I have always been somewhat skeptical of the way that PB values are derived, in part because noone can really explain to me how they are derived. I have noticed in the past the following seem to impact PB values in a more significant way than you would like to believe.

(1) Playing Time. Consider the following example

Player A - 650 PA, .290 AVG, .370 OBP, .530 SLG, 40 HR, 10 SB, RF ODR = 5.0

Player B - 325 PA, .290 AVG, .370 OBP, .530 SLG, 20 HR, 5 SB, RF ODR = 5.0

You would think that (PB Value Player A) = 2.0*(PB Value Player B). However I have noticed that the PB value of Player B will be far lower than half of the value of Player A. I did not pull up an example with real PB values but from past experience I know there are a lot of examples out there.

(2) Position. If you play SS or CF and get a lot of playing time (600-700 PA) and hit at even a marginal level you are going to get a high PB value. And I'm not sure this is a bad thing as long as you know how to deal with it in a sensible manner. Just don't be shocked when the Doug Glanvilles and Carl Everetts of the world have a similar PB value to the Juan Gonzalez's of the world. But the sensible among us know which is a better player.

This is one reason presented by another person for the difference in Suzuki and Cameron and I believe this is the reason for the difference. I don't think its a HR vs. SB thing. In fact in the past I felt Tony Womack was OVERRATED because of stolen bases. Try comparing a slugging corner OF to a slugging CF (make sure to make the comparison fair by comparing people with similar OPS, playing time, and ODR) And I think you will find that the CF has a higher PB value most times because of the CF scarcity issue.

<3>Random Stuff that I can't explain

Sometimes I will look at two players (even at the same position) and consider their numbers and I am sure that player A has the best stats and the same playing time but player B has a higher PB rating. Sometimes I just throw up my hands and say "I think the PB value is wrong and this other player will help me more." my opinion is true this year more than in the past and especially with pitchers. Joe Mays vs. Randy Johnson -- who can explain why Joe has a higher PB value. Look at some of the long relievers that started often, threw 100-140 innings, and compare them to starting pitchers that threw 100-140 innings. The long relievers have very low PB values by comparision. The list goes on. I just threw up my hands this year on pitcher PB values and tried to figure out who I thought would help me the most. And to be honest where I was able to I tried to snag some long relievers that I think will be productive on the cheap due to their PB values.

In summary I think PB value can be used as a guide in some cases. But if you make the PB value your slave instead of a tool then you may make bad team construction tools. Frankly I look at the Grades for pitchers, the lefty/righty splits for relievers, and the OBP, SLG (including lefty/righty splits) for hitters. Defense and SB are secondary concerns in the sense that I don't want a whole team of iron gloves but if a couple of my good hitters have bad years on defense I just deal with it.