posted on Saturday, February 02, 2008 2:35 AM by Jim

Hey, PECOTA is alive

That headline works better if you say it aloud.

Nate Silver at Baseball Prospectus released his PECOTA projections, and here's a glance at how it sizes up the positional battles set to take place in Tucson:

MIDDLE INFIELD:
  • Danny Richar: .260/.315/.415
  • Juan Uribe: .249/.299/.419
  • Pablo Ozuna:  .269/.305/.341
  • Alexei Ramirez:  .295/.342/.452
  • Orlando Cabrera: .269/.323/.372
If Ramirez were to hit that projection, I'd probably wet myself.  There were six shortstops with an OPS higher than .794 -- their names are Hanley Ramirez, Carlos Guillen, Derek Jeter, Miguel Tejada, Troy Tulowitzki and Edgar Renteria.  Not bad company, especially when Ramirez is set to make $4.5 million over the next four years.

Alas, that line is merely a shot in the dark, since there's not nearly a large enough sample size to translate Cuban stats into major-league numbers.

I'd also take that line for Richar in a heartbeat.  PECOTA is middle of the road for Richar, as projections systems are giving him anywhere from a .685 OPS (ZiPS) to .788 (Bill James).

LEFT/CENTER FIELD:
  • Jerry Owens: .264/.321/.338
  • Carlos Quentin: .269/.349/.439
That should be a no-brainer, but regardless, I think Owens can outdo that line.  Projections systems aren't too kind to him, but he was pretty consistent after his second call-up in 2007.  Once again, The Bill James Handbook is the kindest to a Sox youngster (.340 OBP).

BACK OF THE ROTATION:
  • John Danks: 5.54 ERA, 21 HR in 113 1/3 IP
  • Gavin Floyd: 5.23 ERA, best improvement rate of the bunch.
  • Charlie Haeger: 6.29 ERA, 1.76 WHIP
  • Lance Broadway: 6.15 ERA, 57 BB, 63 K in 102 IP
  • Jack Egbert: 4.99 ERA, 6.38 K/9IP
Just like ZiPS, PECOTA thinks Egbert will be the Sox's third-best starter, not that far behind Mark Buehrle (4.77 ERA) and just ahead of Jose Contreras' 5.00.  Also compared to other projection systems, it loves Floyd and is lukewarm on Danks.

For comparison's sake, it projects a 4.88 ERA out of Brandon McCarthy, which is the second-highest of any projection system.  All of them peg McCarthy to stay under 5.00; only Marcel gives Danks an ERA under that bar (4.95).

Comments

# re: Hey, PECOTA is alive

Saturday, February 02, 2008 12:20 PM by dudeman
Damn, if Ramirez can put up those numbers the Sox should be a top five offense. I don't subscribe to BP... can someone tell me what Pierzynski projects to do? If he can improve over last year (in a platoon role) this offense can be really good. Unfortunately, if PECOTA is accurate about our starters, it may have to be.
What do the bullpen stats look like?

# re: Hey, PECOTA is alive

Saturday, February 02, 2008 2:58 PM by larry
one other fun stat about quentin: he's projected to lead the majors in HBP with 12. in only 365 plate appearances.

# re: Hey, PECOTA is alive

Saturday, February 02, 2008 3:43 PM by Jim Margalus
Pierzynski: .260/.304/.396. It's the harshest projection for him by far.

Bullpen stats -- it's about in line with the other projections systems for Jenks, Linebrink, Thornton and MacDougal. The two where it sticks out:

*Boone Logan: 1.37 WHIP, by far the lowest of any projection system, which is curious since it also pegs him for the lowest K rate.

*Ehren Wassermann: Significant increase in both walk and hit rates, although if it's factoring in a declining defense, I can see it.

# re: Hey, PECOTA is alive

Saturday, February 02, 2008 4:22 PM by Jim Margalus
Oh, that HBP rate seems to be selling Quentin short -- it's half his minor-league HBP rate. He averaged a plunking once ever 15.8 plate appearances over his last three seasons in the D-Backs' system.

He makes Luis Terrero or Aaron Rowand look like Sammy Sosa after the beaning.

# re: Hey, PECOTA is alive

Saturday, February 02, 2008 11:41 PM by Gregory Pratt
How, uh, accurate is PECOTA? In general? I always see their predictions but what is their track record, generally?

# re: Hey, PECOTA is alive

Saturday, February 02, 2008 11:42 PM by Gregory Pratt
Adding to that last comment: I think the Ramirez project is horseshit. That will not happen.

I am not either way inclined to trust or distrust PECOTA's system; just curious how often it turns out accurate or within a reasonable margin of error.

# re: Hey, PECOTA is alive

Sunday, February 03, 2008 12:37 AM by Jim Margalus
"Deadly accurate!" At least that's what it always says on their book covers.

Here are how the projection systems have stacked up the last couple of years:

2007: http://www.baseballprospectus.com/unfiltered/?p=564 (hitters)
http://www.baseballprospectus.com/unfiltered/?p=569 (pitchers)

2006:
http://www.baseballthinkfactory.org/files/newsstand/discussion/2006_projection_results/

PECOTA has the best track record of the bunch, it seems.