BP Releases PECOTAs for Pirates Hitters

by Charlie Wilmoth on Feb 19, 2009 9:01 PM EST

Baseball Prospectus has published PECOTA projections for hitters (subscription only). It wouldn't be right to leak details, but here's the cheat sheet:

-P- Nate McLouth stands a very good chance of matching last year's production and even has an outside shot at being one of baseball's best position players this year. PECOTA, which has always been a big believer in McLouth, thinks he's for real.

-P- Andy LaRoche will be fine. He won't be a star, but he'll be a serviceable starter.

-P- Nyjer Morgan projects to be about eight miles below replacement level as a left fielder.

-P- Brian Bixler should hit better than Morgan.

-P- Jason Jaramillo will be a totally functional backup catcher and should be clearly better than Robinzon Diaz.

Tweet Comment 33 comments | 0 recs |

Do you like this story?

Comments

My problem with PECOTA

It is completely worthless.

While I am no fan of Morgan, it has absolutely nothing to do with his PECOTA projection. Projections are a bad joke.

by haven on Feb 19, 2009 9:11 PM EST reply actions

Time will tell.

I’m with PECOTA on this one, though.

by Vlad on Feb 19, 2009 9:43 PM EST up reply actions

Does it like Salazar at all?

Just curious.

by Vlad on Feb 19, 2009 9:43 PM EST reply actions

It loves him, although I’m pretty sure it has him projected as a Diamondback, which changes things. The equivalent translations still look pretty tasty, though. I’m really rooting for him to make the team.

by Charlie Wilmoth on Feb 19, 2009 9:46 PM EST up reply actions

You know.....

I hope Salazar and/or Morgan are useful. Emphasis on the “and”. I am a long term pessimist. I doubt either will be useful. But I still don’t care about a projection for either of them. Just seems completely ridiculous.

by haven on Feb 19, 2009 10:45 PM EST up reply actions

I’m looking at his numbers on the Pirates depth chart, presumably adjusted for PNC, and he has better slash stats than everyone but McLouth and Adam LaRoche.

Also, his PECOTA card predicts a huge lefty/righty split. According to them, he’s a good bet to slug 100 points higher vs. RHP than vs. LHP. It doesn’t like Monroe at all, even against lefties.

by OlStubbleBeard on Feb 20, 2009 7:40 AM EST up reply actions

Haven...

What’s the matter with you?

You don’t like PECOTA projections?

I thought that I was the only person on this blog that ridiculed those projections; but, nevertheless, I’m happy to have you join the club.

by thegunner on Feb 20, 2009 12:03 AM EST reply actions

BP Projections

This is the first year I’ve subscribed to BP, so I can’t speak much to the overall accuracy of the projections.

But this article here makes reference to how PECOTA did predicting last year’s top 3 players in each league’s ROY voting. It’s actually pretty impressive. It nails Ellsbury, and Alexei Ramirez, does very well on Longoria (but underestimates some power), and nails Votto in the NL. Soto’s projection was also very close. The only one they missed on was Jurrjens.

I would think the rookie projections are even harder than more established major leaguers, but as I mentioned, I can’t speak to the overall accuracy year in and year out.

I certainly wouldn’t ridicule them, though.

by lost_zero on Feb 20, 2009 2:17 AM EST up reply actions

Accuracy of PECOTA predictions

I plead ignorance but how accurate have these predictions been historically (I realize there is no foolproof predicting method)?

I tried looking it up online and found a lot of statistical arguments but I got a generous C in statistics and was wondering if someone could give me a simple answer about it?

This is the link I found:

http://www.baseballprospectus.com/article.php?articleid=2515

I made most of my life decisions at a Foghat concert... I stand by them.

by Chester J Lampwick on Feb 20, 2009 12:49 AM EST reply actions

See here, as well as the Wikipedia entry on PECOTA, which contains a number of other links. It seems to be the best projection system, and it soundly beats other stats-oriented projections and projections by non-statistically-oriented folks as well. I don’t fully understand the skepticism about it, except to say that I think a lot of people who react negatively do it don’t really understand it or think that it purports to do something it doesn’t. (Well, maybe on the BP annuals it occasionally does purport to do something it doesn’t, but that’s the only place.) Nate Silver, the creator, would agree with a lot of the criticisms that are leveled against it, because he doesn’t see it as this absolutely accurate system but rather as a very helpful tool to deal with a lot of uncertainty. In fact, all the uncertainty is built into the system, in that it weighs the probability that a player will break out or collapse, rather than just giving a single set of numbers.

For people who don’t like it, I’d say, show me something better.

by Charlie Wilmoth on Feb 20, 2009 2:27 AM EST up reply actions

Agreed

I’ve seen several comments from Silver and the other BP writers where they compare their own feelings about a guy compared to PECOTA and say “well, PECOTA doesn’t know this, so I think it is underestimating this or that….” For example, if someone is playing with an injury, PECOTA doesn’t “know” that, per se. It just sees that performance as part of the historical record of an individual’s performance.

The thing I like about it is it seems to do a better job of accounting for regression to the mean than most people ever allow themselves to believe. There was a long discussion on onlybucs today about Nyjer Morgan and other players average with RISP. PECOTA would probably view those ABs as a small sample size and the future performance to move close to that player’s overall history of produciton. Whereas we mere mortals could be inclined to chalk some of it up to “clutchiness” or “unclutchiness”.

And, yes, unclutchiness is totally a valid word.

by lost_zero on Feb 20, 2009 2:35 AM EST up reply actions

think that it purports to do something it doesn’t

Or, rather, think that it purports to do something it doesn’t actually purport to do. Sorry, poor writing on my part.

by Charlie Wilmoth on Feb 20, 2009 3:24 AM EST up reply actions

I have a lot of respect for Nate Silver...

based on his work at fivethirtyeight.com. I’m curious, can you give us some hint what “serviceable” means regarding LaRoche this year? Is it the just a little better than Jose Bautista numbers that seem to keep cropping up?

by mocasdad on Feb 20, 2009 6:39 AM EST up reply actions

Thanks

That was helpful. Should have learned by now to always turn to wikipedia first.

I made most of my life decisions at a Foghat concert... I stand by them.

by Chester J Lampwick on Feb 20, 2009 11:43 AM EST up reply actions

For people who don’t like it, I’d say, show me something better.

It isn’t that there is something better. It is trying to project someone’s exact BA/OPB/SLG/SB/VORP and the like just seems to have no value to me. Say Nyjer Morgan isn’t good. Base it on past performance. Trends. Any sort of analysis you want. But I see no real value in saying his PECOTA projection for 2009 is .214/.308/.333. I made that up of course…… I just don’t see the value with specific projections in a general sense because there are way to many variables. But if this is the way people want to spend their time then so be it……

by haven on Feb 20, 2009 4:02 PM EST up reply actions

What’s the difference between predicting somebody’s slash stats and predicting it will be 41 degrees tomorrow? Yeah, the specificity is pretty useless, but the specificity isn’t the point. As Charlie remarked, PECOTA and its creator are well aware that it is dealing with probabilities, not specifics.

by matskralc on Feb 20, 2009 6:16 PM EST up reply actions

Matthew Berry at ESPN thinks only two Bucs will perform above fantasy value this season.

I’m not trying to champion Nyjer Morgan. However, I believe he does get a particularly bad rap on this site.

I believe Nate McLouth and Brandon Moss will start the majority of games in two of the outfield positions for the Pirates in 2009.

So what does third outfielder Morgan need to do to be considered a success in 2009?
An OBP of .350, whilst scoring 100 runs combined with swiping 50 bases in 75 attempts?

And what OPS/HR/RBI totals would Pearce/Hinske/Salazar/Monroe have to produce to top that as far as win shares go?

by RDV across the sea on Feb 20, 2009 10:55 AM EST reply actions

Runs scored are a team-dependent statistic.

So that wouldn’t necessarily have much to do with the quality of Nyjer’s performance. And 50 SB with 25 CS is basically a break even net value for baserunning – he needs to put up a SB% that’s at least in the 70s in order to have much leg value.

As a CF or benchie, if Morgan could manage a .700 OPS, then he’d be having a better-than-expected season. The bar for starting LF is hella high, though, and I think he’d need to either have a huge power spike or bat something like .330 in order to keep his head above the waterline there. And unfortunately, I don’t see either one happening.

by Vlad on Feb 20, 2009 11:29 AM EST up reply actions

for fantasy purposes

Nyjer Morgan might give you +SBs, but he’s going to give you —HR and —RBI (compared to a replacement level player in almost any fantasy league). I don’t think he hurts your batting average, but he shouldn’t get enough at bats to help it either even if he hits well (btw, he won’t). He won’t help your runs either (I don’t think any Pirate scores even 95 this year).

by wickethewok on Feb 20, 2009 11:49 AM EST up reply actions

Btw, “—” is supposed to be a “minus-minus”.

by wickethewok on Feb 20, 2009 11:49 AM EST up reply actions

What I'm trying to get at Vlad, is what do the Bucs have that is better than Nyjer Morgan?

Let’s assume that Morgan and McLouth are a hybrid LF/CF position. NH has stated on numerous occasions that LF at PNC Park takes CF skills. McLouth has the better skills, Morgan has slightly better athleticism. Anyway for this argument’s sake, let’s assume both can handle either position.

Let’s give Moss credit and assume he does OK in RF (doubles and OBP especially) and becomes a fixture.

Which player and with what statistics can supplant Morgan if he has an OBP of .350 and steals 70% of bases he tries?

Does any body have win share statistics for this kind of player against players with better power numbers?

by RDV across the sea on Feb 20, 2009 12:01 PM EST up reply actions

I think the main argument against Morgan is that he is playing rather than Steve Pearce who could potentially have an impact on some future Pirates teams.

But since management is already determine for Pearce to start in AAA and play everyday, I don’t really have a problem with Morgan playing some games until guys come up from the minors as the season progresses.

I made most of my life decisions at a Foghat concert... I stand by them.

by Chester J Lampwick on Feb 20, 2009 12:33 PM EST up reply actions

I'm at work, so I don't have WS numbers on me.

But I think the odds of Morgan posting a .350 OBP and a 70% SB rate are very low. His minor league career suggests that he’s probably more of a .320 or .330 guy on true talent in the majors. So even if that number is better than what we’d get from Pearce or Salazar (and it might or might not be), it still doesn’t necessarily tell us much about who should be starting in LF.

I’d love to see him do it. I just don’t think the odds are in his favor.

by Vlad on Feb 20, 2009 12:45 PM EST up reply actions

Salazar..

is a “nice” player who will perform adequately, but he is no more than a platoon player who should only face RH pitching. Moss and Pearce/Monroe should likewise be platooned based on the other team’s starting pitcher.

by thegunner on Feb 20, 2009 3:33 PM EST up reply actions

A Pearce/Moss platoon would make some sense.

I wouldn’t bench anybody for Monroe, though. He’s awful.

by Vlad on Feb 20, 2009 4:05 PM EST up reply actions

"I’m not trying to champion Nyjer Morgan. However, I believe he does get a particularly bad rap on this site."

Charlie’s disdain for Nyjer Morgan is about as bad as Jennifer Aniston being a jilted lover.

I got it the 1st time….it seems he goes out of his way to diss Morgan. Still like the site though and believe that Morgan is a situational pinch-runner/5th OFer at best.

by BadAndy on Feb 20, 2009 12:39 PM EST up reply actions

The problem is . . .

. . . Nyjer Morgan has hit 3 HR in his last 4 full seasons (minor leagues included). And he’s being penciled in as a starting corner OF on a purported ML team, a position normally requiring a power bat. (Plus, I’d bet you some of those 3 HR where inside the park jobs.)

Since you mention Jennifer Aniston, I feel I have to point out that Nyjer Morgan’s odds of hitting a ball out of a ML park in 2009 are statistically pretty much the same as my odds of having sex with Jennifer Aniston in 2009. That can’t be good for either Morgan or me (it maybe very good news to Jeniffer Aniston, though).

Plus he can’t run the bases, has no technique when it comes to steals (he slides right over the bag on his way to left field), gets no jump on the ball in the OF, throws to the wrong base, and Ks more than 2.5 times as often as he walks.

He is the epitome of a Jim Tracy kinda ballplayer. Looks good, so he must be good & don’t confuse me with facts.

I have come to grips (sadly) with the fact that nothing Steve Pearce can do in spring training will give him a chance to compete for a starting OF job. Hopefully, Hinske will be allowed to compete — in any fair competition he will wipe Morgan off the map, just like McLouth did to Mogan last spring. If not, my backup hope is that Morgan is just a pawn in Huntington’s chess match, and is just supposed to keep LF warm for Cutch to come up at the All Star break.

by WstCstBucco on Feb 20, 2009 3:26 PM EST up reply actions

don't be so hard on yourself

Nyjer Morgan’s odds of hitting a ball out of a ML park in 2009 are statistically pretty much the same as my odds of having sex with Jennifer Aniston in 2009.

nyjer had 160 chances last year to hit a hr in the big leagues. how many times were you even in the same ballpark as ms. aniston?

by johnnycuff on Feb 20, 2009 3:51 PM EST up reply actions

Baserunning

That’s really one big thing about Morgan that people forget. He’s fast, but he’s a pretty terrible base runner. In the minors he’s got 234 steals in 314 attempts, which is almost exactly 75, which is about the break even point (I think) . In the majors he’s only 16 of 24, which is 66.7. We’re talking about a guy who’s only tool is speed, and his speed isn’t even particularly useful. So what’s the point of having him around?

http://www.whygavs.com
http://mlb.fanhouse.com

by whygavs on Feb 20, 2009 4:29 PM EST up reply actions

ugh

Almost exactly 75%, that should say.

http://www.whygavs.com
http://mlb.fanhouse.com

by whygavs on Feb 20, 2009 4:30 PM EST up reply actions

The link for above Berry article at ESPN is:

http://sports.espn.go.com/fantasy/baseball/flb/story?page=mlbdk2k9tmrlovehate

by RDV across the sea on Feb 20, 2009 11:00 AM EST reply actions

If Morgan starts in LF...

this team WILL lose 100 games. He might be marginally useful if he and Nate switched spots…at least Morgan’s usual numbers would play better in CF than LF.

However…he has very little skill in the outfield…just speed. There’s a reason he’s not playing CF…and it’s his defense. Poor routes to fly balls…and an arm that doesn’t even measure up to Jason Bay’s when he had shoulder problems all season.

Put bluntly…the more Nyjer Morgan plays this season…the less chance there is of Cutch or Steven Pearce actually appearing in a Pirates uniform before September, if at all. As soon as Morgan goes to the bench…he becomes a glorified pinch runner. I just pray that there is some stupid GM (if one exists not named Littlefield) that takes Nyjer off our hands during spring training or early in the season. Nyjer Morgan WILL NOT be on the team when the consecutive losing seasons streak ends.

by Thunder on Feb 20, 2009 9:43 PM EST reply actions

His numbers would play better in CF...

…but Nate’s would play worse by the same amount in LF, so the net offensive gain for the team from that move would be negligible. Both guys are still getting the same number of PAs in the same lineup positions.

Basically, you’d have to either think that Nyjer is a better CF defender than Nate, or that it’d be easier to slot McCutchen into CF if he’s displacing Nyjer.

by Vlad on Feb 23, 2009 9:41 AM EST reply actions