Comparing Dodger Pitching Projections for 2009

by meercatjohn on Feb 16, 2009 10:06 AM PST Comment 15 comments

We haven't had much of a debate about the subject but over at Dodger Thoughts they have been discussing our projected pitching staff over and over. A selected few think our staff is in big trouble but they are shouted down by the many. So taking their cue I've assembled the projections for what I think are the three most respected projected systems. ~~These are all pay sites so I'm only going to post the ERA.~~ Per Paul Scott below let me make this clear. THT is a free site that charges $10 for the projections to be downloaded. HQ and BP are full pay sites. For that reason I'm only going to show the ERA projections and not the underlying statistics they used to come up with projections.

[Update] I've added the IP to get a better idea of where the numbers are coming from.

You have The Hardball Times which brought you x-FIP with the most negative projections. Their x-FIP darling Troncoso does not fare well. Pecota from Baseball Prospectus seems to sit in the middle. The good news is that my favorite site for projections, Baseball HQ has a very optimistic forecast for us.

The crazy outliers are Cory Wade for THT compared to the other two. James McDonald is liked by HQ but hated by THT and Pecota. I've been a HQ believer in the pitching projections all this decade and they have done me well in my roto battles. I myself was a bit worried about our rotation and while Wolf/Schmidt do not inspire much confidence I'm relieved to see that HQ thinks our big three are really a big three.

FirstName	LastName	THT	HQ	Pecota
Jonathan	Broxton	2.82/73	2.95/58	2.83/65
Hong-Chih	Kuo	3.35/77	3.62/87	3.43/70
Chad	Billingsley	3.75/174	3.28/203	3.55/180
Hiroki	Kuroda	4.15/173	3.9/189	4.16/175
Clayton	Kershaw	4.3/150	3.78/174	4/155
Yhency	Brazoban	4.31/42	4.8/	/
Randy	Wolf	4.79/160	4.6/174	4.5/130
Travis	Schlichting	4.84/64	/	5.41/60
Scott	Elbert	4.85/52	3.27/44	4.63/45
Jason	Schmidt	4.98/78	4.34/58	4.88/65
Claudio	Vargas	4.98/108	5.28/29	4.6/80
Ramon	Troncoso	5/67	3.89/44	4.3/40
Cory	Wade	5.38/76	3.41/58	3.92/50
Eric	Stults	5.59/146	4.97/29	4.8/55
James	McDonald	6.29/131	3.89/44	5.11/70
Charlie	Haeger	4.96/157	5.32/44	5.85/40
Guillermo	Mota	4.29/62	4.5/58	4.52/50

0 recs | Comment 15 comments | Share

Story-email Email Printer Print

Comments

THT

is not, and never has been, a pay site.

by Paul Scott on Feb 16, 2009 10:32 AM PST reply actions 0 recs

It is not a pay site

Sure, you can buy their book (completely separate from their site – all of which is free) or download a portion thereof (what you are talking about for $10), but calling THT a “pay site” in the same way as BP or HQ is simply not fair.

by Paul Scott on Feb 16, 2009 10:43 AM PST reply actions 0 recs

re:

The post is related to projections and the projections are $10. So is not, and never has been is an incorrect absolute.

by meercatjohn on Feb 16, 2009 10:45 AM PST up reply actions 0 recs

This is not that hard

For BP and HQ (and ESPN) you go to their site and you can get some web-based content for free, others sections and articles require a subscription.

You cannot “subscribe” to THT and 100% of their web-based content is free. Yes, you can buy a few books from them each year and you can also buy, separately, a portion of their preview book as an excel file. But that’s it. No portion of their website is restricted from access on condition of payment.

The sentence “Baseball Prospectus, Baseball Headquarters, and The Hardball Times are all pay sites.” is very clearly not correct. THT is completely different from the other two sites (and from ESPN).

Perhaps all you meant to say was “access to each organizations projections is a for fee, and therefor I am only printing part of of it.” That is fine and is probably all you meant. But it is absolutely not fair to THT to lump them in with BP and HQ by stating “[t]hese are all pay sites…”

by Paul Scott on Feb 16, 2009 10:59 AM PST reply actions 0 recs

re:

Maybe you can represent THT when they sue me for false misrepresentation. They can pay you the $10 I just sent them for the projections.

by meercatjohn on Feb 16, 2009 11:13 AM PST up reply actions 0 recs

I Want Some

of what they are drinking at HQ. Oh wait…I have already partaken.

by Bob Hendley on Feb 16, 2009 12:11 PM PST reply actions 0 recs

re:

It is the kool-aid of choice for optimistic Dodger fans.

by meercatjohn on Feb 16, 2009 12:19 PM PST up reply actions 0 recs

HQ has Elbert with a 3.27 ERA?

I assume they think he’ll be a reliever then, because thats beyond optimistic if its a starter projection.

by Brendan Scolari on Feb 16, 2009 12:59 PM PST reply actions 0 recs

re:

Yes, both Elbert and McDonald have < 50 innings projected.

by meercatjohn on Feb 16, 2009 1:22 PM PST up reply actions 0 recs

Is Tron Tron projected as a starter?

by Bob Hendley on Feb 16, 2009 1:24 PM PST up reply actions 0 recs

Ok thanks.

That makes more sense.

by Brendan Scolari on Feb 16, 2009 7:12 PM PST up reply actions 0 recs

I should probably combine

the ERA/IP together because some of the low ERA’s are a direct result of fewer innings. THT has McDonald with 131 innings thus the 6.29 ERA. HQ manually tweaks their numbers and it is obvious that THT does not. McDonald would never get 131 inning if he was pitching that poorly.

Troncoso has 69 inning per THT, 44 innings per HQ, and 40 innings per Pecota.

by meercatjohn on Feb 16, 2009 1:27 PM PST reply actions 0 recs

What does it matter?

I guess I could pay for these sites (and $10 contribution to THT), to get the answer, but what does the number of innings matter for the projecting of performace, other than defining whether one is a SP or reliever? As you say, no way James is put out there for 131 innings if he is cunnythumbing to the tune of 6.29 ERA. Are these parts of a larger model?

by Bob Hendley on Feb 16, 2009 4:17 PM PST up reply actions 0 recs

If you're a reliever you're ERA will be lower than if you were a starter.

The IP help to make more sense of the projections like McDonald’s and Elbert’s.

by Brendan Scolari on Feb 16, 2009 7:14 PM PST up reply actions 0 recs

Just wanted to add

ZIPS, from BBTF, is free.

by StolenMonkey86 on Feb 16, 2009 8:10 PM PST reply actions 0 recs

plus

MARCELS, CHONE and OLIVER (and all three are accessible through Fangraphs or their originating site and freely downloadable excel/cvs.)

PECOTA is the only truly different projection system, with the heart of its engine being comparables. All others use some sort of [year span] numbers regressed to [some mean]. At least one (BBHQ) starts with a regression, but then employs highly questionable subjective input. MARCELS is one of the simplest. None-the-less, the free projection systems come very close (and sometimes beat) the for-pay ones in correlation to actual results.

As the free projection systems are becoming more numerous, someone should employ a WOC model and average all freely available models and see how that tracks.

by Paul Scott on Feb 17, 2009 9:57 AM PST up reply actions 0 recs