ZiPS projected record for the Orioles with Jimenez+Cruz: 78-84

[Frobby]

There's why Machado is projected to drop from 6.4 to 3.0 WAR. They don't think he'll even play until the summer.

Hard to say if they gave him a low PA projection because of his injury, or whether their computer just looked at him having 202 PA in 2012 and 710 last year and came up with 455 as an average. Probably the former, but I don't know.

[Going Underground]

Zips cleaners project Oriole record?

I think the record will depend on the health of Manny. Also,not a lot os speed in the lineup. I project 86 but I am always wrong.

[Orioles4Life21]

ZiPS should zip it.

[DrungoHazewood]

I think I'll come up with my own projection system that applies the same set of assumptions across the whole league. Then tweak/rig it until the Orioles are projected to have 90+ wins and a playoff birth. I would totally rock and it would be the best and most accurate system ever and Fangraphs would totally list it at the top of their projections.

[waroriole]

Hard to say if they gave him a low PA projection because of his injury, or whether their computer just looked at him having 202 PA in 2012 and 710 last year and came up with 455 as an average. Probably the former, but I don't know.

Considering the average of 202 and 710 is 456, it could just as easily be the latter. If that's the case, their credibility is basically zapped.

[skanar]

something that I've always questioned about.

In BB-ref's breakdown, both oWAR and dWAR include the positional adjustment. When combining the two, one of the positional adjustments must be removed.

[weams]

My guess is this projection is heavily based on two things.

Low OBP of the offense.

Low K rate of the pitching.

And error?

[waroriole]

Considering the average of 202 and 710 is 456, it could just as easily be the latter. If that's the case, their credibility is basically zapped.

Damn it. I totally whiffed there. There credibility would be Zip.

[weams]

I hate that WAR has become a stat people follow.

They get Paid to get the WAR. It has a monetary value.

<iframe width="420" height="315" src="//www.youtube.com/embed/387ZDGSKVSg" frameborder="0" allowfullscreen></iframe>

[CA-ORIOLE]

In BB-ref's breakdown, both oWAR and dWAR include the positional adjustment. When combining the two, one of the positional adjustments must be removed.

Not sure I understand. I assumed each positional adjustment/contribution was pro-rated (but not additive).

[Tony-OH]

Anyone know what ZIPS projected the Orioles in 2012 and 2013?

[tntoriole]

This kind of "statistical analysis" always makes my old head hurt. But can someone explain to me how a projection system like ZIPS comes up with the projection that Michael Ohlman is going to play 93 games for us this year and hit .262 with 8 home runs? or Michael Almanzar at 136 games played, .255 average and 15 home runs? It just sounds nonsensical on its face and may be useful to fantasy game players, but it is hard to see how this has much relevance to the actual decisions that are likely to be made by Buck and DD.

http://www.fangraphs.com/projections.aspx?pos=all&stats=bat&type=zips&team=2&players=

[skanar]

Not sure I understand. I assumed each positional adjustment/contribution was pro-rated (but not additive).

I'm not 100% on this, but I believe the idea is that oWAR can be compared between positions, subject to the full positional adjustment - so an equal performance by a SS and a RF get different oWAR by the full difference between their positions. Likewise for dWAR.

So you have oWAR = (offensive production + positional adjustment) and dWAR = (defensive production + positional adjustment). If you were to simply add them, you'd get (off prod + def prod + 2*positional), so you have to remove one positional adjustment from the sum.

[CA-ORIOLE]

I'm not 100% on this, but I believe the idea is that oWAR can be compared between positions, subject to the full positional adjustment - so an equal performance by a SS and a RF get different oWAR by the full difference between their positions. Likewise for dWAR.

So you have oWAR = (offensive production + positional adjustment) and dWAR = (defensive production + positional adjustment). If you were to simply add them, you'd get (off prod + def prod + 2*positional), so you have to remove one positional adjustment from the sum.

Yeah, I don't see how that is workable (or at least accurate), especially if a guy plays multiple positions. I'd have to assume that it is apportioned over all of the positions. You can compare oWAR to oWAR, but some guys value may be run down by playing some positions poorly. Same with dWAR. I always assumed this to be a weakness with WAR in that it didn't provide same allowance for defensive flexibility ..... which arguably can provide a team value.

[skanar]

Anyone know what ZIPS projected the Orioles in 2012 and 2013?

2012: 69-93

2013: 79-83

Here's the win differential between actual # of wins and the ZiPS prediction last year:

Boston: +15

Oakland: +12

Cleveland: +12

Baltimore: +6

Kansas City: +5

Tampa: +4

Detroit: +3

Texas: +2

New York: +1

Seattle: +0

Minnesota: +0

Astros: -8

Chicago: -13

Los Angeles: -15

Toronto: -15

Looks like they were far off for 6 teams and reasonably close for the rest. Astros were predicted to have 59 wins and actually had 51; it's hard to fault a projection system for not predicting just how bad one of the worst seasons ever would be.

Edit: OK, I read my source a little more carefully, and these are a bloggers' additions of the ZiPS projected WAR values, not Szymborski's official projected standings. Those are behind the ESPN Insider paywall. Presumably, any differences should be small.