2012 Tampa Bay PECOTA Projections

by Erik Hahmann on Feb 9, 2012 10:00 AM EST in Features & Analysis

Yesterday, Baseball Prospectus released it's 2012 PECOTA player projections. I'm not going to comment on the quality of them one way or another -- there's a discussion at The Book Blog, if interested -- but they are yearly among the most accurate projection systems out there.

The Rays' PECOTA projections are below the jump and edited a bit as to not cram all the information into an unreadable table. Also, I only included the players with MLB experience currently on the roster.

There may be two stats in the tables that you're unfamiliar with if you're not fluent in Prospectus speak: TAv and WARP. We tend to be a more FanGraphs-heavy site, so, if it helps, think of TAv (True Average) as the BP equivalent of wOBA and WARP (Wins Above Replacement Player) as the equivalent of WAR.

For reference, here's a chart of the average TAv from each position (1-11) in 2011:

POS	TAV	PA	GMS	VORP
1	.1476	5923	1297	140.35
2	.2519	19140	2429	676.61
3	.2839	20709	2429	686.93
4	.2575	20831	2429	484.88
5	.2567	20233	2429	554.69
6	.2536	20715	2429	684.73
7	.2628	20763	2429	484.34
8	.2664	21247	2429	758.89
9	.2804	20780	2429	769.66
10	.2787	9758	1134	274.12
11	.2266	5100	1847	-30.1

Now, on to the projections

Hitters:

NAME	HR	RBI	SB	AVG	OBP	SLG	TAv	WARP
B.J. Upton	17	70	41	0.245	0.332	0.398	0.272	2.8
Ben Zobrist	18	71	18	0.249	0.344	0.413	0.282	3
Desmond Jennings	13	66	36	0.257	0.334	0.392	0.273	2.3
Evan Longoria	28	87	8	0.262	0.347	0.482	0.303	5.6
Carlos Pena	32	78	3	0.22	0.346	0.458	0.294	1.8
Matt Joyce	22	76	11	0.25	0.341	0.449	0.29	2.7
Sean Rodriguez	15	53	12	0.229	0.31	0.394	0.263	1.6
Luke Scott	19	58	2	0.238	0.321	0.432	0.279	1.5
Jose Molina	4	29	2	0.221	0.266	0.305	0.217	-0.4
Jeff Keppinger	3	29	1	0.266	0.312	0.355	0.252	0.8
Reid Brignac	5	27	3	0.229	0.272	0.335	0.228	0.3
Sam Fuld	1	9	5	0.239	0.32	0.328	0.25	0.4

The first thing I notice is the seemingly un-optimistic outlook for Longoria's slash line. Each of those totals would be below his career average. His BABIP was incredibly low last season and he battled injuries, yet he still managed to hit 31 homers. If his luck improves just a little and he manages to stay healthy, these projections should prove to be on the low side. On the bright side, his 5.6 WARP projection ranks second in baseball to only Albert Pujols, and his TAv is 18th. So even with a seemingly pessimistic projection, he's still quite good.

Unsurprisingly, Ben Zobrist has the second highest WARP on the team. His 3.0 total places him with Dan Uggla and Alex Avila in the projections. And any Rays fan would gladly take the 32 projected home runs for Carlos Pena in a heartbeat.

Pitchers:

NAME	W	L	IP	BB9	SO9	ERA	WARP
David Price	14.3	12.3	214.3333	2.9	8.1	3.17	4.5
James Shields	13.1	12.5	201.6667	1.9	7.4	3.47	3.4
Jeremy Hellickson	10.2	9.8	156	2.8	7.2	3.36	3
Wade Davis	8.9	10.2	144	3.3	6.4	4.03	1.5
Matt Moore	8.9	9	138	4.6	10.4	3.56	2.3
Jeff Niemann	7.4	7	125.3333	3	6.6	3.83	1.3
Joel Peralta	3.2	1.6	61.6667	2.4	7.8	3.27	0.7
Alex Cobb	3.6	4.6	60	3.5	6.2	4.41	0.3
Burke Badenhop	2.8	1.4	57.3333	3	5.4	4.11
Alexander Torres	3.3	3.5	54.6667	5.3	7.9	4.45
Kyle Farnsworth	2.7	1.4	52.6667	2.9	8.3	3.67	0.3
Jake McGee	2.7	1.2	52.6667	3.8	8.6	3.84	0.2
Fernando Rodney	2.4	1.1	48.3333	4.9	8.1	4.24	-0.1
J.P. Howell	2.3	1.1	44	3.6	9	3.52	0.3
Brandon Gomes	2	1	39.6667	3.6	8.4	3.88	0.1
Dane De La Rosa	1.1	0.4	29.6667	3.4	8.3	3.74	0.2
Josh Lueke	1.3	0.6	26.3333	3.2	8	3.75	0.1
Cesar Ramos	1	0.4	22	3.6	4.8	5.06	-0.3

David Price's strikeout and walk rates go in opposite directions from his 2011 campaign, while James Shields' walks drop along with his strikeouts. The strikeout and ERA projections for Moore would be welcomed, but I don't think he'll have as much of a control issue as this states. He hasn't come close to walking that many people since High-A ball. There's been some concern over the reliever projections across the board, so take those for what you will.

Nevertheless, trying to predict anything in baseball is a rough job and I applaud those who try. The full PECOTA spreadsheet can be found here for subscribers. What did you all think of the projections?

Thanks to R.J. for helping out with the tables.

Tweet Comment 31 comments | 0 recs |

Do you like this story?

Comments

40 WARP total

Not sure if the PAs add up at the team level at this stage, but if so, that’s pretty darn solid. BPro uses a relatively high rep level, so this is at least a 90-win projection.

@Sky_Kalkman

by Sky Kalkman on Feb 9, 2012 10:07 AM EST reply actions

Why are they so low on Molina's defense?

by mr. maniac on Feb 9, 2012 10:13 AM EST reply actions

Maybe they think it's B. Molina.

by Peter Piontek on Feb 9, 2012 10:17 AM EST up reply actions

Great article from Matt Swartz today analyzing how projection systems fared last season.

Link.

In the end, there’s little difference between all the top projection systems. But I think you can argue that ZiPS and PECOTA were the best for offense (with Oliver and Steamer not far behind).

And in a somewhat of a surprise (to me, at least), it looks like Steamer and CAIRO were the way to go for pitching projections. Or just use a regressed version of SIERA.

I love Casey Fossum. Now try and take me seriously.
Follow @steveslow

by Steve Slowinski on Feb 9, 2012 10:38 AM EST reply actions

It’s funny how close these all are.

In the last 4 years, of which there has been little substantive change to the mean projections, ZiPS has had:

A really good hitting, really good pitching year
A really good hitting, really mediocre pitching year
A really mediocre hitting, really good pitching year
A really mediocre hitting, really mediocre pitching year

--
Dan Szymborski
Dan's Stuff is on: ESPN, BTF, Twitter

by D.Szymborski on Feb 9, 2012 10:53 AM EST up reply actions

Cool...I was wondering how this stacked up with past results.

It really does seem like the top projections are all crazy close, and the answer to “who’s better?” on a yearly basis comes down to how the bounces go during the season.

Although the FB velocity bit about the Steamer projections is fascinating.

I love Casey Fossum. Now try and take me seriously.
Follow @steveslow

by Steve Slowinski on Feb 9, 2012 11:07 AM EST up reply actions

I have a follow up story going up shortly

Consider

Follow @jasoncollette

by Jason Collette on Feb 9, 2012 11:24 AM EST up reply actions

Those are the various projections for Longoria

Follow @jasoncollette

by Jason Collette on Feb 9, 2012 11:25 AM EST up reply actions

They had Ben Zobrist as a -3 2B and a +2 RF

I think you can add another win on to his WARP at a minimum. I’ve also converted these to wOBA if anyone is interested:

Row Labels HR BA OBP SLG wOBA WARP
Evan Longoria 28 0.262 0.347 0.482 0.355 5.6
Carlos Pena 32 0.220 0.346 0.458 0.346 1.8
Matt Joyce 22 0.250 0.341 0.449 0.343 2.7
Ben Zobrist 18 0.249 0.344 0.413 0.335 3
Luke Scott 19 0.238 0.321 0.432 0.328 1.5
B.J. Upton 17 0.245 0.332 0.398 0.323 2.8
Desmond Jennings 13 0.257 0.334 0.392 0.319 2.3
Juan Miranda 8 0.230 0.320 0.401 0.312 1
Robinson Chirinos 3 0.245 0.321 0.401 0.306 0.6
Sean Rodriguez 15 0.229 0.310 0.394 0.299 1.6
Sam Fuld 1 0.239 0.320 0.328 0.299 0.4
Jeff Keppinger 3 0.266 0.312 0.355 0.299 0.8
Brandon Guyer 4 0.254 0.299 0.398 0.297 0.3
Stephen Vogt 6 0.249 0.284 0.388 0.293 0.7
Elliot Johnson 6 0.235 0.293 0.366 0.291 0.9
Jose Lobaton 2 0.230 0.305 0.348 0.289 0.3
Matt Mangini 5 0.242 0.286 0.362 0.283 0.6
Leslie Anderson 5 0.246 0.281 0.368 0.280 0
Reid Brignac 5 0.229 0.272 0.335 0.269 0.3
Derek Dietrich 8 0.210 0.251 0.362 0.260 0.2
Tim Beckham 3 0.221 0.269 0.319 0.259 0.1
Nevin Ashley 4 0.208 0.282 0.303 0.256 0.2
Jose Molina 4 0.221 0.266 0.305 0.255 -0.4
Hak-Ju Lee 1 0.226 0.275 0.297 0.255 -0.3
Jeff Malm 5 0.182 0.251 0.287 0.236 -1.9
Drew Vettleson 3 0.179 0.230 0.260 0.219 -2.6
Ryan Brett 0 0.192 0.238 0.249 0.216 -1.8
Yoel Araujo 4 0.159 0.220 0.251 0.211 -2.5
Joshua Sale 1 0.168 0.214 0.225 0.197 -3.2
Jake Hager 1 0.180 0.206 0.238 0.195 -2.4

I got away from the one thing that kept me on the straight and narrow, and that was my relationship with the Lord

by Sandy Kazmir on Feb 9, 2012 10:48 AM EST reply actions

The Zobrist quibble aside, nothing else seemed particularly glaring other than they seem to have lower expectations across the board than I (or Zips) expect

If they have a higher replacement level then it could stem from that, otherwise these seem fine as projections. I’d still average this with Zips, Marcel, and Oliver and you’ll probably get closer to the truth than any one system.

I got away from the one thing that kept me on the straight and narrow, and that was my relationship with the Lord

by Sandy Kazmir on Feb 9, 2012 10:50 AM EST up reply actions

It seems consistent (besides for defense), which makes it fine in my eyes.

by mr. maniac on Feb 9, 2012 10:55 AM EST up reply actions

using all of them is definitely my preference.

by rglass44 on Feb 9, 2012 11:03 AM EST up reply actions

Laughed at this part for the obvious reason:

WARP (Wins Above Replacement Player) as the equivalent of WAR.

Also worth noting: BP uses a higher replacement level than Fangraphs (and B-Ref, I believe), hence why the WARP values may seem depressed if you’re comparing it to another value metric.

by R.J. Anderson on Feb 9, 2012 11:27 AM EST reply actions

Was wondering about that

Having only 2 guys in the 3+ WARP territory(and Zob barely got there) was a little sobering. Still super excited about this team

by BJ the Bossman on Feb 9, 2012 11:37 AM EST up reply actions

Here's the breakdown of batters with 200-plus plate appearances from last year:

FG doubles up BP in four-to-five win players, and also has more five-plus win players.

by R.J. Anderson on Feb 9, 2012 11:54 AM EST up reply actions

Add another WARP on right off the bat for his defense

He’s much closer to +11 at 2B than -3

I got away from the one thing that kept me on the straight and narrow, and that was my relationship with the Lord

by Sandy Kazmir on Feb 9, 2012 11:56 AM EST up reply actions

So combining with R33J's chart he'd be basically top-15% in baseball

I got away from the one thing that kept me on the straight and narrow, and that was my relationship with the Lord

by Sandy Kazmir on Feb 9, 2012 11:58 AM EST up reply actions

Yeah.

Like I said, I’m not sure where they are getting the defensive numbers.

by mr. maniac on Feb 9, 2012 11:58 AM EST up reply actions

I don't know how the defensive WARP part is computed.

To answer the Molina question, I do not believe his framing numbers have been take into consideration yet, although that is something they are working on.

by R.J. Anderson on Feb 9, 2012 11:59 AM EST up reply actions

I am guessing it has to do with FRAA

Which Colin wrote up here: http://www.baseballprospectus.com/article.php?articleid=11589

I’m bearish on defensive metrics nowadays thanks to Colin’s other work, so I couldn’t tell you which is best. My guess: if you average them and ignore anything after a decimal point you’ll usually have a decent idea, but that’s just a guess.

by R.J. Anderson on Feb 9, 2012 12:03 PM EST up reply actions

Do that then add +-1.5 for error bars and I think that's where I'm at

Basically you’ve got plus guys (3<x), average guys (-3<x<3), and bad (x<-3). When you color it in this way it’s pretty easy to combine with visual confirmation provided you’ve seen 20 – 30 full games.

I got away from the one thing that kept me on the straight and narrow, and that was my relationship with the Lord

by Sandy Kazmir on Feb 9, 2012 12:14 PM EST up reply actions