2013 ZiPS projections

[filihok]

Dan Szymborski’s ZiPS projections are currently being released at FanGraphs.

The first team released was the Giants

http://www.fangraphs.com/blogs/index.php/2013-zips-projections-san-francisco-giants

[KHenry14]

Why in the world would they attempt to project stats for minor leaguers??

[filihok]

Why in the world would they attempt to project stats for minor leaguers??

Well, because one might want to know how a minor leaguer is expected to perform

[KHenry14]

Well, because one might want to know how a minor leaguer is expected to perform

But how can you possibly accurately predict an immature player who's still learning the game? Too many variables if you ask me to gain any kind of insight.

[filihok]

But how can you possibly accurately predict an immature player who's still learning the game? Too many variables if you ask me to gain any kind of insight.

http://www.fangraphs.com/library/index.php/the-projection-rundown-the-basics-on-marcels-zips-cairo-oliver-and-the-rest/

- ZiPS – The work of Dan Szymborski over at Baseball Think Factory, the ZiPS projections uses weighted averages of four years of data (three if a player is very old or very young), regresses pitchers based on DIPS theory and BABIP rates, and adjusts for aging by looking at similar players and their aging trends. It’s an effective projection system, and is displayed at FanGraphs for off-season and in-season projections.

Here's a link to the 2012 ZiPS projections

http://www.baseballthinkfactory.org/oracle/discussion/2012_zips_projections_-_san_francisco_giants

Hector Sanchez was a rookie in 2012.
ZiPS projected him to hit .288/.365 (OBP/SLG)

Sanchez actually hit .295/.390 (OPB/SLG)


ZiPS isn't always that accurate, but it works.

[Honus Wagner Rules]

But how can you possibly accurately predict an immature player who's still learning the game? Too many variables if you ask me to gain any kind of insight.

Bill James discovered back in the 1980's that one can look at player's minor league stats and by taking into account the context, one can get a fair idea of how a young player may do in the majors. But these are only projections or estimates. They are not to be taken as absolutes or exact stats to expect. Today the tools are more sophisticated of course. To me these projections are useful in the sense that one can fit a player into a type of profile. For example take Brandon Belt. What kind of hitter is he? What previous players had similar minor league stats in similar minor leagues? How did these other players do in th majors? Based on Belt's minor league numbers he has under performed in the majors. We all know Bochy jerked Belt around most of 2011 and Belt also got hurt which led to Belt's poor 2011 season. He was much better in 2012 if a bit streaky. But we saw flashes of brilliance as well. I still truly believe that Brandon Belt will be a great hitter, in the mold of Fred McGriff or a pre-concussion Justin Mouneau. I expect .280-.300 BA, .380-.400 OBP, .500-.550 SLG% in his prime seasons. I hope I am right.

Kind of side note. Baseball men have been doing prospect "projections" on an qualitative or intuitive level for a long time. Today I was researching Jim Thome and it's kind of funny how he was viewed as a prospect. In the minors he hit for little power but high batting averages. Baseball people were "projecting" him to be George Brett type of hitter.

This is what Bill James said about Home in the Baseball Book 1992 :

10. Jim Thome, Cleveland
I could have him rated way too low. He's the opposite of (Leo) Gomez and (Dean) Palmer, a George Breat, Carney Langsford type of hitter as opposed to Mike Schmidt type. Has an outstanding chance to be the 1992 Rookie of the Year.. I project him to hit around .290 with only five home runs but a chance to surprise on the power.

Well, Thome morphed into a Mike Schmidt type hitter.

And this is from Craig Wright who for worked for the LA Dodgers in the early 1990's. He is writing about Dodgers GM Fred Claire approached Wright to ask him about his thoughts on a prospect named Jim Thome. Craig is commenting on Thome power potential.

Just a note on Thome's power potential. There have been a lot of hitters who made significant strides in their power in their mid to late 20s. Thome strikes me as that kind of hitter. Besides the potential to hit for a good average, decent power, and get on base a lot, I could see a scenario where he'll eventually be poking over 20 homers to go with it, and that will make him an All-Star. He is a strong kid, 6-4, 220. I've seen him swat at least one prodigious 2nd deck homer.

Fred called me shortly after getting that report and asked if I really believed that Thome would develop 20-homer power. That sounds like such a funny question today, but viewed in context, Fred's question was quite reasonable. Going into the 1993 season Thome had hit just 9 homers in his 493 ABs at AA and AAA, and in his major league trials he had hit just 3 homers in 215 ABs. That’s just 12 homers in his 700 ABs above A-ball. I reiterated my belief that Thome had good power potential, and pointed out I expected better than 20-homer power.

Now 20+ homers was a lot to talk about back then, which was before the modern home run explosion. In the most recent full season at that time, 1992, the HR% was 2.13%. We crossed the 3% mark in 1998 and have averaged 3.17% since, or about a 50% gain. In modern terms I was essentially saying I expected Thome to develop 30+ home run power. As optimistic as my analysis was, I had no clue we were talking about someone who would develop the monster power he has shown in his career.

Again, no one had any clue that Thome would develop the home run power to hit over 600 career home runs and I doubt any sabermetric projection done using Thome's minor league stats would have projected Thome to be a legit 40+ home run hitter.

[Jackaroo Dave]

Why in the world would they attempt to project stats for minor leaguers??

Another reason is to evaluate and refine their projection techniques.