Baseball Prospectus 2008

[Yoda]

I don't think I am getting a copy this year. I already have the online subscription and I have all the projections on a spreadsheet so I should be good to go. I'll look at the cards when they come out.

[Omaha Red Sox]

Got an email yesterday that mine's in the mail.

[duckbillgates]

Can somebody sell me on this? I've read about it for years in the Cafe and I flipped through it in a bookstore this weekend and was unimpressed.

The projections seem ridiculously conservative and somewhat arbitrary. In looking at a few players, the projections didn't seem to make much sense in conjunction with the player analysis, i.e. glowing analysis with conservative projections and vice versa.

Basically, I felt they were looking at the right stats but didn't feel they had any greater insight than most of the fantasy baseball magazines (which isn't a whole lot).

[garf112]

Most of the projections are going to be "conservative" as he usually regresses folks back to their mean.

As far as recommending that you buy the Prospectus. I don't know. I feel like Shandler's concepts are better than his projections. To me, his ideas are another fantasy "tool" that I have to create my own projections/ rankings. Maybe he has some revolutionary new article in there this year, but I doubt it. I think he is very good at explaining his methodology. If you feel like you have a good idea on how to use BABIP, BB rate, K rate, GB%, FB%, etc, than you probably don't need to buy the book.

Just my opinion, though.

[The Mighty Trow]

I've never looked at the Prospectus or Forecaster, so I went out and bought both of them yesterday. My humble opinions after two days reading:

Prospectus has more of a real life feel to it. Often discusses player contract terms and value, which as far as I know have nil fantasy impact. This is what I read to learn more about baseball the game, not the fantasy game.

Forecaster is marketed toward fantasy baseball managers. Heavy on the statistics, light on the summaries. Gives you oodles of metrics which are intended to be used as a tool for making your own forecasts and projects.

Both are very enjoyable reads. Forecaster is heavier on the numbers -- if you don't like math, you might not like this book so much. I think they're both very useful tools for forming your own opinions and I'm glad I got them.

[jfg]

You can't use BP to get a good handle on your numbers while drafting but I think PECOTA is the most accurate for actually ranking the players when you use PFM or put the numbers in your own formula. What ends up happening is that their are a bunch of guys that pop at you and you say "what the heck is that all about" and then you draft them a couple rounds before people start putting that player on the radar and in the end you win leagues.

I think PECOTA is the best at predicting young players and players about to take a downturn. The only thing with young players is that it definitely overrates them even when playing time is considered. Like I said, you can't use the numbers when drafting but rather use those numbers as an indicator that good things should come- just probably not great things.

Two things that have popped out at me: Matt Wieters as a top 15 hitter and a pretty low projection for Longoria. I rarely go against PECOTA but I have Longoria in all three of my leagues.

[hot4tx]

o I bought the Baseball Prospectus for the first time this year and I am finding it very, very interesting. I know many of you use and/or swear by it and its PECOTA projections and I, as a PECOTA noob, have a few questions for those of you who do.

First of all, let me make sure I'm understanding the gist of what is being done - if I understand it correctly they are finding similar players (using stats, bio info, inflation adjusted first signing bonus, etc, etc) and weighting the amount of similarity to those players as of a certain point in their careers? Then taking the following year(s) stats for those players, at the similarity-weighted amount, and using that to project this year's stats for the player in question?

If so I find the idea of analog career paths as a prediction of a players future a completely fascinating idea. I wonder why I never considered it before, since I do just that for many oil companies.

If that is what they are doing I have several other questions both about their projections and about how you practically use their projections - ie do you alter the Plate Appearances and rates as you see fit to create one projection or do you continue to view it as projected distribution?

TIA.