Your Official Edwin Jackson Versus Andy Sonnanstine Forecast Post

by R.J. Anderson on Jun 3, 2009 4:07 PM EDT Comment 64 comments

People keep talking about it, so using PECOTA's comparables from 1980 onwards, and dumping the extremely small sample sizes, I compared the 23-28 year-old projected career paths of Edwin and Sonnanstine. Hoping to see if there's anything worthwhile here. Let's take a look, first at Sonnanstine. The comparables are:

Player	23	24	25	26	27	28	29
Doug Drabek	4.11	3.91	3.62	3.76	3.28	3.36	2.97
Jose Lima	4.99	3.83	4.09	3.76	6.12	5.63	5.29
Esteban Loaiza	4.74	5.87	4.03	5	3.93	4.7	4.51
Charles Hudson		3.5	3.52	4.14	4.65	4.34	3.91
Bill Laskey		3.46	4.21	4.15	4.65	5.99	2.11
Rick Aguilera	3.28	3.59	3.83	3.86	2.44	2.99	3
David Bush		4.15	4.75	3.98	4.57	4.93
Brad Penny	3.39	4.52	3.92	3.59	3.64	3.89	3.63
Josh Fogg			4.86	5.08	4.61	5.13	5.03
Moose Haas	4.49	3.56	3.51	3.3	3.51	3.59	3.89
Chad Ogea	5.4	4.22	4.71	4.42	4.98	6.16
Oil Can Boyd	3.8	3.42	3.58	4.17	5.24	5.18	4.61

Here's a look at the average of those versus Sonnanstine to date:

Average	4.275	4.002727	4.0525	4.100833	4.301667	4.6575	3.895
Andrew Sonnanstine		4.26	3.91

So far, Sonny has had one year of topping his comps and one year below the comps. Moving forward, his real "rough year" is projected to take place during the 2011 season. Odds are, he's not a starter with the Rays then, and probably a reliever. Otherwise, things look fine. He profiles exactly as you would expect.

Now, on to Edwin.

The comparables:

Player	23	24	25	26	27	28	29
Llivan Hernandez	5.22	4.36	3.91	4.37	3.91	3.87	4.01
Steve Trachsel	4.51	5.09	4.81	4.69	4.72	4.68	4.92
Mac Suzuki	4.89	5.72	5.04	5.8	5.2
Ryan Dempster	4.36	4.45	4.72	5.1		3.38	3.79
Jason Schmidt	4.7	4.13	4.17	4.45	4.74	3.64	3.11
Luis Leal	4.8	3.44	4.06	4.04	4.23	5.42
Dave Weathers	3.27	4.72	4.82	4.77	5.1	3.34	4.79
Joe Mays	4.82	4.77	4.27	4.92	5.35		5.03
Don Aase	3.94	4.29	4.02	3.33	3.76	3.13	3.28

PECOTA threw a ton of old-age comparisons my way. We do get a few modern cases though, so that's nice. Here's how Edwin stacks up versus the average:

Average	4.501111	4.552222	4.424444	4.607778	4.62625	3.922857	4.132857
Edwin Jackson	4.9	4.88

Above his average, but not terribly so. This is the year that the averages dip down by a tenth of a run, which is interesting since Edwin has improved by more like a full run. After this year though, he forecasts to shoot back up for two years, before settling into a pretty good pitcher.Of course, by then he's already a free agent and on his next team most likely. Gil Meche part two?

Here's how these two paths stack up visually:

Sonnyvsedwin_medium

So that's it. Sonnanstine's path is supposed to arch up until he hits 28, then those who survived shot back down. Edwin is supposed to bounce around until age 28 when he has a great year. Do players always follow these paths? Absolutely not. The fact that Edwin was an outlier and reached the majors at age 20 suggests he isn't following a path often traveled. But, this is my way of adding something to that debate. If nothing else, it takes some of the "What if we forecasted him wrong..." talk and adds historical examples to it.

If someone wants to do a more wide scale project with this, feel free to. Links to the PECOTA pages:

Jackson

Sonnanstine

I used June 30th as the cutoff date and FanGraphs FIP. So someone born on 5/5/1980 would have the 2009 season count as his 29-year-old season.

Anyways, have at it.

5 recs | Comment 64 comments | Share

Story-email Email Printer Print

Comments

Meant to include this:

Comparable Players are the backbone of a player’s PECOTA. Only the twenty best comparables are listed here, but as many as 100 players may be used in the generation of his forecast if they are sufficiently comparable.

PECOTA compares each player against a database of roughly 20,000 major league batter seasons since World War II. In addition, it also draws upon a database of roughly 15,000 translated minor league seasons (1997-2006) for players that spent most of their previous season in the minor leagues. (When minor league comparables are used, they appear in ALL CAPS). PECOTA considers four broad categories of attributes in determining a hitter’s comparability:

1. Production metrics—such as batting average, isolated power, and unintentional walk rate for hitters, or strikeout rate and groundball rate for pitchers.

2. Usage metrics, including career length and plate appearances or innings pitched.

3. Phenotypic attributes, including handedness, height, weight, career length (for major leaguers), and minor league level (for prospects).

4. Fielding Position (for hitters) or starting/relief role (for pitchers). PECOTA doesn’t require that a comparable hitter play the same defensive position; it is a factor that is evaluated along with many others, and assigned a relatively substantial weight. Consideration is also given to the ‘similarity’ between two positions; for example, a shortstop will be compared to a second baseman before he is compared to a left fielder. In most cases, the database is large enough to provide a meaningfully large set of appropriate comparables. When it isn’t, the program is designed to ‘cheat’ by expanding its tolerance for dissimilar players until a reasonable sample size is reached.

PECOTA glossary

by R.J. Anderson on Jun 3, 2009 4:13 PM EDT reply actions 0 recs

why can't i post a pic in making a new fanpost?

wtf. and OT fanposts are still cool, right?

by daveh33 on Jun 3, 2009 4:22 PM EDT reply actions 0 recs

Did you hit the pic button?

And yes.

by R.J. Anderson on Jun 3, 2009 4:25 PM EDT up reply actions 0 recs

that's the thing.

there was no pic button.

just link.

by daveh33 on Jun 3, 2009 4:56 PM EDT up reply actions 0 recs

ok

that was weird.

by daveh33 on Jun 3, 2009 5:07 PM EDT up reply actions 0 recs

All those charts won't make Sonny throw more than 86

by Raymondo on Jun 3, 2009 5:17 PM EDT reply actions 0 recs

Even when you may have a legitimate point...

you do yourself a disservice by turning it into a ridiculous one-liner that is only going to piss people off and devolve the debate

by RaysTheRoof on Jun 3, 2009 5:21 PM EDT up reply actions 0 recs

Not my intention at all

I’ve said all along, i realize Sonny has the better numbers, but i don’t think trading a young arm, never been hurt, who has an upper 90s FB and a plus slider, AND is only 25 yo is very smart

by Raymondo on Jun 3, 2009 5:37 PM EDT up reply actions 0 recs

Age doesn't matter as much for pitchers as it does hitters.

So I’m not sure “young arm” and “25 yo” really matters. The other stuff is reasonable.

by R.J. Anderson on Jun 3, 2009 5:39 PM EDT up reply actions 0 recs

Just saying, it's not like he's 30

I love Joyce, but i’m afraid Jackson might finally live up to the billing

by Raymondo on Jun 3, 2009 5:43 PM EDT up reply actions 0 recs

PECOTA uses a lot of metrics to identify comparables

However, since PitchFx has only been around a little while, it doesn’t take into account things like velocity, break on pitches, release points, etc. I think in the future (i..e 20 years down the road) we could get more accurate predictions for pitchers and even hitters once HitFx is available. I just generally think it is difficult to predict young pitchers using only results stats (i.e. FIP) as opposed to process stats (i.e. PitchFx). For these reasons, I suggest PECOTA may be a too optimistic on Sonny and too pessimistic on EJax.

by RaysTheRoof on Jun 3, 2009 5:17 PM EDT reply actions 0 recs

Well, that assumes BPro actually tweaks PECOTA.

Silver has all but left them, and they seem more content with selling the brand than bettering their metrics. That’s why WARP/VORP have been broke for years and still are broke.

I definitely agree the velocity would add a nice dimension to it. I wonder if the James/Neyer Guide to Pitchers lists enough of these guys to get an estimated average velocity.

by R.J. Anderson on Jun 3, 2009 5:24 PM EDT up reply actions 0 recs

Yeah BPro has not been impressive since Silver left.

Most of the really cutting edge stuff is probably being done in house by teams (like the rays) and will remain proprietary for a long time. There is almost certainly more incentive (financially) for these modeling guys to work for a team than to produce models for a free site (fangraphs, BTB) or even a paysite (BPro, etc). Further, I imagine that the sense of satisfaction in helping assemble a team and seeing it win can not be topped by writing and modeling for a public audience.

by RaysTheRoof on Jun 3, 2009 5:30 PM EDT up reply actions 0 recs

Using career velos and whatever I could find in the J/N Guide:

Player Velocity
Doug Drabek 88-92
Jose Lima 87-89
Esteban Loaiza 87-89
Charles Hudson N/A
Bill Laskey High 80s
Rick Aguilera Low 90s
David Bush 88-90
Brad Penny 92-94
Josh Fogg 86-88
Moose Haas N/A
Chad Ogea 87-91
Oil Can Boyd Low 90s
Andrew Sonnanstine 86-87

Working on Edwin now.

by R.J. Anderson on Jun 3, 2009 5:37 PM EDT reply actions 0 recs

And:

Player Velocity
Livan Hernandez 84-86
Steve Trachsel 86-88
Mac Suzuki 89-91
Ryan Dempster 91-93
Jason Schmidt 92-94
Luis Leal N/A
Dave Weathers 87-89
Joe Mays 87-89
Don Aase N/A
Edwin Jackson 92-94

by R.J. Anderson on Jun 3, 2009 5:42 PM EDT up reply actions 0 recs

Interesting.

Dempster/Schmidt compare with Edwin’s velocity, and both hit it big at 28/29.

Edwin’s 25, next year 26, year after 27. His first year after being a free agent will be 28.

So even if we kept him, you really shouldn’t have expected a breakout until he left us. Huh.

by R.J. Anderson on Jun 3, 2009 5:44 PM EDT up reply actions 1 recs

Are you saying we'll never be able to sign our own FA?

by Raymondo on Jun 3, 2009 5:52 PM EDT up reply actions 0 recs

Why would we re-sign Edwin, who figures to be worse than Price, Davis, and Hellickson and cost way more?

Same with Sonnanstine.

Kazmir/Shields re-signing made sense because they were really good.
Garza may make sense in a few years.

The back of the rotation guys? No. We’ve got arms like Talbot, Rollins, and Barnese/Moore to consider down the road as well.

by R.J. Anderson on Jun 3, 2009 5:58 PM EDT up reply actions 0 recs

I don't know about worse than Hellickson or even Davis

A lot of pitchers have dominated A and AA and been atrocious in the majors. Hellickson’s stuff is very iffy from what I read.

by RaysTheRoof on Jun 3, 2009 6:00 PM EDT up reply actions 0 recs

Per Kiley:

Fastball -55/55

Worked 91 to 93 and kept the velocity late into the outing. Worked off of this pitch traditionally, getting ahead by using it aggressively early in the count and then going to the breaking ball for K’s, but would mix in the changeup to keep hitters on their toes. Flashed above-average life and command of the pitch, mostly as late run in on the hands of right-handers. Also showed some solid sink at times. Hellickson’s approach with the fastball was very impressive for a young guy with a smaller frame; to come after hitters with a big boy’s fastball and approach.

Curveball -50/55

His breaking ball is a little tough to figure out, but I identified it as a 73-75 mph loopier early-count curve with 10-to-4 slider action and a hard 77-79 mph curve that had sharper downward bite and was more of an 11-to-5 break.

The slower version had flatter break and was thrown exclusively to the arm side, indicating Hellickson wasn’t following through, making it more of a get-me-over, show pitch.

The harder version was used more often and was a late-count chase pitch that was frequently buried and was easily above-average when on. The pitch would back up and come out flat a few too many times, so the feel isn’t completely there, but over half of them were late, sharp, and overmatched FSL hitters.

Changeup – 50/55

The changeup was also advanced, and elicited some weird swings as well. It is solidly average right now and flashed above with late fade and depth with excellent deception. Like the curveball, the changeup isn’t quite there yet as he wasn’t getting both the fade and depth consistently and sailed way out of the zone at times, but the deception and potential is there, and when it’s in the zone, Hellickson again is hitting spots. It was a clear third pitch by usage, mostly to keep hitters off of expecting all fastballs before the hard curve comes out.

Combined with above average control and questionable command. Sounds like a harder throwing Sonnanstine.

And I really, really hope Davis is better than a 4.8 FIP for most of his time here.

by R.J. Anderson on Jun 3, 2009 6:03 PM EDT up reply actions 0 recs

I think Ejax will be posting lower than a 4.8 FIP over the next few seasons

but I digress, what do the ratings for the pitches mean (50/55)? What is this out of? Is the split for righties and lefties? Sorry not familiar with all of the scouting lingo.

by RaysTheRoof on Jun 3, 2009 6:07 PM EDT up reply actions 0 recs

20-80 is the scale.

So 50/55 means the pitch rates as a “50” in velocity and movement now, and projects to mature to a 55. Remember, he was in A-ball when Kiley saw him, so he wasn’t a finished product.

by R.J. Anderson on Jun 3, 2009 6:08 PM EDT up reply actions 0 recs

So he has three pitches that might be +

but not that project to ++? Fastball sounds better than Sonny’s though.

by RaysTheRoof on Jun 3, 2009 6:11 PM EDT up reply actions 0 recs

Basically.

by R.J. Anderson on Jun 3, 2009 6:14 PM EDT up reply actions 0 recs

Talbot and Rollins are Sonny-like

by Raymondo on Jun 3, 2009 6:06 PM EDT up reply actions 0 recs

I see little issue here.

by R.J. Anderson on Jun 3, 2009 6:07 PM EDT up reply actions 0 recs

When's the GDT going up?

by Raymondo on Jun 3, 2009 6:10 PM EDT up reply actions 0 recs

Seems you have every viable prospect succeeding

how likely is that.

On projections, as the data set increases, it likely becomes more reliable, but what happens when the latest data varies strongly, one way or the other, from the prior data? Cause this is what we’re currently seeing. I suspect some significant change in those projections if ‘09 performance stays in same general vicinities for each these guys. An analysis restricted to ’07 and ’08 data will inherently favor Sonny. Perhaps one can’t adjust projections mid-year, but all it means is that they may well have been wrong as their actual career arcs develop.

Trading E-Jax wasn’t the wrong decision at the time – sometimes things just don’t go as projected or planned. And it may never be the wrong decision, depending on the trade principles career. The point made by trade defenders is just this – and that is correct. But I sense an underlying unwillingness to accept that something the FO did turned out poorly. Happens, accept that too.

by nyyfaninlaaland on Jun 3, 2009 6:58 PM EDT up reply actions 0 recs

well reasoned and fair comment.

by RaysTheRoof on Jun 3, 2009 7:12 PM EDT up reply actions 0 recs

How likely was Edwin suddenly becoming a run+ better?

by R.J. Anderson on Jun 3, 2009 9:05 PM EDT up reply actions 0 recs

As I said

the trade made sense at the time, and I didn’t even mention the added issue of salary impacts going forward which strongly reinforces the point.

But how likely his improvement was isn’t an issue. It may not continue, he may get better. Joyce may explode, settle in at his minor league ranges (still good), or be a bust. How they perform over the near term, likely confined to their tenures in their teams, is how folks will judge. But you can’t say it was a bad deal when made, and you can’t say we should have given them Sonny, because that might not have been an option.

You could say that it might be easier for Jax to learn new pitches than Sonny to learn to throw 95.

by nyyfaninlaaland on Jun 4, 2009 2:27 AM EDT up reply actions 0 recs