hardest to easiest categories to project

[viper]

This is really just for Gary & Todd - unless other spend time to do their own projecting.. Some categories are easier to project than others. Some you feel more confident in than others. I certainly feel that wins is hard to project but is it harder than ERA or SBs.

Could you rank the 10 normal categories with how you feel they rank in ease to difficulty in projecting.

[Todd Zola]

Personally, no categories are "harder" to project. They are all based on historical trends, global and micro.

If the REAL question is what categories do I feel are most reliable, that is I perhaps trust the projection the most, that I can answer.

The determining factor is how much of the category can the player control, and how much is either impacted by luck or team factors, of which there could be a cross-section.

For me, the category most under the player's control is HR. Yeah, there is luck involved, but not nearly as great as the other categories. I'm really not sure I can delineate the other 4. SB involves opportunity, runs necessitates teammates driving you home, RBI require getting hits with RISP and BA revolves around a player's BABIP, which we know can fluctuate around a player's baseline level. I suppose I would lump R and RBI as in the middle with SB and BA at the end.

For pitching, strikeouts are under the greatest control. After that, ERA we know is dependent upon defense and bullpen, along with the timing of hits allowed (LOB%), WHIP has the element of hits which is impacted by luck then there are wins and saves. I think the expression "wins and saves are a crapshoot" has taken on a slightly hyperbolic connotation. Good starting pitchers on teams with a good offense have a chance at a win. Realize that some bad teams have a bad record because of bad pitching. An excellent pitcher on a 70 win team has as good a chance at a win as the equivalent pitcher on a 90 win team, if the two teams have the same level of offense. I'm not sure if the variation with respect to wins is any more than the variation with other team related stat. What does differ is middle relievers getting 8 or 9 vulture wins, there is nothing analogous with hitting. Owners of Johan Santana may disagree after last season. But he did get 16 wins -- what was a reasonable projection, 18? 19? Saves are similar in that mark down 35 for a fulltime starter and accept the fact it may be 30-45.

If I had to rank after K, I next trust WHIP as the BB element is under the player's control. Then I lump ERA/W/SV together.

It's probably not the answer you want, but it is the best I can do.

Perhaps Gary has a different take, he has done the layman's work with respect to really looking at this.

FYI - we have teased it before, but on the "to-do" list is to come up with a way to incorporate these variances into error bars which would then define the "range" that we suggest is the best way to look at dollar values.

[rotodog]

For what it is worth, this is anecdotal only. Although I do not do my own projections because I leave that to people smarter than me, I do have to value those projections.

When it comes to pitchers, I usually devalue wins a little and add that extra to k's or Whip in the absence of K's....

Wins are just so unstable year to year for me...

[Guest]

For hitters -

HR
R/RBI really tie well to OBP/SLG
SB
AVG

For pitchers -
K
WHIP
ERA
SV
W

Basically HR rate and K rate are pretty damn stable, R/RBI and WHIP/ERA all relate well to the peripherals, SB relate well to the OBP stats, AVG tied to BABIP which moves, SV based entirely on role (but we know a lot about those roles in many cases), W very situational especially for guys who aren't top-30 SP.

[Todd Zola]

Meant yeoman's work, not layman's, sorry partner.