Huh. Zips loves Oakland! Score one more for the immortal Szym.

And computers don't lie...the Rays have arrived, yo!

As a Met fan I heartily approve of these projections...

Yes, the Reds are definitely correct in their belief that 2008 is so important that they must play Patterson and Hatteberg over Bruce and Votto to squeeze every last win out...

That Devil Rays projection is the only one in the list that I am very skeptical of.

But overall, this is great work.

Man, I must be out of touch. I've been telling anyone who listens that the Yankees are doomed this year.

I'm a big believer in projections, too.

Yikes, I feel sorry for fans of Florida and Baltimore.

Yikes, I feel sorry for fans of Florida and Baltimore.

I don't mean this to sound snarky, but I can't think of any Marlin fans who post here. I can sort of match the team allegiance to most of the regulars, but I'm drawing a blank on a Marlins fan. Surely, one of you knuckleheads must live and die with the two times World Series champs and recent Yankee killers, no?

Nate Silver may be a bright guy, but this treand of giving projection systems allcaps names of journeymen must stop. PECOTA, KUBIAK, CHONE, CAIRO. Begone!! It reminds me of cutesy names that they gave computer systems twenty years ago or so.

Huh. Zips loves Oakland! Score one more for the immortal Szym.


Don't say it out loud! If the A's go 69-93, I'm never going to hear the end of it. I still don't know why the A's are projecting so well - it's not like their projections are super-awesome.

Nate Silver may be a bright guy, but this treand of giving projection systems allcaps names of journeymen must stop. PECOTA, KUBIAK, CHONE, CAIRO. Begone!! It reminds me of cutesy names that they gave computer systems twenty years ago or so.

See, this is why ZiPS is awesome - it has a lowercase letter.

There used to be a guy named Black and Teal who posted here, but I haven't seen him lately. Wiki Gonzalez also lists Marlins Homer in DC.

PECOTA hates the Cardinals, and ZiPS doesn't hate them nearly as much as I thought it did. I mean, nobody is crazy about them this year, including me, but I'm going to be shocked if the Central plays out like that, no matter how respected PECOTA is.

That Devil Rays projection is the only one in the list that I am very skeptical of.

Looking at their lineup, depending on who gets PT... this *should* be the year they get over that .500 hump.

The 2007 Rays went 66-96, and deserved to go 66-96, what's changed?
1: The 2007 Rays had (aside from 2 SPs) a historically awful pitching staff. (subtract KAzmir an dShields and you have a 6.31 ERA- good for an ERA+ of 72 (there team ERA+ of 82 with those 2 guys included is still one of the 5 worst of the last 10 years)

2: The 2007 Rays had an extremely poor team defense

3: While the team on the whole had above average offensive production, catching and OF (every OF aside from Crawford) was very poor.

Now?
1: I'm a big believer in Garza, I also think Edwin Jackson can give 160-180ip of league average performance. I believe the BP will be much improved- that hideous 82 team ERA+ should be anywhere from 90 to 100 in 2008.

2: The worst offenders on dee are either not playing Dee or have been moved to a position they can better handle.
Bartlett at SS is a huge improvement over Harris, anyone replacing Upton at 2B is an improvement.

3: I believe pudgicito will be at least league average for a catcher in 2008...

My projection system differs only in the NL West. I ran about 500 simulations in my head and it came out like this:

ARI 104.4-57.5 (I know it doesn't add up perfectly to 162--rounding error)
COL 81.6-80.4
SD 80.9-81.1
SF 79.1-82.9 (surprising until you consider Lincecum averaged a 22-5 record in my sims)

Don't know what projection systems THT, DMB, Szym and others are using, but they are way off. Check them numbers, will ya?

Don't say it out loud! If the A's go 69-93, I'm never going to hear the end of it. I still don't know why the A's are projecting so well - it's not like their projections are super-awesome.

But if you're right, you can tell everyone to stick it! That could be fun.

Are the Dodgers still in the league?

Nate Silver may be a bright guy, but this treand of giving projection systems allcaps names of journeymen must stop. PECOTA, KUBIAK, CHONE, CAIRO. Begone!! It reminds me of cutesy names that they gave computer systems twenty years ago or so.

Oh right, mighty big words from a guy who calls himself Gary Geiger Counter.

Topic? -- All projection system agree who the worst team in the AL is.

I mean, nobody is crazy about them this year, including me, but I'm going to be shocked if the Central plays out like that, no matter how respected PECOTA is.

Oh, I don't know; PECOTA only puts them a fraction of a game behind Pittsburgh, and that's the least favorable projection model for the Cards. I don't think that's too unfair. I don't think I'd *pick* the Cards for last (unless Pujols goes down for the season), but I think HOU/PIT/STL is pretty much a tossup, with CHC/MIL the pretty clear top 2 and the Reds a solid third, clearly below the top 2 and clearly ahead of the bottom 3, which is what PECOTA reflects.

I'm a little surprised at how many of the projection systems put the Cards a solid third.

(surprising until you consider Lincecum averaged a 22-5 record in my sims)

so, basically you see him as better than 1968 Bob Gibson? 1978 Ron Guidry? 1985 Dwight Gooden? 1999-2000 Pedro?

22-5 with, what? 3 runs per game support?

so, basically you see him as better than 1968 Bob Gibson? 1978 Ron Guidry? 1985 Dwight Gooden? 1999-2000 Pedro?

1972 Steve Carlton?

Doc, where would I be without hypocrisy?

Surely, one of you knuckleheads must live and die with the two times World Series champs and recent Yankee killers, no?

You mean the Red Sox?

Surely, one of you knuckleheads must live and die with the two times World Series champs and recent Yankee killers, no?


You mean the Red Sox?

Well, sure, but you Sawx fans are like flies anymore. You don't have the mystery or legend that the rare and elusive Marlins fan has.

Division/league strength --

Looking at the sum of sims in Part II where all 6,000 projections get put into one great spreadsheet orgy . . .

AL: .507
NL: .497

By division:
ALE .520
NLW .505
NLE .502
ALC .496
ALW .492
NLC .486

Love the NL West pie chart! Could get hairy out there.

so, basically you see him as better than 1968 Bob Gibson? 1978 Ron Guidry? 1985 Dwight Gooden? 1999-2000 Pedro?

22-5 with, what? 3 runs per game support?

Like I said, somehow my projection system likes him quite a bit. Oh by the way, Upton and Young were 1-2 in the MVP voting too.

CBW, can you do something about getting Petit to be the #5 starter? It would really help my fantasy team. Thanks. Also, if you could trade Haren to the Mets for Pelfrey that would be great, too.

We won't even need a 5th starter w/ Webb and Haren and RJ taking 40 starts apiece.

Lassus--sleeper picks?

Chad Bradford and Jay Marshall....might want to look into Byung Hyun Kim as well.

Off to TEP....

Oh by the way, Upton and Young were 1-2 in the MVP voting too.

So a breakthrough season for BJ Upton and a big bounceback for Michael Young. Should be fun to watch.

I've been telling anyone who listens that the Yankees are doomed this year.

I mentioned it in the Yankees comment, but so much of the Yankees' success is going to be based on how Hughes/Chamberlain/Kennedy pitch. I was surprised at how bullish their projections are in almost every system. 450-500 league average innings with the Yankee lineup is a good way to win a lot of games, but there's reason to be skeptical that they'll get anything close to that if you look at the history of some pretty good pitchers and how they did before the age of 23.

Before I forget, thanks for all that effort, SG's computer! (And SG, too.)

Nate Silver may be a bright guy, but this treand of giving projection systems allcaps names of journeymen must stop. PECOTA, KUBIAK, CHONE, CAIRO. Begone!! It reminds me of cutesy names that they gave computer systems twenty years ago or so.

As we say at work, the better-sounding the acronym, the worse the product it represents.

Surely, one of you knuckleheads must live and die with the two times World Series champs and recent Yankee killers, no?

You mean the Red Sox?

Red Sox are seven-time World Series champs. Only two teams have won more.

Red Sox are seven-time World Series champs. Only two teams have won more.

Ah, but the Athletics franchise has 9!

As we say at work, the better-sounding the acronym, the worse the product it represents.

I call this the PATRIOT Act rule.

Naming these sorts of things can be fun, so I don't mind the naming stuff.

Mine was called "Cayce" if anybody cares.

Naming these sorts of things can be fun, so I don't mind the naming stuff.

I was about to say, that's half the fun of going through all the work of making a projection system.

Red Sox are seven-time World Series champs. Only two teams have won more.

Let's see -- the Cardinals and the A's, obviously. I could have sworn there was a third team, but maybe not.

Edit: Or, what Shooty said.

Wow - AL East and NL West both projecting to have all teams but one above .500

Pretty cool if it happens, though it sure is a rough assignment to be a Blue Jays fan.

Villageidiom: Yankees (26), Cardinals (10), A's (9). So I count 3. :)

Let's see -- the Cardinals and the A's, obviously. I could have sworn there was a third team, but maybe not.

Jeeze, that's like trying to name the five boroughs and leaving out Manhattan.

Numerical Estimate of Individual Future Input.

I'm too lazy to come up with the formula, but there's the acronym.

Jeeze, that's like trying to name the five boroughs and leaving out Manhattan.

You don't talk to the people in Brooklyn enough.

Numerical Estimate of Individual Future Input.

I'm too lazy to come up with the formula, but there's the acronym.

0*0/0=0?

0*0/0=0?

Oh yea, he has zero value but his value is closer to infinity if you consider the intangibles and division by zero

And count me as surprised as how well the Mets project. For a team with no depth, thats surprisingly good mean projections.

Oh yea, he has zero value but his value is closer to infinity if you consider the intangibles and division by zero

I think you just invented a perpetual motion machine.

First, I appreciate the effort. Always an interesting read.

Second, the following is a foolish remark if not worthy of being termed ridiculous:

Their biggest weakness may be depth in the starting rotation, but you can say that about everyone except Milwaukee probably.

Milwaukee has pitching depth? Milwaukee? The Milwaukee Brewers? The team in the National League??

Milwaukee has pitching depth like Keanu Reeves has talent. It's kind of there but Chr*st you gotta look hard to find it.

Pitching depth?? Milwaukee???? Honest??????

As you can see this is news to me. I didn't think having a bunch of guys hanging around the clubhouse counted as pitching depth.

If we use the list in part 2 to play over-under, which teams look like sure overs?

I am picking Atlanta, Cincinnati and Washington.
Under on the two NY teams and Cardinals.

Milwaukee has pitching depth? Milwaukee? The Milwaukee Brewers? The team in the National League??

To be fair, I didn't explicitly say it was good depth.

I didn't know that Capuano is likely out for the year when I wrote that comment and the Brewers' MLB.com depth chart shows 8 guys as possible rotation candidates. But I will defer to you on this as you clearly know more about your team than I do. It's tough to be up on all 30 teams for something like this.

And count me as surprised as how well the Mets project. For a team with no depth, thats surprisingly good mean projections.

Diamond Mind consistently under-projects injury time, which will make teams with no depth project better than they should (and vice versa).

SG:

Understood. But if you saw the names Sheets (Senor Fragile), Suppan (Senor S*ck), Bush (Senor Stupid), and Vargas (Senor Lento)I struggle to equate that to "depth".

Apologies if my comments were overly harsh.

Diamond Mind consistently under-projects injury time, which will make teams with no depth project better than they should (and vice versa).

This is generally true, but I did build depth charts that account for bench and spot starter time so these simulations do account for team depth at least somewhat. No position player is projected to play more than 90% of the time, and I have spot starters behind pitchers in varying percentages. In the case of the Mets I limited Pedro to 170 innings and El Duque to 120.

I don't account for catastrophic injury, but I don't know that any forecast really should.

Apologies if my comments were overly harsh.

No offense taken. Compared to the crap I get when I criticize Derek Jeter's defense on our blog that was nothing.

As we say at work, the better-sounding the acronym, the worse the product it represents.


Why do you hate Pakistan?

SG,

A question. Did you account for the injuries to Lackey and Escobar in your sims?

Thanks. And thanks for the work.

But if you saw the names Sheets (Senor Fragile), Suppan (Senor S*ck), Bush (Senor Stupid), and Vargas (Senor Lento)I struggle to equate that to "depth".


How serious is Gallardo's injury? Is Parra not a factor this year?

Gallardo is already throwing. He should miss a few starts in the rotation and make his first major league start a few weeks into April. Parra's status hasn't been announced yet. He has looked fantastic in the spring (last I heard), and is a candidate to make the Brewers rotation.

The Brewer's preference would probably be that Bush and Vargas look great in their last few ST starts and make the rotation with Parra sent to the minors. I'm hopeful that the Brewers feel that the NL Central race will be a tight one, and that they should have their most talented starters throw the most innings, and forget about service time and options, etc.

Vargas is out of options, so is almost guaranteed to make the Brewers, either as back end starter or long man out of the bullpen. Bush still has an option, but apparently would have to be placed on revocable waivers to be sent down (I believe that the gentleman's agreement is not to block these moves, we'll see). Villenueva could go either way, depending on his last few ST appearances.

Basically, Sheets and Suppan are locks and Vargas is almost a lock. The bullpen is, in some order, Gagne, Torres, Riske, Mota, Shouse, Turnbow. Stetter, McClung and Choate are options to fill out the rest of the pen. So it's likely that Parra, Villenueva and Bush are fighting for the 5th rotation spot once Gallardo comes back and the other two would be sent to AAA.

A question. Did you account for the injuries to Lackey and Escobar in your sims?

I had Lackey missing 15% of his starts and Escobar missing 20%. Not sure if that's enough or not.

Gallardo will come back but he's young. Regression has to be expected.

Parra has been the best starter in camp and has better stuff than any of the vets but Ned has made it clear he prefers experience.

Apparently experience is being bad takes precedence over the chance of being good.

(Insert Yosemite Sam soundtrack)

Man, I must be out of touch. I've been telling anyone who listens that the Yankees are doomed this year.

I'm a big believer in projections, too.

I'm with you. No playoffs for Yankees. If by doomed you mean that. 80 wins without A Rod, 87-88 with.

I'm a little surprised at how many of the projection systems put the Cards a solid third.

Help, I do not know how most season projections systems work. Are the Cards getting a bonus from their 83 win World Championship season of a couple of years ago? Is there an input in these projections called "magic"?

To be fair, I didn't explicitly say it was good depth.

I didn't know that Capuano is likely out for the year when I wrote that comment and the Brewers' MLB.com depth chart shows 8 guys as possible rotation candidates. But I will defer to you on this as you clearly know more about your team than I do. It's tough to be up on all 30 teams for something like this.

I think Milwaukee does indeed have starting pitching depth. Consider Milwaukee's 5th starter performance last year. I don't have it in front of me, but it was in the top 5-6 in terms of ERA. Is there a better measure of pitching depth? The team does have depth.

I think the Cards are getting a bonus of not being Houston or Pittsburgh. As for them over Cincy...I dunno, but it's not crazy.

I took the combined projections (6000 runs) and used log5 to backsolve for each team's schedule-neutral win percentage.

adj w% Team
0.598 NYA08
0.582 BOS08
0.570 NYN08
0.565 DET08
0.556 CLE08
0.550 LAA08
0.542 TOR08
0.525 TB08
0.524 ATL08
0.524 CHN08
0.518 PHI08
0.516 LAD08
0.510 SD08
0.507 ARI08
0.506 MIL08
0.504 OAK08
0.500 COL08
0.487 SEA08
0.478 MIN08
0.471 TEX08
0.470 CHA08
0.468 STL08
0.466 KC08
0.460 CIN08
0.447 HOU08
0.446 SF08
0.440 BAL08
0.430 WAS08
0.420 PIT08
0.416 FLA08

7 of the 8 best teams in baseball are in the American League, and 4 of them are in the AL East.

The Cubs have the easiest schedule and the Orioles have (by far) the toughest schedule.

The Cubs have the easiest schedule and the Orioles have (by far) the toughest schedule.

If they have the 10th best record with a neutral schedule, they could have the 3rd or 4th best this year.

I was surprised about the gap ZiPS had between them and Milwaukee. I just don't see it being that large.

Parra's status hasn't been announced yet. He has looked fantastic in the spring (last I heard), and is a candidate to make the Brewers rotation.

Courtesy of the JSonline Brewers blog today.
Parra, who is fighting for a spot in the starting rotation once the season starts, allowed five runs on five hits in three innings. He walked four, didn't record a strikeout and allowed a home run in the third to Iannetta. After his outing, Parra slumped in his clubhouse folding chair with a dejected expression on his face and said he was simply "thinking too much" and "trying too hard."

"There's really no excuse, man," Parra said. "There's no excuse."

I just watched his outing, it was not too good. I still think he's the best 5th man.

Villageidiom: Yankees (26), Cardinals (10), A's (9). So I count 3. :)

I was counting the Yankees as a conglomerate, not a team. ;-)

I was debating b/w Parra and Volquez for the last pick of a 6x6 mixed (al+nl) league; needed a sleeper as my 6th starter. Went with Volquez.

Milwaukee has pitching depth?

Well, sorta. ZiPS likes their rotation depth quite a bit -- I think it was 5 above-average starters and another three guys not far below. But yes, that included Capuano, Bush, Sheets, Suppan so take that as you will.

I think Milwaukee does indeed have starting pitching depth. Consider Milwaukee's 5th starter performance last year. I don't have it in front of me, but it was in the top 5-6 in terms of ERA.

A lot of that is going to depend on how you define 5th starter. Milwaukee avoided the disastrous starters -- they had no games started last years by guys with ERA+ of 85 or below. However, they had a full 89 starts from 3 guys each with an 88 ERA+. So while having the fewest starts <85, they might have been second to Florida in number of starts by guys <90.

But still, we worked through this the other day -- last year the average was an average of 41 starts per team from a guy with an ERA+ of 85 or worse. I don't think we ever tracked down the ERA+ of that group, but I'd wager 75 or worse. That's an ERA of 6. I don't remember if we tracked down the median and there are a few major outliers -- Tampa, Florida, Texas if I remember right.

This means a couple things:

1. Any team that can avoid disaster starts either through an amazingly healthy starting 5 or non-sucky depth has a major advantage.

2. Nobody should ever use the argument "this disappointing team got 40 starts last year from S, T, U, V, W, X, Y and Z who all sucked; that won't happen again this year."

7 of the 8 best teams in baseball are in the American League, and 4 of them are in the AL East.

Anyone in the Central division want to trade divisions with the Jays?

I don't account for catastrophic injury, but I don't know that any forecast really should.

Oh, I disagree. At least if the goal is forecasting/projection. (If the goal is "who's the best team on paper right now" or "how good a job did your GM do in assembling talent" then maybe not)

Ideally a projection would use a ton of randomly generated playing time scenarios as well as things like "young player breakouts" and "cliff-divers". It would also be cool to make in-season playing time more dependent on previous in-season performance -- i.e. the computer currently will limit the guy hitting 350 through May to his assigned playing time while in real life, he's not coming out of the lineup until he craters. Then run the sims million of times. That would give us a better picture of "reality".

Now (1) obviously real-life gets run only once, so it would be rather hard to judge how well your sim reflects "reality" and (2) at best it seems to me that you're only going to bunch the mean projections more tightly while increasing the variance. Also (3) it would be a monumental piece of work with a pretty tiny payoff.

But it will make you look less like a fool when the Giants win the Series because you'll be able to point out that they did just that in the one simulated season where they were 100% healthy and every other team's entire 25-man roster tested positive.

Seriously, a lot of the team "flukes" are probably due to things like health, monumental breakouts, the fact that Billy Grabarkewitz turned in a season of 640 PA (with 95 BB!!) and a 134 OPS+ in a career in which he never had more than 200 AB or an OPS+ over 100 again in a season again.

I was debating b/w Parra and Volquez for the last pick of a 6x6 mixed (al+nl) league; needed a sleeper as my 6th starter. Went with Volquez.

Parra is the better of the two.

Vargas released today. Boy, was I wrong. I'm surprised that they couldn't get anything for him. I'm sure we'll get some details later on today.

Help, I do not know how most season projections systems work. Are the Cards getting a bonus from their 83 win World Championship season of a couple of years ago? Is there an input in these projections called "magic"?

it may have to do with something like the team isn't quite as bad as the naysayers are predicting, above average offense (likely top 5 in the nl) and slightly below average rotation with an above average bullpen. Sounds like a .500 team to me.