Three Words: PE-CO-TA (Player Projections)

[CubFanPhilly]

Why did you give Ramirez 100 fewer AB's than Lee?

 

I just guesstimated based on his career stats. I think he has at least one DL stint every year of his career. I also think he's likely to draw more walks with no real threat behind him. Will anyone pitch to A Ram with Jones behind him?

 

But certainly Ramirez could have a healthy year and get to 600+ ABs and that'd be a great thing for the Cubs.

[CubinNY]

Does anyone have data on the accuracy of PECTOA projections over the last three years? What I am looking for are standard deviations or + or - percentages.

 

I am interested in determining if these data are worthwhile or are just fun food for thought stuff. I bookmarked the ones from the baseball think factory for 2005 at the begininning of last season but when I returned the page wasn't there.

 

Thanks in Advance

 

After year one of PECOTA, BP compared it to other prediction systems here.

 

Correlation coefficients for OPS: .700-.711

Correlation coefficients for ERA: .479-.486

 

PECOTA performs better than most, but its certainly not ironclad. On the pitching side, no one's got a good prediction system. The Bill James Handbook took its first crack at pitcher projections this year, without the help of James who doesn't believe a good system can be made.

 

.7 values are very good with that size of a population,

[stitchface]

That looks like a bullpen ERA around 4 - how is that so terrible? What was the bullpen ERA last year? Looks pretty optimistic on Williamson to me.

[Transmogrified Tiger]

That looks like a bullpen ERA around 4 - how is that so terrible? What was the bullpen ERA last year? Looks pretty optimistic on Williamson to me.

 

The Cubs pen had an ERA of 4.24 last year, tied for 9th in the NL. A 4.00 ERA would've put them alone in 8th last year, between San Fran's 3.98 and Pittsburgh's 4.06.

 

Another interesting note. Aside from the two big pitcher's parks(Washington and SD), the bullpens with the best ERA's pitched the fewest innings.

[USSoccer]

That looks like a bullpen ERA around 4 - how is that so terrible? What was the bullpen ERA last year? Looks pretty optimistic on Williamson to me.

 

The Cubs pen had an ERA of 4.24 last year, tied for 9th in the NL. A 4.00 ERA would've put them alone in 8th last year, between San Fran's 3.98 and Pittsburgh's 4.06.

 

Another interesting note. Aside from the two big pitcher's parks(Washington and SD), the bullpens with the best ERA's pitched the fewest innings.

 

That's not suprising, considering when you have a lot of IP from your pen, it's usually because your starters stink, and middle relief isn't ever good.

[stitchface]

That looks like a bullpen ERA around 4 - how is that so terrible? What was the bullpen ERA last year? Looks pretty optimistic on Williamson to me.

 

The Cubs pen had an ERA of 4.24 last year, tied for 9th in the NL. A 4.00 ERA would've put them alone in 8th last year, between San Fran's 3.98 and Pittsburgh's 4.06.

 

Another interesting note. Aside from the two big pitcher's parks(Washington and SD), the bullpens with the best ERA's pitched the fewest innings.

 

That's not suprising, considering when you have a lot of IP from your pen, it's usually because your starters stink, and middle relief isn't ever good.

 

 

break it down by month - perhaps the bullpens wore out?

[Transmogrified Tiger]

That looks like a bullpen ERA around 4 - how is that so terrible? What was the bullpen ERA last year? Looks pretty optimistic on Williamson to me.

 

The Cubs pen had an ERA of 4.24 last year, tied for 9th in the NL. A 4.00 ERA would've put them alone in 8th last year, between San Fran's 3.98 and Pittsburgh's 4.06.

 

Another interesting note. Aside from the two big pitcher's parks(Washington and SD), the bullpens with the best ERA's pitched the fewest innings.

 

That's not suprising, considering when you have a lot of IP from your pen, it's usually because your starters stink, and middle relief isn't ever good.

 

Which begs the question, if you want to improve bullpen performance, are you better off improving the amount of innings your starters are able to go(whether it's via new players, lack of injuries, new training methods or philosphical approaches) or getting new relievers(especially given the wild variation in relief performances)?

[stitchface]

I think 4 isn't that terrible - the cubs need their starters to dominate or they're in trouble anyway. ie they need to be on that list of least used bullpens. I could see Wuertz, Williamson, and Ohman beating those numbers. Possbily also Novoa. I don't have a lot of confidence in Dusty but I think this pen comes together fairly naturally and could have some great roles.

[USSoccer]

That looks like a bullpen ERA around 4 - how is that so terrible? What was the bullpen ERA last year? Looks pretty optimistic on Williamson to me.

 

The Cubs pen had an ERA of 4.24 last year, tied for 9th in the NL. A 4.00 ERA would've put them alone in 8th last year, between San Fran's 3.98 and Pittsburgh's 4.06.

 

Another interesting note. Aside from the two big pitcher's parks(Washington and SD), the bullpens with the best ERA's pitched the fewest innings.

 

That's not suprising, considering when you have a lot of IP from your pen, it's usually because your starters stink, and middle relief isn't ever good.

 

Which begs the question, if you want to improve bullpen performance, are you better off improving the amount of innings your starters are able to go(whether it's via new players, lack of injuries, new training methods or philosphical approaches) or getting new relievers(especially given the wild variation in relief performances)?

 

I think it's all about starting pitching. You need to make sure you have a staff that is going to consistently go at least 6 IP. You have to avoid middle relief. Middle relievers are almost always the worst players on your team-they are almost all failed starters and innings eaters. If you have SP's who blow up, or aren't durable enough to go 6-7 IP, you have to use the middle relievers or overextend your setup men, which catches up to you by the summer.

 

Given that relievers are so up and down, it seems to be the smart way to build a staff would be to get durable, low BB SP's with solid WHIP and K/9 numbers. Spend money on that, and you can go cheap in the bullpen and get away with it. You can hide your middle relief and save them for mopup type outings, and fully and most effectively use your setup guys and specialists in situations that most favor you, based on matchups, which will also boost bullpen performance.

[snoodmonger]

I expect Lee to hit above .300 this year again. Not much--.315-.320 at the most, but I'd be surprised if he dropped down to .298.

[snoodmonger]

That looks like a bullpen ERA around 4 - how is that so terrible? What was the bullpen ERA last year? Looks pretty optimistic on Williamson to me.

 

The Cubs pen had an ERA of 4.24 last year, tied for 9th in the NL. A 4.00 ERA would've put them alone in 8th last year, between San Fran's 3.98 and Pittsburgh's 4.06.

 

Another interesting note. Aside from the two big pitcher's parks(Washington and SD), the bullpens with the best ERA's pitched the fewest innings.

 

That's not suprising, considering when you have a lot of IP from your pen, it's usually because your starters stink, and middle relief isn't ever good.

 

Which begs the question, if you want to improve bullpen performance, are you better off improving the amount of innings your starters are able to go(whether it's via new players, lack of injuries, new training methods or philosphical approaches) or getting new relievers(especially given the wild variation in relief performances)?

 

I think it's all about starting pitching. You need to make sure you have a staff that is going to consistently go at least 6 IP. You have to avoid middle relief. Middle relievers are almost always the worst players on your team-they are almost all failed starters and innings eaters. If you have SP's who blow up, or aren't durable enough to go 6-7 IP, you have to use the middle relievers or overextend your setup men, which catches up to you by the summer.

 

Given that relievers are so up and down, it seems to be the smart way to build a staff would be to get durable, low BB SP's with solid WHIP and K/9 numbers. Spend money on that, and you can go cheap in the bullpen and get away with it. You can hide your middle relief and save them for mopup type outings, and fully and most effectively use your setup guys and specialists in situations that most favor you, based on matchups, which will also boost bullpen performance.

 

I agree. The best thing you can do for a pen is have starters that last 6+. If you have that, you can skate with a top-notch closer, a good set-up guy, and an effective LOOGY. (Like that Wetteland/Rivera tandem in 1996)

[Rob]

Soooo... Pecota likes Ryan Howard

 

538 PA

 

.286/.375/.612

[Derwood]

so, in Value Over Replacement Player, what's the standard for the replacement player? You never see anyone with a negative VORP rating, so this "replacement player" must be horrible.

 

Seriously, I don't understand the formula. Someone explain it to me.

[Sarcastic]

so, in Value Over Replacement Player, what's the standard for the replacement player? You never see anyone with a negative VORP rating, so this "replacement player" must be horrible.

 

Seriously, I don't understand the formula. Someone explain it to me.

 

I thought Patterson had a negative VORP...

 

-10.6 according to BP. I have no idea how they figure the replacement level for each position, though.

[JGalt73]

so, in Value Over Replacement Player, what's the standard for the replacement player? You never see anyone with a negative VORP rating, so this "replacement player" must be horrible.

 

Seriously, I don't understand the formula. Someone explain it to me.

 

Below is a great explanation:

 

A replacement-level player is one who is "easily available" to any team--a AAA journeyman or end of the bench player. Replacement level is significantly below average--about 80% of average for the position. If you think of it in OPS terms, roughly 70 points of OPS below the average for the position is replacement level....

 

Pitchers "contribute" runs by preventing them from scoring. If replacement level is a 6.00 RA, and our star pitcher has a 3.50 RA over 180 innings:

 

RepLvl pitcher: 180 IP * 6.00 RA / 9 = 120 runs allowed

 

Star pitcher: 180 IP * 3.50 RA / 9 = 70 runs allowed

 

------------------

 

Compared to RepLvl, Star pitcher prevented 50 runs from scoring

 

Thus, his VORP is 50.0

 

Offensive VORP is measured in runs provided above those that would be gotten from this "replacement player". A negative VORP is Neifi Perez bad with the bat.

[Ender]

Maddux had a 69% strand rate last year which generally means the bullpen let him down. He also had a 15% HR rate on flyballs which is abnormally high, a little bit of that is probably the wind blowing out but not all of it. His LD% was 22% which is very high for him, he's usually in the 17-20 range. His xERA was 3.29 which is very strong. I think projection systems think he just had a bit of bad luck last year leading to the higher ERA. With a 1.22 WHIP and a 52% ground ball percentage his ERA should have been below 4.

 

Seeing stats like that two years in a row is scary though and its starting to look like a trend, however its not unheard of for a pitcher to have two bad luck years in a row and looking at the numbers behind his stats that appears to be what happened.

[CubinNY]

Maddux had a 69% strand rate last year which generally means the bullpen let him down. He also had a 15% HR rate on flyballs which is abnormally high, a little bit of that is probably the wind blowing out but not all of it. His LD% was 22% which is very high for him, he's usually in the 17-20 range. His xERA was 3.29 which is very strong. I think projection systems think he just had a bit of bad luck last year leading to the higher ERA. With a 1.22 WHIP and a 52% ground ball percentage his ERA should have been below 4.

 

Seeing stats like that two years in a row is scary though and its starting to look like a trend, however its not unheard of for a pitcher to have two bad luck years in a row and looking at the numbers behind his stats that appears to be what happened.

 

Nice analysis, but I think some of Maddux's more obscene numbers may be due to his advanced age. It seemed to me, and I could be wrong, that he left a lot of balls up in the zone last year. He also missed in the strike zone too (as did Prior). Overall, I think his numbers are pretty reflective of his actual performance.

 

I hope for some rebound this year, but I am not optimistic. At this point he is a six inning pitcher. We just have to hope he keeps his ERA around 4.

[sethuel1]

Maddux had a 69% strand rate last year which generally means the bullpen let him down. He also had a 15% HR rate on flyballs which is abnormally high, a little bit of that is probably the wind blowing out but not all of it. His LD% was 22% which is very high for him, he's usually in the 17-20 range. His xERA was 3.29 which is very strong. I think projection systems think he just had a bit of bad luck last year leading to the higher ERA. With a 1.22 WHIP and a 52% ground ball percentage his ERA should have been below 4.

 

Seeing stats like that two years in a row is scary though and its starting to look like a trend, however its not unheard of for a pitcher to have two bad luck years in a row and looking at the numbers behind his stats that appears to be what happened.

 

Nice analysis, but I think some of Maddux's more obscene numbers may be due to his advanced age. It seemed to me, and I could be wrong, that he left a lot of balls up in the zone last year. He also missed in the strike zone too (as did Prior). Overall, I think his numbers are pretty reflective of his actual performance.

 

I hope for some rebound this year, but I am not optimistic. At this point he is a six inning pitcher. We just have to hope he keeps his ERA around 4.

 

When Maddux hung em last year, he really hung em. A lot of solid contact. Very un-Maddux

[stitchface]

I just don't see his age having that huge negative effect. That being said, the only explanation I can offer is Questec. I don't see a big dropoff from Maddux coming but I don't expect a huge improvement either. Has his K rate changed substantially?

[T2E]

Soooo... Pecota likes Ryan Howard

 

538 PA

 

.286/.375/.612

 

Program glitch?

[Rob]

Soooo... Pecota likes Ryan Howard

 

538 PA

 

.286/.375/.612

 

Program glitch?

 

Maybe, but maybe not.

 

2002 - .280/.367/.460 (A)

2003 - .304/.374/.514 (A)

2004 - .297/.386/.647 (AA)

- .270/.362/.604 (AAA)

- .282/.333/.564 (in Philly)

2005 - .371/.467/.690 (AAA)

- .288/.356/.567 (Philly)

 

 

The kid has some very, very impressive power. And he just turned 26.

[TheDude]

Your starting five:

 

Maddux: 202 inn (30 starts), 3.77 ERA, 32.2 VORP

Prior: 196 inn (30 starts), 3.28 ERA, 41.5 VORP

Williams: 158 inn (27 starts), 4.71 ERA, 9.1 VORP

Wood: 100 inn (15 starts), 3.52 ERA, 22.4 VORP

Zambrano: 224 inn (33 starts), 3.29 ERA, 49.3 VORP

 

6th starter: Rusch 104 inn (12 starts), 4.56 ERA, 9.0 VORP

 

Those are certainly the staff numbers we've been looking for since 2003. Those are staff numbers capable of taking any division in baseball. I'm holding firm to my assertion that this team has to win the one-run games, likely 4-3 or 5-4, to have any chance this year.

 

Also, I've argued in several other threads that Maddux is better than Williams and will have a better year in 2006 than Williams (with much dispute from others), and it is nice to see several "professional" sources make the same prediction.

[goonys evil twin]

Also, I've argued in several other threads that Maddux is better than Williams and will have a better year in 2006 than Williams (with much dispute from others), and it is nice to see several "professional" sources make the same prediction.

 

I think that sources prediction was a bit unrealistic, expecting improvement out of a pitcher Maddux's age. But he's going to have a better year if only because he'll be given a spot all year long, while Williams will have to unfairly battle with Rusch for time.

[Bull]

Your starting five:

 

Maddux: 202 inn (30 starts), 3.77 ERA, 32.2 VORP

Prior: 196 inn (30 starts), 3.28 ERA, 41.5 VORP

Williams: 158 inn (27 starts), 4.71 ERA, 9.1 VORP

Wood: 100 inn (15 starts), 3.52 ERA, 22.4 VORP

Zambrano: 224 inn (33 starts), 3.29 ERA, 49.3 VORP

 

6th starter: Rusch 104 inn (12 starts), 4.56 ERA, 9.0 VORP

 

Those are certainly the staff numbers we've been looking for since 2003. Those are staff numbers capable of taking any division in baseball. I'm holding firm to my assertion that this team has to win the one-run games, likely 4-3 or 5-4, to have any chance this year.

 

Also, I've argued in several other threads that Maddux is better than Williams and will have a better year in 2006 than Williams (with much dispute from others), and it is nice to see several "professional" sources make the same prediction.

 

I have to belive that one of Prior, Wood or Z has a 2.xx ERA this year.

[TheDude]

Also, I've argued in several other threads that Maddux is better than Williams and will have a better year in 2006 than Williams (with much dispute from others), and it is nice to see several "professional" sources make the same prediction.

 

I think that sources prediction was a bit unrealistic, expecting improvement out of a pitcher Maddux's age. But he's going to have a better year if only because he'll be given a spot all year long, while Williams will have to unfairly battle with Rusch for time.

 

The key for Maddux is to keep that HR total down. The control is still there, but that HR total doubled from career norms in the 2003-2005 timespan.

 

[edit - I mean doubled per year, not doubled the career total]

 

BTW - Didn't the Bill James endorsed pitching prediction give Maddux a similar 2006 line?