My so-called secret is a disaster. I used PB value extensively to create my draft list lists for LLs 1 and 2. I adjusted all players to establish value versus average and replacement players -- to diminish the affect of just playing every day. Every player was re-indexed to their position. If PB value really measured the ability to create or prevent offence, my teams would be optimal.

LL1, the team looked great. WS potential. Alas, it looks like I won't make the playoffs. I think bad luck hit this team because it has not scored runs like I expected. I drafted a couple of rounds without my list, and it probably helped. Bad SP certainly hurt but still I think it was a good team.

LL2, stuck to my formulas and drafted a terrible team.

Here's what PB value gets you:

It loves Right-handed (ie, lots of innings) RPs. Turns out, saves are in the formula.

When I did draft position players, they are highly skilled beyond hitting: fielding, Sacs, speed, SBs ... . PB value, in LL2, led to a team with rotton SP, superlative RPs, and a bunt-and-run offense. So, once I'm down 5-0 in the 5th, my pullpen can stop more damage while I try to bunt my way back. loser.

New Wisdom=Old Wisdom. Gotta draft an overall strategy. PV value overstates reliever's value. It might accurately reflect batters value in real NL baseball, but it does not adjust for the more concentrated PB rosters. I'm almost back to the "ignore PV value" mindset, although my experiment failed in large part due to a devotion to a measurement rather than being adaptive. Obviously, any drafting system that does not adapt to the guys you've already picked is a bad system. In LL3, I used my adjusted PB values for the first 4 rounds, then dropped it. The draft list seems to be pretty close to what everybody else had, it's value is subtle at best.