Dissecting Our Beloved Forecaster: TINSTAAP's Dissapointment

I am posting this in draft form first. I may tighten it up later, depending on my mood.

TINSTAAP was predicted by PECOTA through Justin to finish in first place, yet finished the year in sixth place. As has been noted previously, PECOTA is a propietary tool of BP that I (and other players) use extensively in our pre-draft fantasy baseball rankings. I am interested in figuring out what might have gone wrong with its predictions or Matt's use of those predictions. In this venture, I am going to limit myself to the facts of the auction draft. Also, I am not going to look at PECOTA's actually predictions for the players. In most cases, I am going to assume that Matt's value is an adequate reflection of what PECOTA predicted. In this regard, I present you with my analysis.

WINS
Dustin Pedroia $8.00
Pedroia had a surprisingly good year and Matt made up a lot of value here. A young player with good peripherals that may have made him projectable by PECOTA.
Jhonny Peralta $1.00
Peralta had an average to above average year, and, at the cost of $1, is about all you can hope for. He is young and has been relatively consistent, which may have made him projectable by PECOTA.
J.J. Hardy $1.00
Peralta clone. Similarly, great value pick.
Adrian Gonzalez $14.00
Gonzalez is an underrated youngish player who improved on his performance from last year to become a superstar. He has a solid, consistently improving periphals. This may have been the best money Matt spent all year.
Ryan Church $1.00
Matt was the first on the Church bandwagon and was rewarded with spectacular performance early on, with decreasing returns as the season progressed. This was a good value pick, but I know that holding on to Church through all his dramas this season ended up hurting more than helping my team in another league.
LOSSES
Derrek Lee $24.00
I thought Lee had a much better year than he actually did. 20 HR and 8 SB for a first baseman that you bought for $24 is very disappointing. Lee sunk many a fantasy seasons last year. PECOTA probably predicted a bounce-back to his preceding level of production that was not to be. However, Lee is an older player on the decline, and PECOTA was too optimistic.
Hideki Matsui $9.00
Projectors may have expected Matsui to bounce back from his injuries to his career above average to good performance, but he didn't get there because of those darn injuries. Similar to Lee, he is an older player on the decline.
Jeremy Hermida $7.00
I remember Hermida being up and down all season. PECOTA saw a potential for greatness, but he just wasn't consistent enough to meet the price paid for this year.
Adam Jones $12.00
Younger and sexier than Hermida, what PECOTA didn't know that he wasn't ready either, ergo the disconnect between price paid and projected value. Because he was new and desired, he turned out to be an even bigger disappointment than last year's fad, Hermida.
Scott Kazmir $24.00
Kazmir had an uneven season and was always at the highest a 50-50 shot to earn $24 draft day money. Pitchers are unpredictable. 'Nuff said.
Brett Myers $14.00
Myers may have had $14 of value over the course of the season, but I doubt it was for Matt's team.
Yovani Galladro $13.00
Season (almost) ending injury to a pitcher. Predictable? Possibly. That may have been taken into account in his auction day price already.
WASHES
Kurt Suzuki $1.00
Suzuki had a mediocre to average year for the catcher position, but Matt did not spend much on him.
Felix Hernandez $19.00
At a lower price range than Kazmir, he may have done well enough to earn his value at least.
Tom Gorzelanny $4.00
Upside pick that didn't pay off.
Daric Barton $4.00
Upside pick that didn't pay off.
Travis Buck $1.00
Upside pick that didn't pay off.
Joel Zumaya $2.00
Francisco Rodriguez $5.00
Relievers don't count for much.
Jonathan Broxton $5.00
Relievers don't count for much.
Joba Chamberlain $10.00
Price seems a little high for a reliever, but he did put in a lot of excellent innings as a hybrid pitcher and reliever.

I do see some patterns here with Matt and/or PECOTA from this small sample size, but nothing earth shattering. They do a good job of forecasting young players who have some major league experience. They do a bad job of forecasting declining middle-aging players who have a relatively recent history of success, but a more recent history of suckitude. They don't do well at picking when young players will actually first experience success in the major leagues. And, as I know all too well (and Matt too, see his team's name), they have no idea what pitchers are going to do well and what pitchers are going to suck.