I’ve said the following in a previous Unfiltered post as well as my most recent chat, but I’ve seen it come up again a few times since then, including in Will’s recent “Focus” post: Baseball Prospectus 2010 will have an index. It really, truly will. By Odin, Zeus, and Babe Ruth, there will be an index.
Look, we feel pretty rotten about the absence of an index in the 2009 edition. We had a perfect storm of things go wrong, and it was put to us that if we insisted on an index the book might be late or not-at-all. Also, they threatened to shoot a puppy. Several puppies. Really cute ones. We figured it was better to have a book without an index than an index without a book. Note that we are with a new publisher this year. The Index Incident is not the whole reason we moved, but it’s not unrelated either. We take anything that compromises the reader experience personally.
This year we had a far more pleasant experience cooking the book and an index was part of it, prepared with care by Christina Kahrl. It is really, really there. Honest. Again, we contritely apologize for last year’s omission, but it is a thing of the past. The index is back and it is here to stay.
It's part of the double-edged sword guys. You can't have it both ways. When you're at the top like BP is, you're going to get heat when an index is omitted. Not to mention the numerous typos throughout the book. I was not one that ever complained about the index last year; I'm just giving a reason why you got the heat you did.
Eh, every book has typos. I got a kick out of Hubbard misspelling Psychlo as Pyschlo and other variants in his own sci-fi book. I don't remember being overly distracted about the typos in last year's annual. Include an index and axe FRAA (or at least, have player comments match what FRAA says) and I'll be happier :)
Richard, I think I've commented on this before, but let me do so again. We want the writers to feel free to disagree with stats like FRAA. Their worldview should not be limited to numbers on a page, but should also embrace what they see. Obviously if a player hit 25 home runs then he hit 25 home runs, but with some of the fielding metrics and projections there is more room for interpretation. If that means saying they disagree with the PECOTA projection for a player, fine. If it means saying that FRAA over- or underrates a player's defensive abilities, that's great too. Sometimes it's kind of important that we do. For example, part of the Mark Teixeira comment is spent discussing why FRAA (and UZR as well) didn't like him this year. That's a spot where the metrics and perception clash, and it's important to acknowledge that and try to understand what is happening.
I don't mind (and in fact, like it) if writers disagree with the metrics... but sometimes the comments are written in such a way that don't even acknowledge the metrics. For example, the comments will say someone has a bad year of fielding when their FRAA stayed stabled for the last three years. Yet, they won't even say why or how the fielding got worse... it's that kind of stuff that bothers me, where it almost seems like the writer didn't look at the statistics.
Btw, I have this pet peeve about a lot of fantasy mags too.
I'll go ahead and make this quick note on defensive metrics and BP, and then I'll crawl back into my spider hole and keep working on them:
Starting Friday, you should get your first glimpse at our new defensive metric. I have already started writing the first article of the set, and I'm happy with what's shaping up so far.
It is still a ways away from being finished, and it is going to take some time to roll it out to you, both in terms of doing the work to get it up and running, and then taking the time to explain the work without simply dumping a lot of formulas and name-dropping some fitting techniques. And without the benefit of time travel there's no way to get it into the '10 Annual.
I could sit here and talk your ear off about this, but I doubt you'd believe me until you actually saw it in action and so it's probably better if I just finish this article and let it speak for itself. But this is certainly being worked on.
I mean, I don't know what we're CALLING it yet. I don't have some clever name for it. Maybe we end up calling it FRAA as well - after all, it's Fielding Runs Above Average as well. At some point you run out of useful names for things anyway. So maybe at BP every plus/minus metric for fielding gets called FRAA, and we simply differentiate which version of it we're using based upon the dataset. This is just me spitballing here - maybe we put the call out for the full backronym squad and try to figure out how to make something fit OZZIE. Who knows.
But yes, it will use a full compliment of play-by-play data, including batted-ball type and hit location.
I'll leave the PECOTA questions for Clay. I can confirm that the new system isn't operational yet and won't be used for PECOTA this season.
There are a lot of replacement players in the world. Some are guys who can hit a little but can't play a position well enough that you'd want them out there every day. Some are slick fielders with Mario Mendoza-like hitting numbers. So when we compare someone to replacement level, what we're really doing is comparing them to a composite of all of those real-life replacements.
And what you don't tend to see in the wild is players who can't hit or field their position, though. Replacements typically can't do one or the other, but not both.
See if this helps you out any. You can break it out further into components, like so:
where the runs above replacement and the positional adjustment are handled on a per-PA basis.
Now we certainly don't have to do it this way. We could put RAR and PosAdj on FRAA and combine that with BRAA. We could adjust batting for replacement and fielding for position and then combine. Or we could list them out separately.
But at the end of the day, it simply does not matter - it's a matter of presentation, not of substance. Any which way you shuffle it around, it all comes out the same in the wash.
(I suspect - although I don't know - that presenting BRAR in that way is done for the sake of familiarity; it's the way VORP has been presented, for instance.)
Thank you so much for your response on this. I've had this question in mind now for months, and I've asked it before on other articles but never got a definitive answer, though I suspected it as much. And I don't necessarily disagree. In fact, it makes alot more sense than my own thoughts about the subject. I dispute with you the fact that there are guys hanging around in the minors who are average offensively but don't really have a position they can field. Most of them do find jobs. I know for a fact that there are at least 14 of them. They are the DH's for AL teams.
Young guys who can't field have it much worse than older players, simply because there is an athletic aspect to it. Players get less athletic as they age, so, if you are 22 and can't run faster than Benji Molina, odds are that you aren't going to get faster as you grow older.
I also wonder how pitching is handled. are FRAA included in their WARP? would it matter? does pitchers hitting stats in the NL change a Pitchers WARP? how is that handled historically?
Wait, so you mean that this year we get an index? I'm going to be really upset if there isn't an index. You jerks, how come you won't put in an index? Why do you hate America?
We edit like crazy, amazin. However, most books of this size are prepared over the course of a year or more, not in a mad rush of just a couple of months. We added an additional layer of copyediting last year, and the number of errors was vastly reduced. I trust the same will be true this year. I offer this explanation not as an alibi but as a simple statement of the limitations inherent in the process.
I remember being off-put (not quite put off, though) by the number of typos in the 2008 book and noting the lack of noting such a typographical abundance last year.
I write and edit technical reports for a living, and there is a margin of error that I can be tolerated in projects of the scope of a BP annual.
I get more pent up on errors like how Matt Brown is wrongly given a birth year of 1972, severely skewing his PECOTA take by making him 10 years older than he really is. (BTW - this typo has been around for a while now) A word typo can be corrected by the reader who will likely (on most common typos) still be able to discern the context of the sentence. I numerical typo, unfortunately changes the context of the surrounding numbers.
That's my two cents (Canadian currency).
Shouldn't the rule be that if you're going to write a post complaing about typos, there shouldn't be any typos in it? Even if "there is a margin of error that I can be tolerated?" But don't get too "pent up."
touche. But I also might note that I really don't care about typos in blog posts.
And if you read what I wrote, I was mostly "complaining" about numerical typos, more than spelling typos.
Although I initially missed the index, it certainly fell into the "forgivable sins" category for me. I also found the Excel index to be very useful. Is it possible to get that for 2010 as well?
31 comments have been left for this post.
It's part of the double-edged sword guys. You can't have it both ways. When you're at the top like BP is, you're going to get heat when an index is omitted. Not to mention the numerous typos throughout the book. I was not one that ever complained about the index last year; I'm just giving a reason why you got the heat you did.
Eh, every book has typos. I got a kick out of Hubbard misspelling Psychlo as Pyschlo and other variants in his own sci-fi book. I don't remember being overly distracted about the typos in last year's annual. Include an index and axe FRAA (or at least, have player comments match what FRAA says) and I'll be happier :)
Richard, I think I've commented on this before, but let me do so again. We want the writers to feel free to disagree with stats like FRAA. Their worldview should not be limited to numbers on a page, but should also embrace what they see. Obviously if a player hit 25 home runs then he hit 25 home runs, but with some of the fielding metrics and projections there is more room for interpretation. If that means saying they disagree with the PECOTA projection for a player, fine. If it means saying that FRAA over- or underrates a player's defensive abilities, that's great too. Sometimes it's kind of important that we do. For example, part of the Mark Teixeira comment is spent discussing why FRAA (and UZR as well) didn't like him this year. That's a spot where the metrics and perception clash, and it's important to acknowledge that and try to understand what is happening.
I don't mind (and in fact, like it) if writers disagree with the metrics... but sometimes the comments are written in such a way that don't even acknowledge the metrics. For example, the comments will say someone has a bad year of fielding when their FRAA stayed stabled for the last three years. Yet, they won't even say why or how the fielding got worse... it's that kind of stuff that bothers me, where it almost seems like the writer didn't look at the statistics.
Btw, I have this pet peeve about a lot of fantasy mags too.
I'll go ahead and make this quick note on defensive metrics and BP, and then I'll crawl back into my spider hole and keep working on them:
Starting Friday, you should get your first glimpse at our new defensive metric. I have already started writing the first article of the set, and I'm happy with what's shaping up so far.
It is still a ways away from being finished, and it is going to take some time to roll it out to you, both in terms of doing the work to get it up and running, and then taking the time to explain the work without simply dumping a lot of formulas and name-dropping some fitting techniques. And without the benefit of time travel there's no way to get it into the '10 Annual.
I could sit here and talk your ear off about this, but I doubt you'd believe me until you actually saw it in action and so it's probably better if I just finish this article and let it speak for itself. But this is certainly being worked on.
BIP data based...? Or just a different version of FRAA/FRAR?
I take it PECOTA will still be using FRAA/FRAR this year, then...?
I mean, I don't know what we're CALLING it yet. I don't have some clever name for it. Maybe we end up calling it FRAA as well - after all, it's Fielding Runs Above Average as well. At some point you run out of useful names for things anyway. So maybe at BP every plus/minus metric for fielding gets called FRAA, and we simply differentiate which version of it we're using based upon the dataset. This is just me spitballing here - maybe we put the call out for the full backronym squad and try to figure out how to make something fit OZZIE. Who knows.
But yes, it will use a full compliment of play-by-play data, including batted-ball type and hit location.
I'll leave the PECOTA questions for Clay. I can confirm that the new system isn't operational yet and won't be used for PECOTA this season.
Oh come on ... just for laughs, call it JETER:
Judged Equity Through Evaluation (of) Range
are we still going to base WARPs off of BRAR + FRAA?
I always found that rather curious.
Lemme see if I can clear that up for you.
There are a lot of replacement players in the world. Some are guys who can hit a little but can't play a position well enough that you'd want them out there every day. Some are slick fielders with Mario Mendoza-like hitting numbers. So when we compare someone to replacement level, what we're really doing is comparing them to a composite of all of those real-life replacements.
And what you don't tend to see in the wild is players who can't hit or field their position, though. Replacements typically can't do one or the other, but not both.
See if this helps you out any. You can break it out further into components, like so:
BRAA + FRAA + RAR + Positional Adjustment
What we call BRAR is really:
BRAA + RAR + Positional Adjustment
where the runs above replacement and the positional adjustment are handled on a per-PA basis.
Now we certainly don't have to do it this way. We could put RAR and PosAdj on FRAA and combine that with BRAA. We could adjust batting for replacement and fielding for position and then combine. Or we could list them out separately.
But at the end of the day, it simply does not matter - it's a matter of presentation, not of substance. Any which way you shuffle it around, it all comes out the same in the wash.
(I suspect - although I don't know - that presenting BRAR in that way is done for the sake of familiarity; it's the way VORP has been presented, for instance.)
Thank you so much for your response on this. I've had this question in mind now for months, and I've asked it before on other articles but never got a definitive answer, though I suspected it as much. And I don't necessarily disagree. In fact, it makes alot more sense than my own thoughts about the subject. I dispute with you the fact that there are guys hanging around in the minors who are average offensively but don't really have a position they can field. Most of them do find jobs. I know for a fact that there are at least 14 of them. They are the DH's for AL teams.
Young guys who can't field have it much worse than older players, simply because there is an athletic aspect to it. Players get less athletic as they age, so, if you are 22 and can't run faster than Benji Molina, odds are that you aren't going to get faster as you grow older.
I also wonder how pitching is handled. are FRAA included in their WARP? would it matter? does pitchers hitting stats in the NL change a Pitchers WARP? how is that handled historically?
so many questions....
But thanks again!