Sabermetrics
I have been following this blog from about the beginning of the season and have grow increasingly more interested in sabermetrics and how and when to use each stat. However, I only have a rough understanding of how to apply and what the different stats mean. I was wondering where a beginner, such as myself, would start when trying to build a rudimentary knowledge of how these stats are calculated as well as applied. Anything helps. I would also like to note that the moderators of this site so an excellent job. They have, from what I have seen, only written articles with interesting and insightful information. I am a Mariners fan but I have visited other sites similar to this one and Lookout Landing is far and away the best that I have seen.
27 comments
|
0 recs |
Do you like this story?
Comments
Nice to have you here.
For a better understanding of Sabermetrics, I would spend time at the websites FanGraphs, StatCorner (run by LL’s own Matthew & Graham), and The Hardball Times.
I also highly recommend The Book by Tom Tango.
Oh, and on the left hand side of the main page of Lookout Landing, you will find a “References” section. The links provided here offer a wealth of good information for you.
The best place to start is the recommended reading section found here, USSM, and Statcorner
as well as what you’ve been doing, which is to say following the discussions. Keep track of what numbers people quote and for what purpose, and don’t be shy about asking questoins; the locals are usually more than happy to share their knowledge. Also, there is a Fanpost/Diary that asks this question about twice a season so you may want to search the archives if you feel like putting in the work.
In general terms though, sabermetrics is about using stats that best demonstrate players’ actual skills as opposed to the results they produce (results being so strongly affected by luck and environment). Thus, for example, ERA and WHIP is not a good indication of how good a pitcher is because so many factors play into them.
tRA and FIP are ERA-like stats which incorporate elements considered to be true pitcher skills and scale them to ERA. Swinging strike rate is usually a good indication of stuff, (particularly out of zone swinging strikes) while called strikes (and therefore called strikeouts) tend to be more umpire dependent, with the exception of guys like Bedard, who have crazy good command of a killer breaking ball. It’s useful know a pitcher’s batted ball profile (GB=good, LD=bad, IFF=random) and there are pitchf/x tools out there that allow you to see how effective each pitch type is for a given pitcher (hilariously, Felix’ fastball has been his best pitch by linear weights).
When looking or signs that a pitcher has been inordinately lucky, look at their strand rate (LOB%) being above ~70% (72-74 for high K guys), HR/FB rate well below 11%, and batting average on balls in play, which should be in the neighborhood of LD%+.12 or so (there’s been some new research on this which I’m ignorant of so this may need to be corrected.
Hitters are much easier to analyze. wOBA is scaled to OBP and evaluates a hitter’s production independently of situation and hitting environment. Every result of a plate appearance has a certain ‘run value’ which is to say, the average number of runs that have resulted from such an event over the decades of recorded baseball games. wOBA averages those run values and gives you the production of the hitter, incorporating HR, singles, doubles, triples, walks Ks, GIDP, SB, CS, IBB, etc etc. It’s a great number for production. BABIP is also important to look at, but each hitter has his own special ‘normal’ BABIP unlike pitchers. For power, it is good to look beyond ISO and check out HR/FB (take a peek at Griffey’s over the years). Plate discipline can be examined in terms of walk rate as well as out of zone swing rate and contact rates in and out of zone, and pitches per PA is a nice number to be aware of as well.
OK, that’s a start. it is not exhaustive by any means. For defense, I suggest you mosey over to Fangraphs and read Dave Cameron’s six posts outlining the use advanced defensive metrics that he put up prior to their publishing UZR. The gist is you need about three years’ worth of data to get a meaningful number and look at the numbers as estimates rather than hard data. Players’ defensive contribution is generally thought of in terms of X +/- 5.
"Even the stupidest of men, by some instinct of nature, is convinced on his own that with more observations his risk of failure is diminished."
-Jacques Bernoulli Ars conjectandi 1713
by Bearskin Rugburn on Jun 8, 2009 6:35 AM PDT reply actions
I personally found the book "Baseball Between The Numbers" to be a great place to start
And as long as you are respectful, asking questions is the best thing to do.
Agreed on both counts.
That book was a great start. And just reading LL/USSM and asking questions will help. Comments like “can someone point me to an explanation of UZR?” are not going to get you ridiculed or anything.
There was an open stat question thread a little while ago.
If someone can find the link to that, that’s also a good place to read through. (The recent one, not the first one)
Not sure if that one is such a good place for a beginner - we talked mostly about advanced stuff in there
here’s the link, it could still be helpful if you bumped/re-opened comments.
Offended is completely the wrong word though
LL has a pretty strong no-politics policy, and many people that are vigilant about enforcing it. So I’d say it’s more “set off a lot of red flags” than “offended”.
Nice Guys Finish Third - My semantics are a waste of time.
by pdb on Jun 8, 2009 1:02 PM PDT up reply actions
Nigga Owens stole my bike?
It's hard to convince people to let you eat them if you're an asshole. - Thingray
Thanks for your help.
I have changed my picture to me pretending to play a guitar. I hope this is less political.
Thug Life
Depends.
Are you playing ‘we shall overcome’? ‘Eyes on the Prize’? ‘Fuck the Police’?
"Even the stupidest of men, by some instinct of nature, is convinced on his own that with more observations his risk of failure is diminished."
-Jacques Bernoulli Ars conjectandi 1713
by Bearskin Rugburn on Jun 9, 2009 6:13 AM PDT up reply actions
He's playing Kumbaya, and you'll fucking like it!
Awaiting the day I catch a Russell Branyan foul ball. I will make love to it.
Preserved In All His Greatness - R.I.P. The Reignman 1989 to 1997

by 










