The fact that a single study found no evidence in support of a hypothesis proves nothing.  Ask any scientist or mathematician.  How a study is designed has a dramatic effect on the outcome.  This is especially true when you are studying human behavior and performance.

2.  You keep touting FIP as the definitive measure of pitching performance.  Where is the proof?  FIP is a theoretical model based on assumptions and it depends on what kind of data are chosen as input, like any theoretical mathematical model.  The formula is (HR*13+(BB+HBP)*3-K*2)/IP, plus a league-specific factor (usually around 3.2) to round out the number to an equivalent ERA number.  You do realize that this is an invention, don't you?  How does it stand up to empirical tests?  That is, what is the actual correlation between FIP and some independent, valid and reliable measure of the actual level of pitching performance, entirely independent of fielding?  What do you correlate the FIP data WITH?  There is no definitive reference point for even testing that correlation with any certainty.  Why?  Pitching success has too many confounding factors (pitching coaches, catchers, umpires, managers, the pitchers own "confidence" and "motivation" and other such highly subjective factors, and even teammates and extraneous events such as the pitcher's personal life).  Who is to say which factors most affected a pitcher's performance more than another factor on any particular day or week or year?  I won't bore you with more about this.  To get the full picture, check out the literature on tautology and the philosophy of science written by researchers who are highly regarded by the foremost scientists and mathematicians in the world.

You might also consider this quote from the inventor of FIP:

"What if my regression rates are wrong? Well, that's definitely a possibility. How much you regress has a huge impact on the whole chaining process." - Tangotiger, inventor of Fielding Independent Pitching (FIP) analysis for pitchers

http://www.tangotiger.net/adjacentPitching.html

3. Re: So Lieber's been better, using your data, over the time period you prefer, and he's been better over the 3 year time period, using FIP.  

Lieber HAS BEEN better over the course of his career than Pineiro has been, yes, but not anymore.  Lieber is in decline.  His ERA+ was 98 this year in 78 IP (if the sample size is too small for you, then consider that Lieber's ERA+ in 2006 was 95 in 168 IP).  He will be 38 spring.  He is a #3 pitcher at this point, no better, and the Cardinals need an authentic #2.  

4.  Re: I'm puzzled why Duncan always seems to get the credit when pitchers perform better under his tutelage but NEVER seems to get the blame when pitchers regress.  Reyes and Marquis, for example, are blamed as being "head cases" or "immature" while Duncan is, ostensibly, immune from blame.  Duncan gets no blame for Wells' pitching this year.

I will readily agree that Duncan is not perfect.  If you want to present cases where pitchers performed less effectively under his coaching, then bring them on.  I included only four cases so as not to overtax anyone who reads this string and not to overtax my own time.  I left out some of Duncan's more celebrated successes in Oakland, for example.  

I think Duncan's ego got in the way of his handling of Wells this year.  Duncan had put his personal endorsement on the line and he stuck with Wells far too long.  Wells should have gone to the bullpen, where he had much better success, perhaps trading places with Thompson or Franklin.

I agree with your criticism of the handling of Reyes, too.  I think the guy's confidence was destroyed by the way Duncan tried to force him into the ground ball pitcher mode that is Duncan's dogma.

As for Marquis, consider that the higly regarded pitching coach in Atlanta got fed up with Marquis, and the Cubs left him out of their rotation for the playoffs this month.  Duncan's frustration with Marquis is hardly an aberation nor an alibi.

We seem to actually agree on a few things!  How about that?

Nevertheless, when Duncan connects well with a pitcher, it is evident from a number of cases that such pitchers are more successful than before and after they work with Duncan.  The point is that Pineiro and Duncan worked well together, and the results showed.
 

Anyone who understands regression analysis well will tell you that adding more data points improves predictive power.  It does not add "noise", even though, of course, data that are older chronologically become less and less potent in adding predictive power.

2.  "Keith Woolner wrote some great articles at BP about catcher's impact disproving it in large part (at least to the point that current methods were unable to capture anything above noise)."

Scientists know that when you study human behavior and performance you are in highly complex, subjective territory, with high variances.  One study in this field does not prove or disprove anything.  Your statement actually supports my point: "current methods were unable to capture anything above noise".  Yes, that is true.  And it is merely the outcome of the study.  It is not a legitimate conclusion about the matter being studied.

3.  "Point out where I'm inputing garbage.  I've pointed out where you are (data older than three years).  The stat I've used (FIP) is a better assessment of a pitcher's ability than ERA (aka it reduces the garbage)."

I certainly mean no disrespect.  The "garbage" quote is familiar to anyone who works with mathematical or scientific formulas or computer programs.  What I mean is that FIP is just a model, like any other, that was fabricated based on assumptions and on selective inclusion and exclusion of certain kinds of data.  It would take far too much time and space to address this fully.  I'll just use one example.  The formula only counts homeruns, not other kinds of hits, on the premise that homeruns are entirely independent of the fielders behind the pitcher.  That is true.  The problem is that there is a confounding in the data, because homeruns are also affected by the pitcher's ration of fly balls to ground balls, and that factor is not included in the formula.  Thus, a distortion is created in that groundball pitchers will be given higher performance scores than fly ball pitchers.  In other words, the FIP formula is inherently biased.  As a result, it does not fairly measure the performance of one pitcher with another's.  (See my FIP comments in the earlier post above.)

4.  "Again, we resort to ERA.  Haven't I discussed this enough?  It's not an accurate representation of a pitcher's ability."

I absolutely agree with you that ERA is imperfect.  ERA+ is much better because it corrects for certain confounding factors, but not all.  But no measure anyone has invented so far has the level of mathematical validity, reliability, orthogonality, and power (I'm using these terms as a mathematician would, each with a distinct, precise meaning) that will satisfy those of us who are so engrossed in both baseball and mathematics.

5.  "I like how you cite the impressive body of work and then name 4 pitchers.  Woo buddy!  4 pitchers.  Well damn me but I forgot about thsoe 4 pitchers.  That is an impressive body of . . . or no wait it really isn't....  I used to attribute a lot of success to Dave Duncan (and I'm almost always in favor of grabbing some reclamation pitcher for a cheap 1-year deal) but is there any real evidence that it's Duncan?"

I have my own criticisms of Duncan, but I do believe he makes an exceptional difference in a pitcher's success.  I've already addressed this in the post above.

6.  "To bring this back to the main point, Lieber is a better pitcher (independent of park and defense) than Pineiro."  

I respectfully disagree.  I've addressed this point in more than one post above, already.

Thank you for sharing your perspective and analysis.  This has been an interesting debate.  I hope it hasn't been too tedious or boring for anyone who continued to read hoping for something other than this kind of discussion.