Jump to content

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


skanar

Recommended Posts

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.

I don't know all the logic behind it. Here's the relevant bb-ref page.

Under Version 2.1, May 2012, the 5th bullet reads:

Converted Offensive WAR from afWAR back to oWAR. Note that oWAR + dWAR now double counts position, so adding them will not give WAR.

Which I interpret as I described in the last post. And yes, I think the positional adjustment is a weighted average over playing time across multiple positions. Not sure if that's done by innings or PA, but I'd have to assume innings.

You can't look at oWAR think it measures just 'performance as a hitter,' unfortunately. Luckily, Fangraphs' "Off" metric does exactly that.

Consider JJ Hardy and David Lough. Both hit about the same last season (Lough .724, 96 OPS+/Hardy .738, 97 OPS+). By "Off," Lough was -0.5, which means half a run worse than the average MLB hitter. Hardy was -3.5, or 3.5 runs worse (difference due to Hardy's extra playing time). The "Off" metric ignores position, so Hardy gets no credit for playing SS and Lough isn't hurt by playing RF/LF. The value is in runs, so 3 is fairly small - equivalent to about 0.3 WAR.

By oWAR, which includes the positional adjustment, the story is different. Lough had 1.0 oWAR last season; Hardy had 2.6. That's a big difference, and it's due to their respective positions.

Of course, both Lough and Hardy add positive defensive value that isn't accounted for in oWAR or Off, since the point of each stat is to look at offense only.

I think both stats can be valuable as long as the user/writer is clear on what they mean. Takes into account that a .730 SS is more valuable than a .730 RF, irrespective of their defense. The other looks only at production at the plate and ignores all other factors.

Link to comment
Share on other sites

  • Replies 71
  • Created
  • Last Reply
The last thing I have ever been called is a homer and this projection smells like doo. No way, this team has solid depth. Only missing piece is a closer IMO. We hope that turns out better than we are expecting currently.

I am curious what they have Flaherty and Markakis at. I think both of them improve a good bit.

I tend to agree. But looking at Webb... I'm wondering if he might not be our answer there. Looking at his stats, it appears he has dominated at times over the past four years. He's a pretty giant man. Assuming he has some mound presence. I hear he's a ground ball guy. Who knows?

But 78 wins is low, IMO. That was my number BEFORE the signings of Jimenez and Cruz. Since those signings, I've given them another 6 wins (not just from the 2 players' performance - but by what they do to the gestalt of the team).

Link to comment
Share on other sites

I don't know all the logic behind it. Here's the relevant bb-ref page.

Under Version 2.1, May 2012, the 5th bullet reads:

Which I interpret as I described in the last post. And yes, I think the positional adjustment is a weighted average over playing time across multiple positions. Not sure if that's done by innings or PA, but I'd have to assume innings.

You can't look at oWAR think it measures just 'performance as a hitter,' unfortunately. Luckily, Fangraphs' "Off" metric does exactly that.

Consider JJ Hardy and David Lough. Both hit about the same last season (Lough .724, 96 OPS+/Hardy .738, 97 OPS+). By "Off," Lough was -0.5, which means half a run worse than the average MLB hitter. Hardy was -3.5, or 3.5 runs worse (difference due to Hardy's extra playing time). The "Off" metric ignores position, so Hardy gets no credit for playing SS and Lough isn't hurt by playing RF/LF. The value is in runs, so 3 is fairly small - equivalent to about 0.3 WAR.

By oWAR, which includes the positional adjustment, the story is different. Lough had 1.0 oWAR last season; Hardy had 2.6. That's a big difference, and it's due to their respective positions.

Of course, both Lough and Hardy add positive defensive value that isn't accounted for in oWAR or Off, since the point of each stat is to look at offense only.

I think both stats can be valuable as long as the user/writer is clear on what they mean. Takes into account that a .730 SS is more valuable than a .730 RF, irrespective of their defense. The other looks only at production at the plate and ignores all other factors.

Thanks. The highlighted part was really my point. I'd assume innings at each position is pro-rated. Your other points about oWAR sound correct. It's tricky.

You can't look at oWAR think it measures just 'performance as a hitter,' unfortunately. Luckily, Fangraphs' "Off" metric does exactly that.

Agreed, I typically just go to wRC+ (although that does include a speed factor). Fangraphs probaly does do a better job with the breakdowns.

Link to comment
Share on other sites

ZiPS probably doesn't think Jimenez and Cruz add more than 2-3 wins than what we would've had otherwise (and it might be right).

This team could win anywhere between 75 and 95 games and I wouldn't be surprised.

I will be very surprised if the O's finish below .500. I see them as a 90 win team on paper.

Link to comment
Share on other sites

I will be very surprised if the O's finish below .500. I see them as a 90 win team on paper.

Depends on what you choose to write on the paper.

I can write on mine, "No starting pitcher in 2013 with an FIP below 4.00. Two lowest ERAs outperformed their FIPs by 3/4 of a run. Just added a starting pitcher whose ERA hovered around 5.00 for several years before the last couple months of last year. Bullpen as volatile as any other. Two biggest offensive contributors had freakish, unprecedented seasons that absolutely cannot be predicted to occur again without more data. The younger one has a longer track record of knee injuries than major league success. Every single returning player of the offensive core (plus Markakis) pulled off a miracle and stayed off the DL and the team still only won 85 games."

And it's not hard for me to see 75 wins. After writing all that out, it feels easier to see 75 than 95, but I still think we'll be alright.

Link to comment
Share on other sites

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=

ZiPS projects major league equivalencies, not actual major league performance. In other words, a 4.00 ERA in Norf

Link to comment
Share on other sites

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=

ZiPS projects major league equivalencies, not actual major league performance. In other words, a 4.00 ERA in Norfolk might equate to a 5.50 ERA in Baltimore. So the projection for a minor league pitcher who is capable of 4.00 in Norfolk will be 5.50, regardless of where he actually plays, and the innings projection is for total innings at all levels, majors and minors.

Link to comment
Share on other sites

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

Funny how the top four teams that outperformed their ZIPS projections have ties to Moneyball.

Link to comment
Share on other sites

ZiPS projects major league equivalencies, not actual major league performance. In other words, a 4.00 ERA in Norfolk might equate to a 5.50 ERA in Baltimore. So the projection for a minor league pitcher who is capable of 4.00 in Norfolk will be 5.50, regardless of where he actually plays, and the innings projection is for total innings at all levels, majors and minors.

Thank you. I learn so much from you all!!

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...