Yesterday we looked at The Bill James Handbook Mets hitter projections for next year, so go check 'em out if you missed it. Today we'll take a look at the pitcher projections.

Pitcher	Age	G	IP	H	HR	BB	SO	BR/9	ERA
Luis Ayala	31	78	76	79	8	18	49	12.0	3.94
Pedro Feliciano	32	86	50	46	4	22	45	12.8	3.90
Nelson Figueroa	35	18	33	35	5	14	20	13.9	4.90
Aaron Heilman	30	70	64	58	6	27	56	12.5	3.82
John Maine	28	25	145	135	18	62	122	12.5	3.98
Pedro Martinez	37	20	120	106	13	32	122	10.9	3.36
Mike Pelfrey	25	31	211	225	14	82	140	13.8	4.35
Oliver Perez	27	34	204	186	31	109	205	13.5	4.53
Duaner Sanchez	29	59	47	44	4	19	35	12.6	3.91
Johan Santana	30	34	230	189	25	58	234	9.8	3.01
Scott Schoeneweis	35	71	51	52	6	21	32	13.4	4.43
Joe Smith	25	85	62	58	4	32	54	13.9	4.09
Brian Stokes	29	46	67	76	8	25	47	14.0	4.90

If we took yesterday's hitter projections with the proverbial grain of salt, we'll have to take the pitcher projections with a heaping spoonful of same because pitcher projections are simply spottier than hitter projections. Pitchers tend to be a more volatile breed than hitters for a couple of big reasons I can think of. First, even the most workhorse-like of starting pitchers play far less than regular position players (by a factor of around five). This means that a crummy day for a pitcher is a lot more harmful to his overall season performance than a bad day for a hitter. It also means that if something isn't working right, a pitcher has to wait four days to give it another go in a real game. If a hitter's swing is off he can get right out there the next day and try to swing his way out of it.

Two of these starting pitchers are at least even money to be playing elsewhere in 2009, and any or all of the relievers could be gone, too. That's not to say that the entire relief corps will be replaced, just that there isn't a single name on this list that I'd be surprised to find pitching for someone other than the Mets next season.

• Take a long hard look at Luis Ayala's projection, because most of the other relievers are pretty similar. High-threes ERA, WHIP in the 1.25 range. Ayala's strikeout rate is not encouraging, though he's probably effective enough to pitch in the middle innings. He's not a closer, and he's not even really an eighth-inning guy.
• Pedro Feliciano is a LOOGY, so that ERA could be quite a bit lower than 3.90 if he is used fairly strictly against lefties.
• If Nelson Figueroa is on this team in April then Omar Minaya did another crappy job of bullpen assembly.
• I still like Aaron Heilman, even if nobody else does. He could very well have been injured this year, and if he can get the walks down he'll be very successful again. Trading him at his low water mark in value would be a huge mistake.
• I'll take those numbers from John Maine. Why is everyone clamoring for him to be a closer?
• Biggest surprise here: Pedro Martinez's arguably overoptimistic projection for 2009. 3.36 ERA and a 4:1 K:BB ratio? Even for just 120 innings I'll take those numbers.
• Mike Pelfrey's strikeouts seem high, here, as does the ERA. There'll be plenty of pressure on him this year because now we know he's actually a good pitcher. He flew under the radar a bit in the past because we only *thought* he was good.
• Oliver Perez can take those 109 walks to Milwaukee or St. Louis.
• Duaner Sanchez had some control issues this year and I think he ran out of gas a bit towards the end of the season. Given some time off this winter and the post-surgery season under his belt, he could definitely contribute to this bullpen next year.
• Johan Santana. There are no words. Except those. And those. And these.
• Scott Schoeneweis is still useful as a LOOGY, but the Mets already have one of those. No need to carry both of them again, is there?
• Joe Smith needs to keep the walks down and he'll be fine.
• BJH is not optimistic about Brian Stokes. I'll take the under on this one. I liked what I saw out of him this year and I think he can definitely be part of the solution in 2009.

I love stats. I acknowledge their flaws, and I certainly don't believe they tell you everything there is to know about baseball (does anyone actually?), but they really are just a lot of fun to pore over. Thanks to countless intrepid folks who have contributed the ways and means for distributing and disseminating baseball statistics, we have the tools available to learn far more about the game and its players than we ever have before.

One ridiculously useful site for baseball statistics (among other things) is FanGraphs, which is where all of the stats for this article (as well as the WPA graphs in my daily recaps) were extracted from. I wanted to take some time to look at the Mets' pitchers, their pitch velocities and pitch selection from 2008, so let's do just that.

Fastballs

The fastball is almost every pitcher's bread-and-butter. Better than 85% of pitchers threw fastballs at least half the time. Almost 11% threw their heater three-quarters of the time or more. Three guys threw 90% fastballs, though only one -- Tampa Bay's Grant Balfour (91.3%) did so over a significant number of innings. Eddie Kunz threw a higher percentage of fastballs than anyone else in baseball (92.3%), but he only threw a total of 55 pitches.

So folks throw a lot of fastballs. But how fast do they throw 'em? Here are the average fastball speeds for Mets pitchers this season.

Player	FBavg
Brian Stokes	95.0
Billy Wagner	94.5
Robert Parnell	94
Aaron Heilman	93.3
Edward Kunz	92.9
Mike Pelfrey	92.7
Jorge Sosa	92.2
John Maine	92.1
Johan Santana	91.2
Oliver Perez	91.2
Brandon Knight	91
Carlos Muniz	90.2
Duaner Sanchez	89.8
Joe Smith	89.4
Jonathon Niese	89.4
Claudio Vargas	89.3
Scott Schoeneweis	88.8
Tony Armas Jr.	88.8
Pedro Martinez	87.7
Nelson Figueroa	86.9
Pedro Feliciano	86.9
Ricardo Rincon	86.4
Matt Wise	84.2

It shouldn't be surprising that relievers dominate the top of this list. They come into a game for an inning at a time -- if that -- and can really air it out because they know they've only got 20 or so pitches before they hit the showers. Whereas starting pitchers have to pace themselves, relievers don't have to show such restraint. Anyone watching Mets games the last two months of the season has seen that Brian Stokes throws gas, and this chart corroborates that observation.

Mike Pelfrey had the fastest average heater among Mets starting pitchers at 92.7 MPH; Pedro Martinez had the slowest at 87.7. If that looks a little low, consider that his average fastball in limited action in 2007 was 86.2 MPH, so 87.7 is a clear improvement.

Of some concern is Johan Santana's average heater speed of 91.2. He was at 91.7 in 2007 with the Twins, and while a half-mile per hour doesn't seem like much now, it's definitely something worth keeping an eye on as he adds years to his ledger and mileage to his arm. His slider speed dropped (84.9 to 83.5) as did his changeup (81.9 to 80.0). The good news is that the disparity in speed between his fastball and changeup actually increased, though the velocity dip across the board is hardly encouraging. Whether that had anything to do with the deterioration in his strikeout rate (9.66 to 7.91 per nine innings) is not yet clear.

For those curious types, the fastest average fastball belonged to Joel Zumaya at 97.5 MPH. The slowest belonged to Tim Wakefield (72.9), though the slowest non-knuckleballer was Chad Bradford at 79.6. The slowest non-knuckler, non-sidearmer was Jamie Moyer for the fiftieth consecutive season at 81.2 MPH. The average big league fastball was thrown at 90.5 MPH.

Pitch Types

Let's move on from pitch speed and on to pitch selection. Here is the breakdown of pitch type for all Mets pitchers this season, sorted by highest fastball frequency.

Player	FB	SL	CB	CH	CT
Brian Stokes	69.30%		20.80%	9.80%
Billy Wagner	71.10%	28.50%		0.40%
Robert Parnell	75.30%	21.50%		3.20%
Aaron Heilman	64.00%	12.00%		24.00%
Edward Kunz	92.30%	7.70%
Mike Pelfrey	81.20%	12.80%	0.80%	5.10%
Jorge Sosa	50.30%	44.70%		5.10%
John Maine	70.50%	10.10%	1.00%	18.40%
Johan Santana	59.60%	11.70%		28.70%
Oliver Perez	69.20%	26.90%	1.30%	2.60%
Brandon Knight	62.00%	20.90%	13.20%	3.80%
Carlos Muniz	63.80%	31.80%		4.40%
Duaner Sanchez	46.40%	19.00%	3.70%	30.90%
Joe Smith	66.90%	31.10%		2.10%
Jonathon Niese	63.30%	0.40%	24.70%	9.40%	2.20%
Claudio Vargas	62.90%	23.20%		13.80%
Scott Schoeneweis	78.80%	20.80%		0.50%
Tony Armas Jr.	57.40%	36.80%	0.70%	5.10%
Pedro Martinez	56.10%	1.60%	16.00%	19.00%	7.30%
Nelson Figueroa	55.50%	19.70%	18.50%	6.30%
Pedro Feliciano	55.40%	38.40%	0.20%	6.00%
Ricardo Rincon	31.10%	60.70%		8.20%
Matt Wise	42.10%	0.80%		57.10%

(FB=fastball, SL=slider, CB=curveball, CH=changeup, CT=cutter)

One thing that jumps out immediately is that the Mets have a lot of pitchers who all sport basically the same arsenal: Fastball, slider, changeup. Jon Niese, Brian Stokes and Pedro Martinez preferred the curveball to the slider, and Niese and Martinez threw some cutters, but otherwise everyone threw the same three types of pitch. I'm a little surprised that the Mets lacked a single pitcher who throws a split-fingered fastball. Approximately one in eight (11.7%) big league pitchers threw at least 1% splitters in 2008; none threw them for the Mets. The splitter can wreak havoc on a pitcher's elbow, and who's to say that Rick Peterson didn't have a hand in steering the Mets away from that sort of injury risk. We know he was a mechanics freak, so it's not outside the realm of possibility that he advised the Mets to avoid pitchers who featured a splitter.

Quick Hits

• Jose Contreras led all qualified starters with splitters accounting for 24.6% of his pitches.
• Taylor Buchholz threw more curveballs than anyone else (34.9%).
• Tom Glavine threw 43.2% changeups, more than anyone else.
• The Cubs' Mike Wuertz threw 60.8% sliders, again, more than anyone else.
• Four pitchers were regular knuckleballers: Wakefield, Charlie Zink (Red Sox), Charlie Haeger (Padres) and R.A. Dickey (Rangers).
• Two others threw at least one knuckleball: Josh Banks (Padres) and Ryan Franklin (Cardinals), though the latter threw them so infrequently (.4%) that they were probably changeups or curveballs that were simply mis-categorized.
• The Angels' Darren O'Day threw 2.5% of his 671 pitches for pitch-outs.
• C.C. Sabathia threw more pitches overall than any other big leaguer: 3,814. NL Cy Young candidate Tim Lincecum was second at 3,682, and Sabathia's fellow free agent-to-be A.J. Burnett threw 3,650. Johan Santana was fifth with 3,598 pitches tossed.

Another day, another 900-some-odd feet of Carlos Delgado homerun. He's still not the MVP, but that doesn't mean he won't win the award and it certainly doesn't mean I wouldn't like to see him do so. He's been the third or fourth best offensive player on the Mets this season, which doesn't really smell like MVP to me. Nevertheless, Jimmy Rollins won the award last year with a good story despite turning in an inferior season to both Chase Utley and Ryan Howard, so it's certainly not out of the question that if Delgado stays hot for the next three weeks that he could take home the Mets' first MVP award. Mets probably should have won the last two, so it would only be fitting that the least deserving player of the three would take home the hardware.

The 2008 decline of Oliver Perez continued tonight. He did manage to walk fewer than three batters for only the second time in six starts, but he also allowed seven runs on eight hits in 3.1 innings. After a long stretch of terrific starts, he has drifted back into frustratingly waffling between decent and inept. That may seem unfair considering he only allowed a few runs in some of his more recent starts, but his walks and other peripheral stats paint a different picture, and I can't be the only one surprised by his recent downturn.

Once again, I really have to hand it to the Mets' bullpen. Nelson Figueroa came in and promptly imploded, allowing the last two of Perez's seven runs to score, but he and the rest of the relief corps allowed just one run over the game's final 5.1 innings. Brian Stokes continues to amaze, throwing 95 MPH heat and showing good control to boot. He and Luis Ayala may not be the 8th-9th inning duo we all envisioned in April, but they're getting the job done.

Sorry about the goofy question on the game form; everyone gets it wrong because there's no way to make both answers correct. My bad.

Big winners: Carlos Delgado, +33.7% WPA, Carlos Beltran, +28.7% WPA
Big losers: Oliver Perez, -40.7% WPA, Nelson Figueroa, -14.0% WPA
Teh aw3s0mest play: Beltran two-run blast, +31.3% WPA
Teh sux0rest play: Guzman wall-scraper, -18.1% WPA
Total pitcher WPA: -34.2%
Total batter WPA: +84.2%
GWRBI: Carlos Beltran

---

Game Thread Roll Call

Nice job by LOUtheMETSfan; his effort in the game thread embiggens us all.

Name	# of Posts
LOUtheMETSfan	109
Catsmeat Potter-Pirbright	77
BobbyV_Incognito	69
itsmetsforme	46
Prince	45
Omar21	36
JoshNY	34
pingel	27
gogomets	27
Simons	20
thriller	7
anonymous	6
IanB in MD	2
Reg Dunlop	1

You can't win them all, and four-game sweeps are especially tough, but when you take the first three against an awful team like the Pirates, it's a little disappointing not to come away with the last one. The Mets failed in the two areas they fail most often: hitting with runners in scoring position and relief pitching. The Mets had thirteen baserunners and only plated two of them. Comparatively, the Pirates had 15 baserunners and plated five.

Pitching-wise, John Maine was iffy. The two hits allowed look great, but his velocity was piss-poor and he walked four batters to just three strikeouts in five innings. He gets some bonus points for figuring out a way to get through those five innings with slop, but the Pirates are a crappy, crappy offensive team so it's hard to come away particularly impressed with his outing.

Brian Stokes turned back into a pumpkin, giving up a brutal two-run homer to Adam LaRoche that tied the game in the sixth. Pedro Feliciano and especially Duaner Sanchez were bad, combining to cough up three runs in the bottom of the eighth.

Damion Easley was just brutal at the plate, going 0-for-3 with a walk and accounting for five total outs if you consider his two GIDPs. The walk came in the ninth when the Mets actually managed to bring the tying run to the plate. Unfortunately, Argenis Reyes was that tying run, and he quickly became the third out.

No rest for the weary, as the Mets scoot home to take on the Braves on Tuesday.

Big winners: John Maine, +29.9% WPA, Nick Evans, +13.6% WPA
Big losers: Duaner Sanchez, -37.3% WPA, Brian Stokes, -23.7% WPA
Teh aw3s0mest play: Sanchez double-play in 5th, +21.1% WPA
Teh sux0rest play: LaRoche homerun, -25.4% WPA
Total pitcher WPA: -21.1%
Total batter WPA: -29.9%

---

Game Thread Roll Call

Nice job by IanB in MD; his effort in the game thread embiggens us all.

Name	# of Posts
IanB in MD	48
Greenpoint Ian	44
Catsmeat Potter-Pirbright	43
gogomets	27
elifriedman	17
kingcritical	13
anonymous	10
ams258	10
Reg Dunlop	2
JoshNY	2
mmxii	2

Joel Sherman breaks down the Mets' three options to remedy the bullpen situation, as suggested by Jerry Manuel in his post-game news conference yesterday. Those options are:

1. Have Eddie Kunz, who has all of three major league appearances, reprise his Double-A closing role. This is the most likely choice.

2. Keep Brian Stokes in the rotation and call upon either John Maine or Oliver Perez to serve as a multi-inning fireman. For now the Mets don't want to mess with Mike Pelfrey by requesting a rotation-to-pen change.

3. Summon top pitching prospect Jon Niese for the rotation and use Stokes plus either Maine or Perez to serve as the main late-inning relievers.

Sherman goes on to discuss why these are all potentially horrible ideas, particularly the ones that involve moving young starters to the bullpen. The Kunz move makes the most sense to me, at least in that he hasn't yet completely proven himself to be a failure like everyone else in the bullpen.

Billy Wagner expects to return on Monday, which is six whole games away. He is scheduled to make at least one rehab appearance with the Brooklyn Cyclones between now and then.

Jerry Manuel called Ramon Castro a pussy, in so many words.

At MetsGeek, Alex Nelson looks at the pitchers the Mets will face against the Nationals.

I don't want this to get political or anything, but this is funny as hell. And it's Olympic-themed!

Brian Stokes was far from terrible in his first big league appearance this season, striking out six and allowing four runs in 5.2 innings. He wasn't exactly spectacular, but he kept the Mets in the game long enough for them to rough up the Marlins' bullpen and ultimately walk away with an 8-6 win.

The legend of Dan Murphy continues to grow, as the Met rookie hit a two-run pinch-hit homerun in the seventh inning, adding a GWRBI to his list of early accomplishments.

Carlos Beltran went 3-for-3 with an intentional walk and Carlos Delgado hit another long homerun, as his bat continues to implore the Mets to pick up his 2009 team option.

The bullpen was shaky again, particularly Joe Smith, but Aaron Heilman nailed down his second save in as many games and the Mets moved a game-and-a-half ahead of the Marlins while remaining a game behind the Phillies.

Big winners: Carlos Beltran, +15.7% WPA, Dan Murphy, +14.1% WPA
Big losers: Brian Stokes, -12.3% WPA, Joe Smith, -10.1% WPA
Teh aw3s0mest play: Amezaga double-play in 8th, +15.3% WPA
Teh sux0rest play: Mike Jacobs bomb, -25.3% WPA
Total pitcher WPA: +1.5%
Total batter WPA: +48.5%

---

Game Thread Roll Call

Nice job by Simons; his effort in the game thread embiggens us all.

Name	# of Posts
Simons	50
Catsmeat Potter-Pirbright	47
tbach81	35
pingel	32
JoshNY	27
Greenpoint Ian	15
Endys Game	9
Shomov	7
anonymous	7
DoctorK16	6
LOUtheMETfan	5
mmxii	3
BlackOps	1
Prince	1

In light of Joba Chamberlain's recent trip to the disabled list, Tim Marchman looks at the curse that seems to have befallen young phenoms over the past twenty years.

Luis Castillo is close to returning. This is good news for Castillo and nobody else.

Ryan Church might not return at all this season. He has been taking batting practice and shagging flies, but the Mets have been tight-lipped about his upcoming medical evaluation and he has yet to be cleared for a return to the field.

Aaron Heilman still wants to start, and with the potential for two holes in the 2009 rotation who's to say he won't?

Jon Niese probably won't start for the Mets on Saturday. It'll probably be Brian Stokes or Claudio Vargas.

At The Hardball Times, Geoff Young looks back at the 1983 Lynchburg Mets, a team that went 96-43 and produced the likes of Dwight Gooden, Lenny Dykstra, Calvin Schiraldi, Ed Hearn and others.

Spring training games -- or, more specifically, player performances in spring training games -- are meaningless, and I don't know that I can stress this point enough. It's easy to get wrapped up in spring stats, to look at Raul Casanova and remark, "Wow, this guy is slugging .629! How the heck could Tampa Bay let this guy go?" The obvious answer, at least to anyone with at least an elementary understanding of sample sizes and random variation, is that Casanova, given a couple hundred plate-apps, would more closely replicate his pathetic career marks of .234/.301/.380 accrued over 1026 at-bats than the .314/.368/.629 that he has hit in an impossibly inconsequential 35 spring at-bats.

Casanova is just one of hundreds, if not thousands, of players who have ever had a spring training performance remarkably out of line with their career norms. "But he's in the best shape of his life", a random uninformed friend or possibly Dusty Baker might reply. Perhaps, but to make an ostensibly intelligent roster decision based on such a small number of plate appearances against arbitrary and often woeful competition is lunacy, especially when you consider how much money is riding on the success or failure of baseball teams in general and the Mets in particular.

I don't mean to single out Casanova, whom the Mets might have to promote to begin the season with Ramon Castro probably heading for the disabled list. Casanova might actually be a bad example because, given a blessing of good team health, he wouldn't be competing for a big league spot at all. A better example is the guy who is trying to distinguish himself from the pack in order to slide into a big league role that would really be available to him even if the rest of the team were healthy.

Two guys who *are* competing for an actual spot are Angel Pagan and Brady Clark, both of whom might make the club with Moises Alou out of action for the next month and the Mets' likely inclination to carry an extra position player in the season's early going. Neither is a terrific player, but the Mets don't really have any superior options immediately available to them. Or, at least options that wouldn't require giving up some non-existent minor league talent in return.

The Mets are also trying to finalize their Opening Day bullpen, a task which is still on the "To Do" pile and the results of which may hinge -- somewhat remarkably -- on the principals' performances in the last two exhibition games of the spring. That's right: forget everything anyone has done in their major or minor league careers up until this point. Those large sample sizes collected over a period of months or years? Not arbitrary enough. Instead, let's pick two random games -- say, the next two -- and base some fairly important baseball operations decisions on the decidedly random performances in said two games. It sounds completely ludicrous, but it's exactly what the Mets are doing. To wit:

The Mets don't know [what's going to happen], either. Two exhibition games now remain, and manager Willie Randolph has made it clear that he's willing -- almost eager -- to use both of them. His decisions need perspective; these games will provide it.

The article linked above was written by Anthony DiComo for Mets.com. I can't really tell if DiComo is editorializing something that Randolph has actually said -- that is, that he will base his decision(s) for the final bullpen spot(s) on the microcosmic results of the next two games -- or if this comment is DiComo's alone. I think we know that Randolph hasn't made up his mind yet, and he will really wait to see what happens tomorrow and Saturday before deciding to go one way or another. I don't think that I am exaggerating the truth when I say that Willie Randolph is going to construct the back end of the Mets' bullpen according to the events of two utterly meaningless spring training games. If he weren't, he would have already made his cuts and his promotions at this point. If we assume that he doesn't yet know which relievers will round out the bullpen, then it's pretty clear that everything that happened before today has not provided sufficient empirical evidence to sway him one way or another.

I'm not sure if there's a portmanteau rich enough to describe my feelings about this, but here goes: it's prepoculous. Preposterous and ridiculous. That's right: suck it, Carroll!

"His decisions need perspective; these games will provide it." How? How could they possibly provide any perspective other than that of sheer randomness? If Matt Wise goes out there and pitches like dogshit tomorrow while Mr. 6.46-career-ERA-mostly-as-a-reliever Brian Stokes pitches three scoreless innings, Stokes makes the Opening Day cut and Wise takes a long walk off a short pier? The level of asininity required to make a call like that is beyond comprehension, and yet it is almost certainly going to happen sometime in the next two days. Perhaps multiple times.

All we can do is throw up our hands in unbridled bewilderment and hope that the better players happen to actually perform better in these next two arbitrarily-selected talent showcases.