PECOTA hates us

[waroriole]

I wonder if Buck ever attended a BP convention? Sorry I think a Star Trek convention would be less boring.

Well since Buck is a manager, I guess he's got better things to do. You clearly do not.

[Guest rochester]

I dunno man... when projections say the team (best) or individuals may have a great season I love them; if they say they'll regress I don't like them because I don't want to believe it; if about the same meh.

I don't get the bashing of guys that love baseball and find various ways, like math, to enjoy the passion. Of course, I believe intangibles are a part of winning and losing.

IMO, I believe Buck uses every single thing available that may eek out one more win.

[El Gordo]

Give me Pearce and Clevenger and I'll plug you in to my model.

Pearce . 780 OPS if he makes the team, Clevenger .680 OPS.

[Ibanezsondajuice]

Hopefully they'll have some historically good offensive and defensive seasons so the projections will be off again and maybe we'll win about 85 games. Or maybe we'll set records in one-run games and extra innings again. It's possible one of those things will happen again. Makes more sense than building on our recent good fortune by adding more talent, right?

[Frobby]

I'm looking now at the detailed PECOTA forecast for Nick Markakis. As some of you know, PECOTA doesn't make one projection of a player, it makes a range: 90th percentile result, 80th, 70th, etc. in terms of a range of probabilities. Here is what they have for Nick:

90th: .815 OPS

80th: .788

70th: .770

60th: .753

50th: .738

40th: .723

30th: .707

20th: .688

10th: .662

That just strikes me as unbelievably pessimitic. Only a 10% chance that Nick exceeds .815 OPS? For a guy who was at .834 as recently as 2012? Only a 20% chance of topping .788 for a guy who has topped that figure in six of the eight seasons he has played? PECOTA obviously thinks he's on a permanent, drastic slide.

[Can_of_corn]

I'm looking now at the detailed PECOTA forecast for Nick Markakis. As some of you know, PECOTA doesn't make one projection of a player, it makes a range: 90th percentile result, 80th, 70th, etc. in terms of a range of probabilities. Here is what they have for Nick:

90th: .815 OPS

80th: .788

70th: .770

60th: .753

50th: .738

40th: .723

30th: .707

20th: .688

10th: .662

That just strikes me as unbelievably pessimitic. Only a 10% chance that Nick exceeds .815 OPS? For a guy who was at .834 as recently as 2012? Only a 20% chance of topping .788 for a guy who has topped that figure in six of the eight seasons he has played? PECOTA obviously thinks he's on a permanent, drastic slide.

They have always been pessimistic about Nick.

[Redskins Rick]

Why we do we care what PECOTA thinks? Seriously, they have an opinion, like most people do.

Personally, I could care less, it's not going to change the outcome one iota.

[Frobby]

Why we do we care what PECOTA thinks? Seriously, they have an opinion, like most people do.

Personally, I could care less, it's not going to change the outcome one iota.

PECOTA doesn't have an "opinion," it simply spits out numbers based on other numbers that are the inputs to the system. You are right, it doesn't change the outcome. But if I tell you there is a 90% chance of rain tomorrow, do you take an umbrella when you leave the house?

[El Gordo]

I'm looking now at the detailed PECOTA forecast for Nick Markakis. As some of you know, PECOTA doesn't make one projection of a player, it makes a range: 90th percentile result, 80th, 70th, etc. in terms of a range of probabilities. Here is what they have for Nick:

90th: .815 OPS

80th: .788

70th: .770

60th: .753

50th: .738

40th: .723

30th: .707

20th: .688

10th: .662

That just strikes me as unbelievably pessimitic. Only a 10% chance that Nick exceeds .815 OPS? For a guy who was at .834 as recently as 2012? Only a 20% chance of topping .788 for a guy who has topped that figure in six of the eight seasons he has played? PECOTA obviously thinks he's on a permanent, drastic slide.

Must be that Gregg Jefferies comp. Wasn't there a joke once about Nick becoming Mark Grace? That would be a good thing IMO.

[DrungoHazewood]

PECOTA doesn't have an "opinion," it simply spits out numbers based on other numbers that are the inputs to the system. You are right, it doesn't change the outcome. But if I tell you there is a 90% chance of rain tomorrow, do you take an umbrella when you leave the house?

I used to be a stalwart defender of PECOTA. But now I'm of the opinion that comp-based projections are no better than much simpler systems like Marcels. And they have the added "benefit" of occasionally spitting out frankly nonsensical projections when the right/wrong comps happen to line up and aren't sufficiently regressed.

[Hallas]

Jones is projected to hit right about at his career OPS. Hardy is just a tick off his last two seasons. Manny and Davis seem low, I would certainly hope they do better than that. I hope Manny breaks out, but he did have a disappointing 2nd half with the bat. I hope Davis holds most of his gains, but he also backslid in the 2nd half.

Davis, during his second half slump, struck out at roughly the same percentage (31.6% 2nd half 2013 vs 30.6% career) walked a ton more 12.1% vs 7.7% career) and still hit 16 HRs in 65 games. The main reason his numbers took a minor dive was because his BABIP was below his career average (.309 vs .335). I think he still showed most of the skills that made him the elite hitter his 2nd half even if the numbers weren't all there.

I would have really liked for him to continue the gains he made in plate discipline last April/May to transfer going forward, but he seriously regressed there. Even with his regression in this department, I don't see how you can project him for .801 OPS and 26 HRs unless you seriously think he's only going to play 130 games. Even if you consider his weak ML numbers as a 23 and 24 year old, you have to put some weight in his MiL numbers, during that time period, which were strong.

[Can_of_corn]

Davis, during his second half slump, struck out at roughly the same percentage (31.6% 2nd half 2013 vs 30.6% career) walked a ton more 12.1% vs 7.7% career) and still hit 16 HRs in 65 games. The main reason his numbers took a minor dive was because his BABIP was below his career average (.309 vs .335). I think he still showed most of the skills that made him the elite hitter his 2nd half even if the numbers weren't all there.

I would have really liked for him to continue the gains he made in plate discipline last April/May to transfer going forward, but he seriously regressed there. Even with his regression in this department, I don't see how you can project him for .801 OPS and 26 HRs unless you seriously think he's only going to play 130 games. Even if you consider his weak ML numbers as a 23 and 24 year old, you have to put some weight in his MiL numbers, during that time period, which were strong.

Which can be attributed to the increased amount of defensive shifts he faced.

In other words I don't think the lower BABIP was the result of luck.

[Hallas]

Which can be attributed to the increased amount of defensive shifts he faced.

In other words I don't think the lower BABIP was the result of luck.

There may be something to that, as his BABIP on grounders was terrible. I think that the difference between his 'expected' babip and his actual last season was quite high, however. Also, the shift can only happen with no runners on. So I still believe that his BABIP will rebound given enough games.

That said, he might benefit from bunting down the 3rd base line once in a while if he goes up with no one on base.

[Can_of_corn]

There may be something to that, as his BABIP on grounders was terrible. I think that the difference between his 'expected' babip and his actual last season was quite high, however. Also, the shift can only happen with no runners on. So I still believe that his BABIP will rebound given enough games.

That said, he might benefit from bunting down the 3rd base line once in a while if he goes up with no one on base.

http://espn.go.com/blog/boston/red-sox/post/_/id/34412/those-ortiz-shifts-they-do-hurt

No American League hitter was impacted more by shifts than Orioles slugger Chris Davis. With no shift, Davis had a .425 BABIP and a 17 percent extra-base hit rate (not including homers). Against the shift, his BABIP was .302 (123 percentage points less) and his extra-base hit percentage dropped to 10.1 percent.

[Hallas]

http://espn.go.com/blog/boston/red-sox/post/_/id/34412/those-ortiz-shifts-they-do-hurt

I believe Davis faced the infield shift all season, so I don't believe the 2nd half is representative of more shifts than the 1st half. Also, without information on his BABIP on groundballs against infield shifts and on normal infields, it's hard for me to gauge the magnitude that the shift had against Davis.

edit to add: Davis doesn't hit too many grounders, so I have a bit of a hard time believing that his overall BABIP would be affected by 100 points due to a defensive alignment that only affects 32% of his batted valls.