I like their theories and articles, and as far as accurate projections go, I dont think you can do much better. But at the same time they tend to be a little pessimistic when it comes to projections. I like to make my own projections based on a few opinions and my own judgements. You should never take any set of projections as your one and only.
RynMan wrote:I like their theories and articles, and as far as accurate projections go, I dont think you can do much better. But at the same time they tend to be a little pessimistic when it comes to projections.
Funny enough I just finished comparing Prospectus projections to my own and came to the same conclusion as RynMan. This is the first year I have ever used it (as it came highly recommended from several cafe people), but either they think the new steriod testing program is going to kill homers or they are a little conservative. That said it is a good source of information.
Major League Manager
Joined: 7 May 2004
Bases this season: 0
Home Cafe: Baseball
Location: Finally back home. A sweet 15 mi from Miller Park.
There is a major difference in the projections done by Baseball Prospectus and others out there. It isn't a subjective process. They don't say, "We've got a good feeling about Player X this season," and then generate statistics to meet that. I'm not saying any one process is better than another, but just that their's is markedly different from what most people do when making projections. It's a systematic analysis of past performance at major and minor league levels, factoring in age and playing experience. I've never read their book, only a subscriber to their website. Their PETCOA projections don't just end at 2005 but offer a five year outlook. A better description comes from their website:
[The PETCOA] Forecast is a representation of the hitter's expected performance in the upcoming season at various levels of probability. For example, if a hitter's 75th percentile EQA forecast is .296, this indicates that he has a 75% chance to post an EQA less than or equal to .296, and a 25% chance to post an EQA better than .296. Higher percentiles indicate more favorable outcomes.
PECOTA runs a series of regressions within the set of comparable data in order to estimate how changes in peripheral statistics are related to changes in equivalent runs. For example, if it first estimates that Pat Burrell will produce 100 equivalent runs next year, it then tries to determine what home run total, walk total, and so on are most likely to be associated with a 100 run season.
A player's [projected] numbers are adjusted to the park and league context associated with the team listed at the top of the forecast page. Park factors are based on a three-year average over the period [2002-2004], except for teams that have changed ballparks.
PECOTA forecasts playing time (plate appearances) in addition to a player's rate statistics. These forecasts are based on a player's previous record of performance, and the comparable player data, and do not incorporate any additional information about managerial decisions.
[The PETCOA] Forecast is a representation of the pitcher's expected performance in the upcoming season at various levels of probability. For example, if a pitcher's 75th percentile ERA forecast is 3.52, this indicates that he has a 75% chance to post an ERA of 3.52 or higher, and a 25% chance to post an ERA less than 3.52. Higher percentiles indicate more favorable outcomes.
PECOTA runs a series of regressions within the set of comparable data in order to estimate how changes in peripheral statistics are related to changes in ERA. For example, if it first estimates that Woody Williams will produce an ERA of 4.29 next year, it then tries to determine what strikeout rate, walk rate, and so on are most likely to be associated with a 4.29 ERA season.
A player's [projected] numbers are adjusted to the park and league context for the team listed at the top of the forecast page. Park factors are based on a three-year average over the period [2002-2004], except for teams that have changed ballparks. In addition, the pitcher forecasts include an adjustment for team defense which affects the pitcher's H/BIP.
PECOTA forecasts playing time (games and innings pitched) in addition to a player's rate statistics. These forecasts are based on a player's previous record of performance, and do not incorporate any additional information about managerial decisions.
As far as being pessimistic or conservative, I believe that is where the percentile forcasts come in.
The one area I think they excell in beyond all others is in identifying players the majority of people overlook. There were plenty of opinions to be had on well known players like Barry Zito (how would he respond after his 2003 season) or Ken Griffey Jr. (would he stay healthy in 2004). But thier 50th percentile projection for Wily Mo Pena last year had him hitting 18 homeruns, and the 90th percentile 25. I don't remember much talk about Pena during last year's offseason (I had no clue who he was until I saw their projection).
I pay attention the forecasts with surprising production, either because it is someone I haven't heard of or someone I didn't expect to produce to their projected lever, or those that have older players declining sooner than I would have thought.
Edit: Wow, anytime you type "Baseball Prospectus" the Cafe automatically links it.
My 2005 version is coming in the mail today and the comments are making me eager to read through it as I've not actually read it before.
I think that a lot of projections I have stumbled upon see to predict EVERYONE will do well as when you add up all of the homers, or strikeouts, or SB or whatever, you end up w/ about 50% more than MLB season records. Which does not seem likely, particularly for both batters and pitchers in the same year. Ideally, of course, the projections will tell you WHICH players are going to do better than expected and which aren't but, after all, that's why drafting is so much fun. Until Sexson's shoulder pops out again!!!
One other thing, I had noticed the linking too but...it's to the 2004 edition when 2005 is the one that came out last week!!! Don't buy that one and think you are getting this year's model!!!
If the 2004 Prospectus book is any indication, all you will get are the raw, weighted mean PECOTA projections in the 2005 version. In 2004, they did NOT give the percentile breakdown and they did NOT give their fantasy projections. These have only been available through their site (which is a great site by the way).
Since we're on the subject, the projections currently available at their site are these same weighted mean projections. Whenever they launch their 2005 fantasy content (which is coming soon), there will be another set of projections released, intended specifically for fantasy consumption. The forecasts for playing time will not rely as much on PECOTA, hence will not be as pessimistic. They will also include the fantasy stats (i.e. RBI, R, W, L, etc.).
Unless Prospectus has changed their format, you will be disappointed with the book if you're only buying it for the projections. However the book offers so much great commentary, it's well worth it. In the past, Prospectus has offered a subscription (~$10/year) purely to their online fantasy content (where you could get the fantasy projections), but they seem to not be offering that this year (again things may change).
Thanks!! I was kind of hoping for more article type of stuff as I'm already getting a bit fried looking at my spreadsheets and online lists of players. Given that nobody knows who's gonna do what, I think that I am more oriented towards articles than stats anyway. I don't really strictly rank players but have ideas about where fault lines are (based on rankings, of course...) but I just go w/ who's available and the best player and count on the final rounds to pick up young players, etc. to fill in the blank lineup spots...