1. What pitches he throws, and how often

2. How hard each pitch is thrown on average

3. Fastball velocity by month

The 3rd place system, Oliver, is available to THT subscribers.  Why would anyone pay money for projections that were less accurate last year than what you can get on this site for free?  Well, I actually have an answer for you, and I do recomment buying them. 

One is the player comments.  They were not written by a small staff of writers who may or may not even watch all the teams they were assigned, they were written by 30 bloggers who follow the teams.  And they will be updated as the 2010 season unfolds.

Another reason is the number of players projected - I've only got 2500-3000 players who at least played in AA last year.  Oliver has players from the low minors, and even uses college stats in the projections.

I think the difference here is Fangraphs is using BIS data, which is, last I checked, compiled by scorers watching the game where MLBAM uses an automated pitch classification algorythm.  The knuckleball apparently is as hard to measure as it is to hit, or to catch.  The formula probably leans heavy on the velocity and doesn't want to call what Wakefield throws a fastball.

Ervin's results did show improvement as the season went on, with a 7.81 ERA in the first half, and a 3.90 ERA in the second. (Splits from baseball reference) His velocity seems like a decent gauge for both his health and effectiveness, so how does his velocity look by month?

I've been spending the past week getting the MLBAM pitch f/x data on my computer.  I may be the first person on earth to do so primarily using microsoft tools (excel & access 2007).  I haven't had time for much else, but a few people have asked about projection spreadsheets, I'll try to get those up later today.

Anyway, once I got all the data a query to look at fastball velocity by month was a snap, and Ervin's shows strong improvement as the year went on:

May 91.0 June 91.2 July 92.4 August 92.9 September 93.3

Even at year's end, he still wasn't quite back to where he was in 2008, so unless he's still gaining strength and doesn't get hurt again, it's probably best not to expect a dominant season like 2008.  But I think he should be able to beat the 4.40 computer projection, that doesn't know anything (yet) about pitch f/x or recovery from injuries.  If I were to bet I'd place him around 3.90 for 2010.

A-Rod is in line for some incredible numbers.  He's got 8 years on his contract so the opportunity will be there.  If he shows up and stays reasonably healthy he's going to break the homerun and RBI records.  The only way he doesn't is if he walks away from the money (yeah, right, like Boras would ever allow that) or suffers a Griffey-like rash of injuries.

So this new set of projections looks only at the starting lineup of every team, the top 5 pitchers (rated on runs above average in a neutral environment for 200 innings each), and the top 2 relievers (70 innings each).

The biggest differences:

1. More wins for the Angels.  I'm not unbiased here, not at all.  Angels rule and that is the number one thing I believe in.  I like these better, though still a game behind a very solid Ranger team.  Angels have been beating my projections anyway for about 6 years now, so if forced to choose I'd rather go with the one that rates them higher.  This is not a case of heads coming up 6 times in a row where you'd still bet 50/50 on the next flip.  I believe that my stats are not capturing everything the Angels do well (neither are most other people's stats), and maybe one day with more research we'll figure out what Scioscia and his boys did so well.

2. Royals lower.  76 wins didn't feel right.  Any team that starts Yuniesky Betencourt and Jason Kendall on purpose can't be expected to win too many.

3. Reds lower.  Not sure about this, I like the Reds' young talent but in any case they are probably not contenders.  I left their top pitcher, Edinson Volquez, out of the starting lineup projections, since he had Tommy John surgery and will not pitch until August, if at all in 2010.

4. Blue Jays a bit higher, Marlins a bit higher

5. Eric Bedard is not with the Mariner's top 5.  He has been a great pitcher when healthy the last 4 years, but it is uncertain when he'll return this year.  Maybe at the end of May, maybe not till August.  One thing is for certain, he won't pitch 200 innings or anything close to it.  Even when he comes back, the track record of labrum tears is not very promising.  Even if he pitches half a season, don't expect him to strike out a batter per inning with decent control.  Pitchers with shoulder issues usually have drastic decreases in performance.  Unlike elbow injuries, where a Ben Sheets will probably be his usual self in the time he's able to pitch.

6. Jordan Zimmerman also had TJS and won't pitch much or at all in 2010.  I left him in anyway, because he's the best projected pitcher on the Nationals, and they happen to have a guy with no projection (sorry, I don't do college stats) ready to fill that ace role.

I have included a spreadsheet that takes team records, plays everyone head to head, and spits out normalized standings.  You can make the bad teams .600 and the good ones .750, or the good ones .500 and bad ones .350, and you'd get the same results.  This is the corrector for the fact that projecting only starting lineups gives you across the board optimistic projections.  The key to remember is that everything is relative.

Download it, change the defense from TZ to UZR, add Johnny Damon to your outfield, use CAIRO or ZIPS projected runs instead of CHONE, pretend Bedard starts game 3 and 31 starts after that, whatever.  Have fun with it.  Don't complain to me like a whiny little beeatch if you don't like my standings.  Create your own.  Anybody who wants to use this tool to create projected custom standings and put them on the internet is free to do so.

The problem is, I don't have time for that.  I am not going to spend a month of my baseball researching time trying to get the team projections just right.  So I call them just good enough and go with them.  I've been using the same process for a few years now and I seem to get the standings as close as anybody else.  (Don't take my word for it.)

I've thought of a way to make the predictions more open source, simpler, and it should be unbiased as well.  I thank Dave Cameron and his recent comments about the Mariner offense for the idea.  Take the R150 projections for the expected starting lineup.  Assume that everyone is healthy for every team, and that bench players taking up the other 12 games are average (or at least the same for every team).  I'm thinking for the pitcher counterpart I should look at 5 starting pitchers (figure runs above average for 200 innings) per team and maybe the top 2 relievers (since they will be the ones with the most leverage).

At this point we are deep into a Lake Wobegone league, where every team is above average (well, sorry Kansas City).  But that's OK.  Take these super-optimistic teams, play them in a schedule against each other using the odds ratio, and see what W-L records result.  This is your correction, and all the teams together will have to add up to .500

I will do this for both leagues, and make it open source.  Maybe even have it done in the next two days since I'm still snowed in.  But I will start with the American League offense, from the mighty Yankees at the top, to the very good top rivals in the AL East and Minnesota Twins, to good offenses in Anaheim and Cleveland, all the way down to the pathetic offenses.

There will be no secrets here, you can click on every team page and add them up yourself, but in case of any confusion I will later publish a spreadsheet telling who I have picked as the starter for each team and position.

Yankees +167

Ray +136

Twins +127

Red Sox +120

Indians +101 

Angels +98

Rangers, Orioles +60

Jays +30

White Sox +28

A's +20

Mariners +11

Tigers +8

Royals -23

For the Angels to beat this projection (and they usually do beat my projections) they'll need Kendry Morales to prove 2009 was no fluke, Howie Kendrick to finally play a full season and do so without going into a batting funk, Brandon Wood to justify the faith the organization has placed in him after three years of AAA, Ervin Santana to get his velocity back to where it was in 2008, and Scott Kazmir to get back to where he was in 2007.  Really, if 3 out of those 5 things happen the Angels should be strong.

When you play that optimistic "what if" game though, you have to consider what happens if things break right for your opponents too.  What if Vladimir Guerrero, who hit .300 and slugged .498 in the second half, is free from injury and takes advantage of his hitter's paradise?  What if the real Josh Hamilton is the one who looked like a superstar in 2008 instead of the medical case of 2009?  What if Elvis Andrus, only 20 last year, breaks out as the next superstar shortstop?

If things go right for Texas, they have the talent to go to their first ever world series.

Normally, projections do not forecast the same range of wins and losses as will happen in real life.  We expect that a few teams will win 95+ games, but are not sure exactly which ones, and if you pick any one team (Yankees excepted) the odds are they won't win that many games.

But yet I'm projecting 99 wins for the defending world champions.  I think this is the highest projection I've ever had, for any team.  I had them at 97 last year and they beat it by 6.  I like the moves they have made in the last year.  Curtis Granderson is a tremendous player who helps on offense and defense (at least against righties).  Javier Vazquez was one of the best pitchers in baseball last year, and Nick Johnson is OBP Jesus.  The Yankees are insanely talented, even more so than usual.  The breaks of the season could mean that Boston wins the East, or even Tampa Bay, but the talent spread is so huge in this division that Baltimore and Toronto have basically no chance.

The Twins already looked to be the favorites once more in the central, but upgraded by a win or two yesterday by adding the O-Dog, Orlando Hudson.  He will keep the group of Nick Punto, Brendan Harris, and Alexi Casilla from playing more, which can only help.  And hopefully tells Joe Mauer they want to win and convinces him to sign an extension.

Philadelphia might have blown it by trading Cliff Lee instead of Joe Blanton.  Halladay/Lee would have been the best 1-2 combo since Johnson/Schilling.  We'll see if they can hold off a very strong Atlanta club.

Thanks to resigning Matt Holliday, the St Louis Pujols should have a comfortable season in the NL central.  Most of the teams in that division don't spend enough money to compete.  Then there are the Cubs and Astros, who have spent on some real albatrosses.  I like the young talent in Cincinnati, but they probably aren't ready to challenge prince Albert just yet.

Projected Standings

Janish is not any kind of a hitter, but in the field combines excellent hands (.991 fielding %, only 3 errors) with a strong arm (89 MPH fastball off the mound) for a TotalZone projection of +5.