I see a lot of people use BABIP statistics as a guide for determining a pitcher's future success based on whether his performance to date has been"lucky" or "unlucky". As I understand, the argument is that pitchers have minimal control over balls put into play, so a high BABIP suggests that balls have been dropping in or finding holes in the infield and that over time, that pitcher's "luck" will turn and those same batted balls will start finding the gloves of fielders for outs. So the general theory goes that a high BABIP indicates that a pitcher has been pitching over his head and his success is likely unsustainable. By contrast, a high BABIP suggests that the pitcher hasn't been getting the breaks and is due for better days.
This all makes general sense to me, but I have a question: Aren't good pitchers more likely to post a low BABIP simply because they throw pitches that are harder to hit? By the same token, isn't a below average pitcher more likely to give up hits because his pitches aren't as good and are therefore more hittable? It seems like a big assumption to me that all pitchers will gravitate toward some magical mean (what is it, .290?). Can anyone enlighten me?
I think BABIP is a valuable metric, although I tend to use it more with hitters.
With regards to pitching, you're absolutely right, which is why I don't think it's wise to compare pitchers with league average BABIP. However, given enough track record, BABIP is an invaluable measure of a pitcher or hitter's 'luck' in relation to their historical or career BABIP.
BABIP does not tell the whole story of course - significant changes in the groundout to air out ratio of a hitter or pitcher can tell a story as well, etc.
Much of the analysis of these types of numbers ends up being subjective, but in my own experience comparing a players BABIP to his own historical numbers seems to be a safer bet when trying to analyze the role of luck in player performance. (and therefore forecasting statistical trends based on this data)
1) For a player who is so far outside the norm that its obvious they'll regress. 2) For a player who has a 2+ year history of stats already that is way outside his own norm.
When a pitcher has good stuff it shows up in other areas more than in BABIP. GB%, K rate, BB rate etc. If you look at the leaderboards in BABIP year after year you'll find a mix of good and bad pitchers at the top and at the bottom of the list and many years a pitcher will be at the top of the list one season and then the bottom the next year.
With a hitter you have to look at their career norms more than anything, some hitters clearly display strong BABIP skills probably because of their approach at the plate and/or speed.
I'm just dipping my toe into some of these more obscure stats - BABIP in particular. While I don't want to go overboard, I would find it useful to hear about a handful of other tools/stats (beyond the ususal "scoring" stats, of course) that some of you more experienced managers use. In fact, a board dedicated to that would, in my mind, be a good idea. Perhaps a pinned thread on each stat for folks to ask questions on, and for the more experienced to share their views etc on...
jswede wrote:I'm just dipping my toe into some of these more obscure stats - BABIP in particular. While I don't want to go overboard, I would find it useful to hear about a handful of other tools/stats (beyond the ususal "scoring" stats, of course) that some of you more experienced managers use. In fact, a board dedicated to that would, in my mind, be a good idea. Perhaps a pinned thread on each stat for folks to ask questions on, and for the more experienced to share their views etc on...
There is a thread on Page 1 called Value Over Replacement Player (VORP) that I posted yesterday that you may want to read. The author compares VORP to Runs scored (although he does have a fundamental flaw, IMO). I also suggest searching Leftovers for terms like EqA, SABR, and sabermetrics. A lot of things have been touched upon at one time or another; some dug deeper into (more on that in my next paragraph). Baseball Prospectus is a good place to start perusing around, as you test the waters. A book I recently picked up Baseball Between the Numbers by BP may intrigue you as well.
Your idea for a forum or what have you is a great one as far as spreading of information is concerned, but unfortunately topics such as that one rarely if ever attract much attention as, oh, say, what A-Rods stool sample looked like this morning.
bigh0rt wrote:what A-Rods stool sample looked like this morning.
Was it blond with a big rack?
Remember that BABIP isn't the only "luck" metric you want to look at for a pitcher. There are also the BHIP% and LOB% metrics that can be very helpful. If you need a refresher on BHIP% vs. BABIP, let me direct you to this post...
Someone extraordinarily intelligent wrote:BHIP% calculates the ratio of ground balls that end up as hits (generally, that would be singles). Without going too deeply into the theory, BHIP% says that a pitcher's basic skill set will result in X% of ground balls ending up as singles and that number will be different for each pitcher. There will be inevitable variations, but the theory is that pitchers will tend to regress towards their natural mean if nothing else changes. When analyzing BHIP% numbers, you can check variation within a pitcher's BHIP% calculation to see if there is an explanation for that pitcher's stats. If a pitcher has a particularly low WHIP as well as a BHIP% that is below his personal average, you can bet that his WHIP will probably be on the rise soon provided that no other factors (namely, infield defense) have changed dramatically.
BABIP is a little more broad. Instead of focusing on ground balls that end up as singles, it focuses on all balls hit into play. Home runs are excluded from this calculation as they are never technically in play. BABIP suggests that there is a natural mean for all pitchers when it comes to the number of balls that end up as hits after being placed in play. I don't have the long-term history of the calculation, but I do know that it has been right at .300 for quite a few years now. The theory is that few pitchers have the ability to vary wildly from this average for an extended duration (Smoltz is a good example of a great pitcher that falls into this average). You can use this stat in the same way that you use BHIP% by checking to see if a pitcher's current stats make sense given his BABIP ratio.