The O's at 20% of the season gone.

[Frobby]

The analogy is far from perfect, but last year Steve Pearce had an .875 OPS in his first 91 PA (Paredes has 92 PA now). I think most of us were expecting him to regress back to the .694 OPS hitter he'd been to that point in his career. Instead, he posted a .948 OPS the rest of the season to finish at .930. So who's to say whether Paredes' outlier season can continue? (And yes, I realize Pearce had a better track record than Paredes. I'm just illustrating that baseball produced a lot of surprising results.)

[DrungoHazewood]

The analogy is far from perfect, but last year Steve Pearce had an .875 OPS in his first 91 PA (Paredes has 92 PA now). I think most of us were expecting him to regress back to the .694 OPS hitter he'd been to that point in his career. Instead, he posted a .948 OPS the rest of the season to finish at .930. So who's to say whether Paredes' outlier season can continue? (And yes, I realize Pearce had a better track record than Paredes. I'm just illustrating that baseball produced a lot of surprising results.)

Baseball produces many fewer pleasantly surprising results than expected results. Jim Traber set the world on fire for a while in 1986, then was sub-replacement for the rest of his career. Chito Martinez slugged .514 for a surprising 67 games in 1991, and his career lasted another 91 games. We all know about David Newhan, and his .400 batting average his first month in Baltimore. Ryan Kohlmeier had 13 saves and a 2.39 as a rookie, his career was over 40 innings later. Jeff Manto had a remarkable run in '95 where he hit four homers in four consecutive at bats, ending up slugging almost .500 in 89 games. He never played 50 games in a MLB season again. I do hope Paredes surprises us.

[DJHJR86]

First post here.

I think the O's will find some sort of balance in the pitching within the next month or so. If Norris continues to regress, Gausman will get the nod. We have some decent pitching depth in Norfolk as well, just in case out bullpen starts to flounder. I honestly think we'll be fine in the long run.

[weams]

First post here.

I think the O's will find some sort of balance in the pitching within the next month or so. If Norris continues to regress, Gausman will get the nod. We have some decent pitching depth in Norfolk as well, just in case out bullpen starts to flounder. I honestly think we'll be fine in the long run.

Welcome to the Jungle.

[Aristotelian]

I was a mathematics major in college. And I've been trying to follow the current sabermetrics. What I've found is that most of the sabermetric methods are quite linear and static. What current sabermetrics doesn't really follow are the signs that a change in a player's performance is eminent. So instead of looking at Paredes's record and saying that he has an extensive track record, so eventually, he'll likely revert to this track record: instead, sabermethics needs to look for signs of a change in the track record. In order to measure change, linear methods don't usually work. Instead, one must use diferential calculus. I won't go here into how differential calculus works here, but just to say that differential calculus measures change in performance.

Mind blown. I agree, aberrations can be just as interesting depending on what you think they mean. However, I would guess that it tends to work in the opposite direction; that is, you are more likely to see a precipitous decline due that later turns out to be caused by an injury, rather than a major increase that leads to sustained production. It is much easier for a player to suffer a crippling injury than to do anything to make himself orders of magnitude better at hitting a baseball. The only thing I can think of is PEDs, and I sure hope that is not what's going on with Jimmy.

[Frobby]

Baseball produces many fewer pleasantly surprising results than expected results. Jim Traber set the world on fire for a while in 1986, then was sub-replacement for the rest of his career. Chito Martinez slugged .514 for a surprising 67 games in 1991, and his career lasted another 91 games. We all know about David Newhan, and his .400 batting average his first month in Baltimore. Ryan Kohlmeier had 13 saves and a 2.39 as a rookie, his career was over 40 innings later. Jeff Manto had a remarkable run in '95 where he hit four homers in four consecutive at bats, ending up slugging almost .500 in 89 games. He never played 50 games in a MLB season again. I do hope Paredes surprises us.

I can't disagree with this, of course. It's just that the occasional surprising result is part of what makes baseball fun to follow.

[DrungoHazewood]

I can't disagree with this, of course. It's just that the occasional surprising result is part of what makes baseball fun to follow.

Absolutely. If the surprises were commonplace then they wouldn't be surprises, and the world would be very, very strange. Can you imagine a universe where past performance had nothing to do with future results?

[webbrick2010]

Seems very likely that JP will regress to his career norms, but I find it significant that it has happened. He has made a meaningful contribution to the Orioles 2015 season, similar to Pearce in 2014.

It seems the O's system of having many moving parts and not being afraid of giving an un-proven player a chance is a unique organizational strength.

I suspect that JP will be all but forgotten by August, but perhaps the next 3 month phenom will be Dariel Alvarez or Noland Riemold or Henry Uurita. Credit to Buck for recognizing and playing the hot hand.

[Nevermore]

Last year, split between Omaha, Norfolk, and the majors, Paredes had 10 home runs in 485 PAs. That was mostly in AAA, and the majority of those at bats were in the PCL where he was roughly the 9th-best hitter on the Omaha Royals, as a 25-year-old. Then in late July he was waived by the Royals. You can talk all about the limitations of sabermetrics, and how a 3400 PA track record might be irrelevant. But if Paredes has had a massive step change in ability or talent it happened very recently. There was little change in his track record until late last year. If anything he was trending downwards - in '13 he walked in almost 8% of his AAA PAs while striking out in only 19%, but last year that became 4% and 25%.

Conventiional sabermetrics can't see signs of a recent massive step change. Linear equations don't do that. But differential calculus can with the right equation done at the right time.

[DrungoHazewood]

Conventiional sabermetrics can't see signs of a recent massive step change. Linear equations don't do that. But differential calculus can with the right equation done at the right time.

How would you tell a change in talent from random variation?

[CA-ORIOLE]

How would you tell a change in talent from random variation?

I think they rely heavily on the scouts for this.

[DrungoHazewood]

I think they rely heavily on the scouts for this.

Of course they do.

[CA-ORIOLE]

Of course they do.

Then why are you defining this as a purely statistical argument? It's not like scouts don't have better and better analytical tools in addition to evaluating the statistics.

[DrungoHazewood]

Then why are you defining this as a purely statistical argument? It's not like scouts don't have better and better analytical tools in addition to evaluating the statistics.

I was agreeing with you. I don't understand how the approach Nevermore is suggesting could work.

[CA-ORIOLE]

I was agreeing with you. I don't understand how the approach Nevermore is suggesting could work.

Ok. Fair enough. I'm not attuned to scouting at all but my guess is they might have pretty slick analytic and statistical tools available to them these days and can look at some of these things at a much more incremental level than the larger aggregate and more conventional statistics. Basically to Nevermore's point, statistics and scouting are likely merging in many aspects..