FanPost

Comparing ZiPS 2012 Projections With Actual Mets Performance

Projection systems like ZiPS are fun especially this time of year when the only games played are in our head. But are they accurate? Well ZiPS didn't do too well with the 2012 Mets.

ZiPS missed badly on nine to eleven players depending on how you screen sample sizes and was very close on six players.

But at least when ZiPS missed badly it was more likely to underestimate Mets players than overestimate them. Eight Mets performed much better than ZiPS projected while three did much worse. All in all, 2012 ZiPS expected us to be a lot worse than we turned out.

I started with projected vs. actual pitchers ERA+ and position players OPS+. I subtracted one from the other and divided by the ZiPS projection to create a "pct. missed" metric. Since those ZiPS stats are league adjusted and use 100 as "average" they seems like the best available bet to attempt to measure some sort of percentage change. I suppose the other way to do it would have been to simply subtract the two numbers - but I think that would have measured the pct. difference to league average.

Here is a look at players who had a meaningful role on the 2012 club found on the BTF article/chart on ZiPS from November 3, 2011 (link to full chart below). The source chart appears to have been limited to "carryover" Mets - thus no Frankie Frank, Rauch etc.

Players 15%+ Better than ZiPS

AB's

Proj/Actual

Ops+

Proj/Actual

Stat

Proj/Actual

Pct. Diff.

David Wright

517/581

116/143

HR 19/21

+23%

Scott Hairston

260/377

97/117

HR 10/20

+21%

Jordany Valdespin

491/191

77/93

HR 9/8

+21% sss

IP

Proj/Actual

ERA+

Proj/Actual

BB/K

Proj/Actual

R.A. Dickey

179/233

102/140

45/114-54/230

+37%

Jon Niese

157.7/190

92/113

48/133-49/155

+23%

Matt Harvey

128.7/59.3

83/141

62/118-26/70

+70% sss

Dillon Gee

159/109

82/94

64/122-29/97

+15%

Bobby Parnell

72/68

103/155

30/70-20/61

51%

Players 27%+ Worse than ZiPS

AB's

Proj/Actual

Ops+

Proj/Actual

Stat

Proj/Actual

Jason Bay

430/194

107/47

HR 16/8

-44%

Josh Thole

421/321

93/63

BB 44/27

-32%

IP

Proj/Actual

ERA+

Proj/Actual

BB/K

Proj/Actual

Johan Santana

91/117

108/79

24/69-39/111

-27%

Players within 10% of ZiPS

AB's

Proj/Actual

Ops+

Proj/Actual

Stat

Proj/Actual

Lucas Duda

504/401

108/98

HR 17/15

9%

Ruben Tejada

528/464

83/90

Bb/k 47/82-27/73

8%

Ike Davis

361/519

113/110

HR 15/32

3%

Daniel Murphy

370/571

108/102

HR 8/6

6%

Kirk Nieuwenhuis

437/282

90/91

HR 11/7

1%

IP

ERA+

BB/K

Chris Young

32.3/115

92/93

4/15-36/80

1%

When ZiPS missed - it really missed. ZiPS "closest miss" understated Dillon Gee's ERA+ by 15% but ZiPS ten other whiffs were by 21% or more.

Of those ten "big miss" players you could argue that Jordany Valdespin and Matt Harvey had samples smaller than you'd like. I'd keep Harvey in because even if he'd doubled his innings pitched he would have had to have posted an ERA+ of about 30 over those 60ip to get him within 15% of the ZiPS projection.

But ZiPS didn't always get it wrong. It absolutely nailed Chris Young (1%), Kirk Niewenhuis (1%) and Ike Davis (3%) and was reasonably close for Dan Murphy (6%), Ruben Tejada (8%) and Lucas Duda (9%).

Of course, ZiPS projections include an estimate of playing time. Few ZiPS playing time estimates bore much resemblance to reality. I don't know enough about ZiPS to understand if projections would have changed if they'd been give different playing time numbers.

Some other positive milestones that were unexpectedly reached from ZiPS 2012 perspective:

  • It gave Ike Davis 0% chance of hitting 30 HRs - he hit 32. I would have thought he'd be 1 of only 2 guys they gave any chance of reaching that mark.
  • Gave Wright only a 9% chance of reaching 140 ops+ - he reached 144.
  • Gave Dickey a 10% chance of reaching 130 ERA+ - he hit 140 & a 3% chance of k'ing more than 8 per 9ip, he k'd 8.9.
  • Gave Niese a 40% chance of reaching 100 ERA+ - he posted 113.

So if 2013 proves to be much like 2012, ZiPS needs to be taken with a huge grain of salt. Depending on how you want to handle sample sizes, ZiPS missed badly enough to be virtually worthless on 9-11 players while only being reliable for six.

I guess that's why they play the games on the field instead of on paper.

The projections for last season can be found here: http://www.baseballthinkfactory.org/oracle/discussion/2012_zips_projections_-_new_york_mets/


This FanPost was contributed by a member of the community and was not subject to any vetting or approval process. It does not necessarily reflect the opinions, reasoning skills, or attention to grammar and usage rules held by the editors of this site.

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join Amazin' Avenue

You must be a member of Amazin' Avenue to participate.

We have our own Community Guidelines at Amazin' Avenue. You should read them.

Join Amazin' Avenue

You must be a member of Amazin' Avenue to participate.

We have our own Community Guidelines at Amazin' Avenue. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9351_tracker