Updated PECOTA: Orioles will finish 15 games out

[El Gordo]

I'd like to get on another topic, i.e., what the heck is wrong with what BP is doing with PECOTA this year? Last year, the average AL team had a .763 OPS and scored 781 runs. This year, BP projects that the average AL team will have an OPS of .790 (27 points higher than in 2009) and yet the average runs scored will be 757 (24 runs lower than in 2009). That is just nonsensical.

Looking at the most current Oriole projections, BP thinks BRob will bat 1st, play 90% of the time, and post an .834 OPS. It thinks Jones will bat 2nd, play 80% of the time, and post an .850 OPS; with Pie posting a .786 the other 20% of the time. It thinks Nick Markakis will bat 3rd, play 90% of the time, post an .877 OPS, and hit 24 homers. And yet, they think Markakis will have only 90 RBI. That makes no sense whatsoever. I guarantee you if the O's top 3 hit like that, Markakis will knock in 115-120 runs.

I don't know why PECOTA is coming up with such high OPS numbers and such low run totals. Neither of those things make sense by themselves, and together they make no sense at all.

Maybe they think Tejada will bat #2 and hit into a gazillion DP's, so most of Nick's hits come with 2 outs and the bases empty.:mwahaha:

[Lucky Jim]

I could be wrong, but here's my take:

Yes. There could be reason to expect this. For instance, PECOTA bases its projections on historical comps, and as a result is entirely extrinsic in its measure: it attempts to gauge "true talent" by locating it in a nexus of similar stats and attributes as they've occurred historically.

With young players, and fewer data points, this is almost always going to be somewhat problematic. First, something like PECOTA is always going to normalize anomalous production. The extrinsic odds that Nolan Reimold is Ben Broussard are always going to be greater than the odds that Reimold is Jason Bay. But if Reimold's true talent is Jason Bay, then he should exceed his PECOTA. PECOTA pegs his as, essentially, reproducing 2009. I would modify Drungo's point to say that we tend to believe that "good players get better." If Reimold is, as we believe, a "good player," then we should expect him to outperform a PECOTA that has to figure he's more likely to be Ben Broussard.

Maybe I'm misreading the way the statistics should be used. I'm a complete novice, so that's a far greater probability than Reimold being Jason Bay.

I'd just like to note that you can disregard this post, as I wasn't looking at the correct Reimold numbers. His OPS will stay similar, but it IS projecting 24 HRs from him.

[now]

So you don't think allowing.75 fewer runs would be "any better?". I'd call that significant progress.

Interesting to note that the above would represent half the gain in wins over last year's Oriole team (75 runs = 7.5 wins = half of 15-game improvement). In other words, half of our projected improvement comes from pitching/defense, half from offense. Seems about right.

[Leitch]

It's going to be pretty hard for them to compete with a sub $60 million dollar payroll in the future...

[cindyluvsbrady]

BP has updated its projected standings based on its PECOTA projections:

BOS 94-68

NYY 92-70

TBR 91-71

BAL 79-83

TOR 71-91

http://www.baseballprospectus.com/fantasy/dc/

A season like that would be pretty interesting for the O's all the way through August. Heck, even September would be interesting, with the O's playing a lot of games that month vs. the 3 teams vying for the playoffs.

By the way, I still feel these projections have a strong disconnect between OPS and runs scored.

They are wrong:angryfire:

No one listen to them:rolleyestf:

BELIEVE

[Crazysilver03]

Might want to take a look at this thread...

It's going to be pretty hard for them to compete with a sub $60 million dollar payroll in the future...

They might have one more year, but then they are done.

I "think" you are wrong.

The Rays jettison $40M at the end of 2010 with the expiration of contracts belonging to Crawford, Pena, Soriano, Burrell, and possibly Navarro. All of those players could be replaced with players coming up in their minors in the next season or two.

[srock]

I thought Palmer was tall, lanky and has a nice pitching motion. I also was mad at him because he spent nearly a minute explaining why he didn't have time to sign my autograph while walking into old RFK stadium before a game with the Senator whereas the pitcher walking just ahead of him (John Miller) never said a word but scribbled his John Hancock on a piece of paper for me. Ultimately, I got a guy's autograph who never made it in the majors and a bunch of baloney from a future HOFer. How's that for an answer?:laughlol:

This the greatest Old#5fan Grandpa Simpson story of them all. Totally did not see this coming.

[Maverick2143]

This the greatest Old#5fan Grandpa Simpson story of them all. Totally did not see this coming.

<object width="560" height="340"><param name="movie" value="http://www.youtube.com/watch?v=zs8BAbigSLM&hl=en_US&fs=1&"></param><param name="allowFullScreen" value="true"></param><param name="allowscriptaccess" value="always"></param><embed src="http://www.youtube.com/watch?v=zs8BAbigSLM&hl=en_US&fs=1&" type="application/x-shockwave-flash" allowscriptaccess="always" allowfullscreen="true" width="560" height="340"></embed></object>

[ShaneDawg85]

This the greatest Old#5fan Grandpa Simpson story of them all. Totally did not see this coming.

But the important thing is that I had an onion tied to my belt. Which was the style at the time.

[RShack]

I'd like to get on another topic, i.e., what the heck is wrong with what BP is doing with PECOTA this year? Last year, the average AL team had a .763 OPS and scored 781 runs. This year, BP projects that the average AL team will have an OPS of .790 (27 points higher than in 2009) and yet the average runs scored will be 757 (24 runs lower than in 2009). That is just nonsensical.

Looking at the most current Oriole projections, BP thinks BRob will bat 1st, play 90% of the time, and post an .834 OPS. It thinks Jones will bat 2nd, play 80% of the time, and post an .850 OPS; with Pie posting a .786 the other 20% of the time. It thinks Nick Markakis will bat 3rd, play 90% of the time, post an .877 OPS, and hit 24 homers. And yet, they think Markakis will have only 90 RBI. That makes no sense whatsoever. I guarantee you if the O's top 3 hit like that, Markakis will knock in 115-120 runs.

I don't know why PECOTA is coming up with such high OPS numbers and such low run totals. Neither of those things make sense by themselves, and together they make no sense at all.

I don't know any inside details, but my theory is that some absentee owner bought the thing, and whoever is supposedly minding the store is not doing an attentive and capable job. As for how much of it is due to the operational people not giving a damn as much as a guy whose name is on it would vs. just being incompetent, well, I have no theory about that. But it seems like they're releasing sloppy first drafts without being thorough, like one might expect of slacker college students who haven't yet learned what doing a proper job means.

[Arthur_Bryant]

I don't know any inside details, but my theory is that some absentee owner bought the thing, and whoever is supposedly minding the store is not doing an attentive and capable job. As for how much of it is due to the operational people not giving a damn vs. being incompetent, well, I have no theory about that. But it seems like they're releasing sloppy first drafts without being thorough, like one might expect of slacker college students who haven't yet learned what doing a proper job means.

This is the first year Nate Silver has not been involved. He sold the system to BP several years ago, I think, but he stayed in on the process until this past fall. I can hardly believe that the BP guys would drop the ball their first time flying solo, but it looks that way.

[TonySoprano]

What did PECOTA predict for 2009? 2008 while we're at

it? I'm trying to gauge how accurate they have been before I put any stock in their latest numbers.

[Frobby]

What did PECOTA predict for 2009? 2008 while we're at

it? I'm trying to gauge how accurate they have been before I put any stock in their latest numbers.

PECOTA was way off base for the O's last year -- 821 runs if I recall. I have a spreadsheet showing all the major systems' 2009 projections at home and will post later if I remember. It suffices to say that the O's underperfomed every projection system out there last year, by a wide margin.

[Orioles Magic]

BP has updated its projected standings based on its PECOTA projections:

BOS 94-68

NYY 92-70

TBR 91-71

BAL 79-83

TOR 71-91

http://www.baseballprospectus.com/fantasy/dc/

A season like that would be pretty interesting for the O's all the way through August. Heck, even September would be interesting, with the O's playing a lot of games that month vs. the 3 teams vying for the playoffs.

By the way, I still feel these projections have a strong disconnect between OPS and runs scored.

No way we come close to this kind of record. PECOTA can't account for the O's annual September/October swan dive. Two to three 7+ game losing streaks guaranteed. Just like death and taxes.

[JLowenstein]

What did PECOTA predict for 2009? 2008 while we're at

it? I'm trying to gauge how accurate they have been before I put any stock in their latest numbers.

"Overall, though, the PECOTA projections have tended to get more and more accurate over time. The average error in our won-loss projections since 2003 has been as follows:

2003 5.91 wins

2004 7.71 wins

2005 5.14 wins

2006 4.94 wins

2007 4.31 wins"

"[PECOTA's] average miss on each team's win total [in 2008] was 8.5 games."

This is their 2009 forecast.