Discussion thread for Dbacks ZIPS projections
Very interesting, and worth checking out. Check out Jeff Bailey's projection. Craaaaaazy. Also, Dan Haren is a god.
about 2 years ago
C. Wesley Baier
33 comments
0 recs |
Email
PrintComments
Which of the below is not like the others?
Name P Age AVG OBP SLG G AB R H 2B 3B HR RBI BB K SB CS OPS+
Adam LaRoche* 1b 30 .287 .363 .520 142 513 80 147 38 2 26 109 63 130 1 1 122
Justin Upton rf 22 .282 .363 .521 141 511 79 144 30 7 26 90 64 134 9 3 122
Mark Reynolds 3b 26 .258 .341 .509 146 530 85 137 28 3 33 106 64 182 7 3 113
Kelly Johnson* 2b 28 .279 .352 .472 127 434 66 121 30 6 14 64 48 84 5 2 108
Conor Jackson lf 28 .278 .357 .441 102 356 50 99 22 3 10 59 40 46 4 1 103
Jeff Bailey 1b 31 .271 .357 .441 108 388 49 105 17 2 15 40 45 87 3 2 102
Miguel Montero* c 26 .275 .342 .451 112 346 44 95 23 1 12 56 33 61 0 1 100
by C. Wesley Baier on Jan 20, 2010 2:36 PM EST reply actions
Oh Oh! Pick me!
Is it Reynold’s 182 strikeouts? Are they projecting him to only play 80 games next year?
by Counsellmember on Jan 20, 2010 2:56 PM EST up reply actions
Well, technically speaking, as the players are ranked by straight RC/27, below-average 1B tend to be pretty high on team lists. Most below-average-but-OK first-basemen in AAA would be major leaguers if they could handle a harder position. Brian Myrow is another example; he usually shows up fairly high on lists, but he would really only be a stopgap/reserve type in the majors. The Yankees tried Myrow at 2nd and 3rd awhile back (in the minors) in an attempt to make some use of his bat, but he was pretty awful.
--
Dan Szymborski
dan@baseballprimer.com
by D.Szymborski on Jan 20, 2010 3:44 PM EST up reply actions
Can you explain a bit about the OODIBE section?
I was looking for info on what the various categories listed there mean, in terms of performance, and haven’t found any specifics. Thx in advance!
"Win, or die" -- Marquise de Merteuil
by Jim McLennan on Jan 20, 2010 5:04 PM EST up reply actions
ODDIBE
These results are obtained from using Monte Carlo to sim a million season of the player in question (based on information from the non-parametric model of a player’s possible aging/development/decline that ZiPS derives from the player’s history and MLB history).
Each number represents the estimated likelihood of the player exceeding the number given in the projected PA. For example, while ZiPS sees LaRoche as most likely a .287 hitter in 2010, it sees a 30% chance that he would hit .300 in the projected 513 at-bats.
The first section in ODDIBE compares each player to the starters at a position. Based on the recent historical distribution of play, LaRoche is projected to be in top 20% of starting 1B (in offense) 18% of the time, the next quintile 37% of the time, the middle quintile 25% of the time and so on.
--
Dan Szymborski
dan@baseballprimer.com
by D.Szymborski on Jan 20, 2010 5:25 PM EST up reply actions
Appreciate the explanation
It’s an interesting number, gives some idea of the “solidity” of the projection.
"Win, or die" -- Marquise de Merteuil
by Jim McLennan on Jan 20, 2010 6:40 PM EST up reply actions
Also
Do you have a quick proj for Max Scherzer in Detroit? He’s not listed with them, as the trade post-dated your publication of the Tigers’ numbers.
http://www.baseballthinkfactory.org/files/oracle/discussion/2010_zips_projections_detroit_tigers/
"Win, or die" -- Marquise de Merteuil
by Jim McLennan on Jan 20, 2010 5:11 PM EST up reply actions
You might want to either
Shoot dan an email to answer the questions in this thread, or ask them on primer. Or I will.
by C. Wesley Baier on Jan 20, 2010 5:13 PM EST up reply actions
ZiPS likes Scherzer a lot – 3.70 ERA in 112 IP for the Tigers. ERA and IP both go down if used as a reliever, obviously.
--
Dan Szymborski
dan@baseballprimer.com
by D.Szymborski on Jan 20, 2010 5:26 PM EST up reply actions
ZIPS loves power pitchers
Max certainly fits that category. Although 112 IP is indicative that we’re not the only ones worried about his arm.
"I've had Bailey's out of a shoe, though."
by Dan Strittmatter on Jan 20, 2010 7:37 PM EST up reply actions
112 IP
Is simply due to him pitching 50 innings in ‘08. Projection systems don’t take injury fears into account.
Actually, they do at times
Webb throws just 162 innings in this projection due to his injury. But, of course, you’re right in this case that Scherzer is docked for your reason because the rest of the innings were in the minors, not on the DL. They dock for injuries, but they also can dock for other things, like not being in the majors. My mistake.
You’d think those minor-league innings would be taken into account… Guess not.
"I've had Bailey's out of a shoe, though."
by Dan Strittmatter on Jan 21, 2010 2:09 PM EST up reply actions
Beat me to it
I’m writing up a piece discussing them in a bit more detail. Should be up tomorrow morning.
"Win, or die" -- Marquise de Merteuil
My opinion of the projections
Too low: Upton, Reynolds, CoJack, Montero, Young, Drew, Webb, E Jackson
Too high: Jeff Bailey, Brandon Allen, Kelly Johnson
Just right: Adam LaRoche, Danny Haren
by C. Wesley Baier on Jan 20, 2010 3:00 PM EST reply actions
Really?
Only 17 appearances and 81 innings from Kennedy? I assume that’s because of his previous injuries?
"I can just see it: Post 'Intervention? Why would I need an intervention?' And then he'd +1 himself."
Like aneurysms
Which happen all the time and recur like crazy.
"I've had Bailey's out of a shoe, though."
by Dan Strittmatter on Jan 20, 2010 4:18 PM EST up reply actions
+1
What's your name? Sandwich. What's your first name? TUNA.
by emilylovesthedbacks on Jan 20, 2010 4:59 PM EST up reply actions
Zips doesn't do well with projecting for playing time
So focus on the rate stats.
by C. Wesley Baier on Jan 20, 2010 3:08 PM EST reply actions
Ah, gotcha
They’re pretty big on Zavada.
"I can just see it: Post 'Intervention? Why would I need an intervention?' And then he'd +1 himself."
And
I’d be happy to see Haren give us a 3.04 ERA and a 1.08 WHIP for the season.
"I can just see it: Post 'Intervention? Why would I need an intervention?' And then he'd +1 himself."
I don't think Haren's numbers will be too far off from that projection
I think Zips is still skeptical of Webby though for some reason. ZIPS just loves power pitchers.
by C. Wesley Baier on Jan 20, 2010 3:37 PM EST up reply actions
Yes, playing time is based on recent playing time and a few other factors (ZiPS looks for clear-cut start-relief conversions and translate the earlier performance accordingly, for example).
As I see it, figuring out who’s going to actually play isn’t really a job for a projection model. It will do fine in general, given that how the player has been used in the past is generally an important factor on how the player will be used in the future, but the model just doesn’t know specifics about, for example, Webb’s shoulder.
--
Dan Szymborski
dan@baseballprimer.com
by D.Szymborski on Jan 20, 2010 5:32 PM EST up reply actions
Thanks for answering all these questions about ZIPS Dan
Much Appreciation.
by C. Wesley Baier on Jan 20, 2010 5:44 PM EST up reply actions
+1
"I've had Bailey's out of a shoe, though."
by Dan Strittmatter on Jan 20, 2010 7:38 PM EST up reply actions
Indeed.
What's your name? Sandwich. What's your first name? TUNA.
by emilylovesthedbacks on Jan 20, 2010 11:40 PM EST up reply actions
Okay, awesome
Thanks a lot for the answers!
"I can just see it: Post 'Intervention? Why would I need an intervention?' And then he'd +1 himself."
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
