191364250889624&set=a.187432391282810.54968.187430347949681#!/album.php?fbid=187432391282810&id=187430347949681&aid=54968
This is a joyous day for me. Baseball has kindled a very latent respect for math (I was one of those lib arts guys at university) so I look forward to learning all about what I’ve been missing.
thank you for doing this. hopefully this gets high on google. it’s a real boon to anyone who isn’t following the conversation in real time.