Batter's Box Interactive Magazine Batter's Box Interactive Magazine Batter's Box Interactive Magazine
Dan Szymborski has posted the 2007 ZIPS projections for Toronto.  The projections, both for the pitchers and the hitters look quite favourable.  Can you guess which hitter has the best projection?


The offensive projections are about as one would have guessed, with good production out of the outfield, the corners and DH leading to probably about 830-850 runs.  Surprises include Russ Adams having a better offensive projection than Aaron Hill.  Alex Rios' projection is very cautious, treating his improvement last year as mostly year to year variation.  If ZIPS is right, the best hitter on the club in 2007 will be Lyle Overbay.

The news on the pitching front is even better, if ZIPS is right.  Among the starters, the club has a top flight ace, an excellent #2, and four candidates for the #3-#5 spots with above league average projections.  Josh Towers, 4.70 ERA, is an interesting one. The pen is as good or better, according to ZIPS, with Ryan, League and Frasor forming an excellent top 3.  All told the projections would suggest a runs allowed figure in the 700-720 range.

In other words, this looks like a legitimate playoff contender to ZIPS.

2007 Blue Jays ZIPS projections | 51 comments | Create New Account
The following comments are owned by whomever posted them. This site is not responsible for what they say.
Pistol - Thursday, January 11 2007 @ 10:44 PM EST (#161837) #
In other words, this looks like a legitimate playoff contender to ZIPS.

Zip-a-dee-zoo-da!

Rios' projection is at his career level.  I think he'll be a lot better than that.  I'm not sure why, but projection models don't like Aaron Hill.  I don't understand why his numbers should drop at 25.  The rest of the hitters are about what I would expect.

The pitcher projections seem pretty optimistic.  It has Thomson, Towers, Janssen, Chacin and Marcum all under 5 ERAs.

If you offered the Jays this type of performance in 2007 right now I think they'd take it in a second.  It'll be interesting to see how these projections translate into wins.
Craig B - Thursday, January 11 2007 @ 10:49 PM EST (#161838) #
I'm not sure why, but projection models don't like Aaron Hill

I suspect that, in general, his other stats don't support the notion that he's a high-average hitter.  High-average hitters usually have significant other offensive skills, which Hill hasn't shown yet.

Anyway, all these are very satisfying.  Note that the Blue Jays have six starters projected to be above-average, as well as Jamie Vermilyea and Davis Romero around average (as starters) with Shaun Marcum not too far off.
John Northey - Thursday, January 11 2007 @ 11:28 PM EST (#161839) #
Interesting to dig through the numbers...

1B/DH
Average player - 283/359/498
Overbay - 301/374/491
Thomas - 255/370/522 but just 92 games played

2B
Average player - 278/335/421
Hill - 268/330/371
Adams - 260/330/380
Roberts - 244/316/382

3B
Average - 278/344/465
Glaus - 249/353/501
Hattig - 246/299/359
Cosby - 250/286/397 - lets hope Glaus stays healthy

SS
Average - 277/329/418
Clayton - 263/315/351
Smith - 240/284/404
McDonald - 245/296/304 - as expected, a sinkhole position
Adams - 260/330/380

LF
Average - 282/352/477
Johnson - 286/346/423
Stairs - 253/337/441
Lind - 296/350/466 - listed as having a range from 272/319/415 to 314/372/515 next season (!)

CF
Average - 275/337/437
Wells - 285/344/501

RF
Average - 279/348/472
Rios - 282/333/444

CA
Average - 267/327/419
Zaun - 242/347/382
Phillips - 236/288/351
Thigpen - 243/315/366

I get a strong feeling we'll be seeing Johnson traded or put into a limited 4th outfield role by mid-season and Lind going to LF. 


Pitching?
Likely Starters ERA
Halladay - 3.20
Burnett - 3.63
Thomson - 4.47
Towers - 4.70
Chacin - 4.77

Prospect Starters
Vermilyea - 4.67
Romero - 4.70
Janssen - 4.71
Rosario - 4.94
Marcum - 4.94
Litsch - 5.05
Taubenheim - 5.27
McGowan - 5.29
Yates - 5.29
Banks - 5.42
Romero - 5.50

A few others are listed as well.  If we see anything like these numbers though the Jays will be playing in October I'm betting.  4.84 is the league average starter projection

The Pen - 4.31 is projected league average ERA
Ryan - 2.05
League - 3.38
Accardo - 3.69
Frasor - 3.77
Downs - 4.33
Tallet - 5.44

Other potential relievers
Blaine Neal - 4.60
Roney - 4.81
Kemp - 4.81
Machi - 4.94

Figure some kid who should be a future starter will get the 7th pen slot, pitching long relief and the like.  Probably will be rotated throughout the season to give as many kids a taste of the bigs until/unless one of them steps up and does well.

So, for the likely 11 man pitching staff only Tallet will be clearly worse than league average according to ZiPS with Downs on the edge for relievers. 

Good for helping us all dream about 2007.
danjulien - Friday, January 12 2007 @ 12:18 AM EST (#161840) #
I played around with the projections and the Lineup Analysis Tool

Jays projected lineup

The team would score 5.237 runs per game!!  The team should use Overbay as a leadoff hitter (has that not been suggested by a Bauxite?)  And Wells as a third hitter doesn't seem to mix...

More Findings are here
China fan - Friday, January 12 2007 @ 12:59 AM EST (#161841) #
    Forgive my lack of reverence for computer projections, but what is the value of a computer model that reaches a conclusion that the Jays have EIGHT above-average starting pitchers?   Including prospects who've barely pitched in the majors?   If these projections had any connection to reality, JP must have been crazy to have offered any money to Lilly, Meche, Thomson or anyone else.  But obviously JP doesn't think that the Jays have eight above-average starting pitchers, and I suspect that he's not a total idiot.  When a computer model has no connection to common sense, what is the point of it?  I know I'll get jumped-on by the stat-heads for saying this, but I fail to see any value in such projections.  Someone please explain to me the value of this exercise.
danjulien - Friday, January 12 2007 @ 01:45 AM EST (#161842) #
I agree...these proections don't seem quite even to the Jays situation but they're fun and they're predictory...it's a system using years of data...but in the end if computers played the games...we'd all have Xbox controllers.  Just enjoy the fun while we wait for spring training
ayjackson - Friday, January 12 2007 @ 02:53 AM EST (#161843) #

Figure some kid who should be a future starter will get the 7th pen slot, pitching long relief and the like.

Rosario is out of options, so I'd imagine that is who will get the spot.  The Chief's starting rotation will look lights out with McGowan, Janssen, Taubenheim, Ramirez and either Towers or Marcum starting the year there.

Paul D - Friday, January 12 2007 @ 09:33 AM EST (#161844) #
Quick comment for John and others.

Despite the fact that it appears as though ZIPS is projecting playing time (for Thomas and others), ZIPS does not project playing time.   So feel free to ignore that stuff.

Ryan Day - Friday, January 12 2007 @ 09:37 AM EST (#161845) #

 I'd take it.

 Granted, I have problems with any system that suggests Adams will be better than Hill (and I like Adams), and the pitching seems a bit too optimistic. Although in that respect, I like and agree with the general idea that the Jays have a bunch of guys who could be above average if a few things break well. I wouldn't bet much money on it, but those projections for Thomson and Towers aren't massively optimistic - they're both in line with relatively recent performances - and if the Jays got just one of them, they'd be in pretty good shape.

Chuck - Friday, January 12 2007 @ 09:42 AM EST (#161846) #

Despite the fact that it appears as though ZIPS is projecting playing time (for Thomas and others), ZIPS does not project playing time. 

I believe that ZIPS projects playing time with regards to injury. Clearly it pays no regard to merit.

Mike Green - Friday, January 12 2007 @ 10:01 AM EST (#161847) #
5.2 runs per game is equivalent to 842 runs in a season.  And I just eyeballed it!

For what it's worth, here are the projections that I disagree with, wearing "the objective analyst" rather than "Blue Jay fan" hat:

Too positive- Lyle Overbay, A.J. Burnett, Jeremy Accardo, John Thomson, Josh Towers, Casey Janssen
Too negative- Aaron Hill, Alex Rios, Shaun Marcum

The overly positive projections for Overbay, Burnett, Accardo and Thomson likely result from a failure to weight adequately the league differences in evaluating prior NL performance.  Thomson's projection is grossly off due to the failure to account for arm injury properly on top of this. Janssen's projection is modestly too good probably because of inadequate weight applied to the weak offences he faced in 2006.  Towers' projection is modestly too good likely because the weighting between 2005, 2006 and career performance is not right; he should be projected at about his career norm.

Hill's projection assumes (as PECOTA wrongly did last year) that he is a punch and judy hitter.  That is simply not correct, and a careful understanding of his double A and triple A performance would help with that.  His career BABIP of .312 is entirely consistent with his power, speed and line drive rate.  There was significant objective evidence that Rios' performance last year was an age 25 leap forward, rather than a fluke.  He changed from a ground ball hitter to a fly ball hitter and pulled the ball much more.  Given his anthropomorphics and age, there is no reason objectively to anticipate a significant decline in power between 2006 and 2007. Marcum's situation is the converse of Janssen's- his projection is modestly too poor likely because of inadequate weight applied to the strong offences he faced in 2006. 

Aside from the Thomson projection (which is more likely his 15% optimistic than his 50%), all of the projections are within a reasonable range. 

Paul D - Friday, January 12 2007 @ 10:10 AM EST (#161848) #
I believe that ZIPS projects playing time with regards to injury. Clearly it pays no regard to merit.

I'm pretty sure that even this isn't true.  ZIPS does not predict anything regarding playing time. 

I'll head over to BTF and see if I can find Szym and ask him.
Mike Green - Friday, January 12 2007 @ 10:13 AM EST (#161849) #
To answer Chinafan's question, the value of this exercise is significant.  The objective analysis shows that the Jays had a number of pitchers whose reasonable projection was equivalent to or better than Meche's, but not as good as Lilly's.  As you can see from the above, that does not mean that the Jays do have 7 above-average starters ready to go.  The objective analysis also substantiates the view of many that the major weakness of the club is at shortstop rather than at starting pitcher. 

Dan Szymborski - Friday, January 12 2007 @ 10:43 AM EST (#161851) #
- ZiPS takes into account injury history to the extent that past games played is accurately reflective of a player's injury history.  If a pitcher has started 12, 14, and 17 games in the past 3 years, ZiPS will also project a number of games in that neighborhood.  If the pitcher has missed these games because of his arm continually breaking down, it's more likely to end up being a correct playing time projection than if the pitcher missed those games because he was holding up liquor stores and consequently, in prison.  There's a general regression to a moderate amount of health, that level being higher for a position player than a pitcher.  There's also an age trend for position and age, based on history - playing time goes up in the 20s and starts dropping off fairly predictably in the early-to-mid 30s, first for C, 2B, and CF, and then everybody else.

- As noted, there's no merit-based projection for playing time.  If there was, Royce Clayton would be listed at 000/000/000 in 0 at-bats since he's not clearly preferable to any realistic option the Blue Jays have in any role, and is probably inferior to a number of unrealistic option the Blue Jays have in any role.  And if anyone says that should be undef/undef/undef in 0 at-bats or brings up L'Hopital's rule, I'm going to drive up there and beat you to death.

- Janssen's opposition wasn't all that weak - the average hitter he faced was a 270/337/432 hitter, which is pretty close to league average.

- Rios' improvement is probably real, but even real improvements in players that age are generally followed by a consolidation year.

Mike Green - Friday, January 12 2007 @ 11:06 AM EST (#161853) #
Thanks, Dan, for visiting.  I have revised my opinion of the Janssen projection.  It is entirely reasonable.
Pistol - Friday, January 12 2007 @ 11:08 AM EST (#161854) #
The overly positive projections for Overbay

Overbay hit .312/.372/.508 last year and is projected at .301/.374/.491.  That doesn't seem like a reasonable projection?

I do remember last year his projection was overly favorable (an OBP in the .390s I believe).
Ryan Day - Friday, January 12 2007 @ 11:14 AM EST (#161855) #

Too positive- Lyle Overbay

  Really? Overbay's projection looks pretty close to what he did this year, and not too far off his career numbers:

Career: 293/372/467

2006:   312/372/508

2007:   301/374/491

Mike Green - Friday, January 12 2007 @ 11:19 AM EST (#161856) #
Nope.  Overbay hit .301/.385/.478 and .276/.376/.449 in the easy league the previous 2 years, and he is 30.  If you make appropriate adjustments to his age 27 and 28 statistics for league, he projects to regress to about .295/.370/.475.  What is most noticeable is that Overbay's power improvement is projected to be more sustained than Rios'. 
China fan - Friday, January 12 2007 @ 11:30 AM EST (#161860) #

    Here are some of my questions about the ZIPS projections.  If I'm misunderstanding something, please correct me.  Otherwise I'll have grave doubts about this computer model.

   1) how can Janssen reasonably be projected as an above-average starter in the major leagues in 2007 when he is only a year removed from Class A and he was an abysmal 1-8 with a 8.58 ERA over his last eight starts in the majors last year?

     2) how can Jamie Vermilyea reasonably be projected as an above-average starter in the majors in 2007 when he has never played a game in the majors and is not even regarded as one of the better prospects in the farm system?  (Triple A hitters managed to pound him for a .291 average last year. I would expect that major-league hitters would do a lot better than that.)

    3) how can Towers reasonably be projected as an above-average starter in the majors in 2007 after a disastrous 2006 season where he sported an ERA of 8.42 and opponents hit a hefty .343 against him? 

Mike Green - Friday, January 12 2007 @ 12:25 PM EST (#161866) #
Dan Szymborski posted this (which I entirely agree with):

how can Janssen reasonably be projected as an above-average starter in the major leagues in 2007 when he is only a year removed from Class A and he was an abysmal 1-8 with a 8.58 ERA over his last eight starts in the majors last year?

- His distance from A is irrelevant, namely because he mastered AA pretty easily and while the ERA in Syracuse was high, his peripherals were excellent.

- Last 8 starts isn't a meaningful split.  An 8.58 in his last 8 starts is no more relevant to a future projection than his 3.09 ERA in his first 9 starts.  While it sounds neat and tidy to say he was figured out, if you weigh later production in a rookie season more heavily than early production, you decrease the accuracy of future projections (trust me, I've looked at this).  There wasn't any difference in quality of opposition faced, either - he faced just as many creampuffs when he was pitching poorly as when he was pitching well.

How can Jamie Vermilyea reasonably be projected as an above-average starter in the majors in 2007 when he has never played a game in the majors and is not even regarded as one of the better prospects in the farm system?  (Triple A hitters managed to pound him for a .291 average last year. I would expect that major-league hitters would do a lot better than that.)

He's actually not projected to be a better-than-average starter, actually.  He's projected to be a below-average starter with some solid relief stints.

3) how can Towers reasonably be projected as an above-average starter in the majors in 2007 after a disastrous 2006 season where he sported an ERA of 8.42 and opponents hit a hefty .343 against him?

Because 2006 isn't the first season that Towers has pitched.  And he wasn't just bad, he was bad and unlucky.  Anyone who thinks his real ability is an ERA of 8.42, I have a secret program to turn $5 into $50,000 to tell you about!
Jordan - Friday, January 12 2007 @ 12:27 PM EST (#161868) #

The projection that impresses me most is Adam Lind's -- if the Jays can get that kind of production out of him this year, especially the 39 doubles, they'll have done very well for themselves.

Rios seems to be short on power -- I really can't see him hitting fewer than 20-25 HRs over a healthy full season -- and .280 ought to be his baseline BA, but it's good to get an objective view of his likely production. Aaron Hill may not exceed that line by as much as many of us think. He unquestionably has the talent to hit .300 with extra-base power, but maybe not this year -- he still seems prone to lengthy slumps. The Russ Adams projection seems right on to me -- if the Jays committed to him at 2B for a full year (which they obviously won't be doing), he's do at least that well. Reed Johnson's line looks exactly right. If this lineup stays healthy, it should get on base and it should score.

On the mound, Burnett's projection doesn't seem outlandish, except that I'd like to see more IP from him in real life. Zips would have you think the Jays bullpen would be one of the best in the league, but even if Accardo does much more poorly than this, as I expect he will, this still projects as a very solid relief corps. I think Towers', Janssen's and Chacin's peripherals are all pretty accurate predictions, but I don't see those lines projecting to sub-5.00 ERAs in the AL -- especially Towers, unless those 28 HRs are all bases-empty. If he's used strictly in relief, McGowan should exceed his projection; I don't see any way in which Vermilyea comes close to his.

Dan, I'm not terribly familiar with Zips -- is there a reason that no pitcher on the staff, including Halladay, is projected to win more than 14 games? I assume that the model more or less dismisses wins as too variable a statistic to predict. Good work, and thanks!

Mike Green - Friday, January 12 2007 @ 12:37 PM EST (#161869) #
I'll answer that one for Dan.  The reason that neither Halladay nor Burnett is projected to win more than 14 games is that neither is projected to make more than 28 starts.  In light of the injury history of both pitchers, that is reasonable.  Winning half of one's starts is very, very good.
Glevin - Friday, January 12 2007 @ 01:43 PM EST (#161871) #

For the record, here are the ZIPS from 2006. http://www.baseballthinkfactory.org/files/oracle/discussion/2006_zips_projections_toronto_blue_jays/

Personally, I put almost no stock in these. If you took an knowledgable person on this site and asked them to project the numbers for the Jays, they would do as well. The vast majority of these projections are reasonable with Rios and Hill being projected to far too low and Chacin, Towers, Marcum, Accardo, and most of the Jays pitchers being rated too highly.

 

SheldonL - Friday, January 12 2007 @ 01:58 PM EST (#161872) #

Does anyone know where Justin Miller is now? Just curious is all...

fozzy - Friday, January 12 2007 @ 02:28 PM EST (#161873) #
According to ESPN, Justin Miller is playing in Japan now. Last year he signed a minor league contract and was invited to the Rays spring training, but didn't make the team (I believe he was one of the last ones cut). In April, about 3 weeks later, they sold his rights to the Lotte Chiba Marines in Japan, according to their AAA website. I can't find anything after that though.

Too bad, his AAA numbers in 2005 and 2006 looked pretty decent, guess he was just AAAA material.
AWeb - Friday, January 12 2007 @ 03:03 PM EST (#161875) #
If you took an knowledgable person on this site and asked them to project the numbers for the Jays, they would do as well.

See, that's always bothered me too. Anyone who follows the game can do what the projection systems do in their own heads, essentially. Take the last few years, possibly weight the most recent year more heavily, figure on the player declining/improving due to age, and that's pretty much it. What a fan can do is add subjective information to the mix, such as how likely a player is to get injured (pitchers), was the first half last year a fluke (Hill, Rios), that sort of thing. Essentially scouting report information, I suppose.

On the other hand....the numeric methods are great because they can do everyone, all at once, very quickly.  And they are better than just guessing "same as last year", which is the baseline any prediction method must get past (it's the same one they use for deciding on weather forecast cutoff dates, i.e., if looking out the window is a better guess at tomorrows weather than the forecast is, it usually isn't reported. Hence the 5-day and 7-day forecasts...anyway).  Projections like this are great for a starting point, and great for a quick look at other teams that I may not know as well.

I'm thinking a project like TangoTigers fan fielding survey would be interesting here. Can a fairly large group of fans do better, on average, at predicting 2007 performance? Could Da' Box readers beat ZIPS? My guess is that they could, but it would be close.
Mike Green - Friday, January 12 2007 @ 04:05 PM EST (#161881) #
Well, we did do polling of Bauxites before 2006 of projected ERA and OBP/SLUG for Jay players.  I calculated the weighted average of the poll results of Roy Halladay's ERA- it was 2.89.  ZIPS had it at 3.16. Score one for ZIPS. My guess is that ZIPS did better overall, but that it would be fairly close. 



Michael - Friday, January 12 2007 @ 07:01 PM EST (#161890) #
ZIPS does project playing time but only to the degree that if the team wanted this guy to play everyday how much would they play sort of way.  So injury history will project lower playing amounts.

ZIPS is fairly accurate as a projection system in the past.  Note also that it is projecting those performances in skydome, a hitters park, so the hitting projections assume a bump from park while the pitching projections are inspite of the park.

Also note that, as with all systems, the projections for hitters are much, much, much more accurate than the projections for pitchers.  These projections iirc have about a .7 correlation for hitters and a .3 or .4 correlation for pitchers.

Still, I think it would do significantly better than most bauxites indivudal predictions and slightly better than an average of all bauxites predictions.

Mike Green - Saturday, January 13 2007 @ 01:03 PM EST (#161908) #
Actually, I am pretty sure that ZIPS would do a much, much better job of projecting Astros players, for instance, than I, or Bauxites as a whole, could do.

 It is very difficult to balance subjectivity and objectivity; what ZIPS does is give a good objective take.  Obviously, there is room for improvement in the system.
Pistol - Saturday, January 13 2007 @ 09:25 PM EST (#161919) #
Interestingly, ZIPS did the best job of projecting pitchers of the several prominent projection systems out there in 2006.
Mike Green - Monday, January 15 2007 @ 05:00 PM EST (#161995) #
For fun, I used the Marcel projections for the Jays available on the Hardball Times and the Lineup Analysis tool available through Baseball Musings to compare with ZIPS.  The figures I used for Glaus and Clayton used 10% McDonald.  I split the catching platoon 2/3, 1/3.  Here is the result.

Honestly, I don't know why they play the games.  It's clear now that the team is going to score 5.2 runs per game!

Here is a summary of the Marcels for the Jay hitters:

Player PA AVG OBP SLUG
Thomas 492 .288 .365 .510
Glaus 586 .257 .355 .512
Lind 232 .303 .368 .490
Overbay 582 .296 .370 .481
Wells 606 .286 .343 .499
Rios 501 .292 .344 .402
Sparky 502 .289 .352 .431
Zaun 421 .261 .352 .412
Phillips 269 .243 .302 .374
Clayton 508 .264 .317 .368
Hill 544 .292 .357 .407
Adams 394 .258 .325 .401
McDonald 301 .248 .301 .301

Overall, it's pretty comparable with ZIPS.  I think the Hill projection is more accurate.
Mike Green - Tuesday, January 16 2007 @ 11:52 AM EST (#162015) #
The 2007 Marcels for pitching are very interesting.  Here they are.  The Marcels seem to be more regressed to the mean than ZIPS projections- no pitcher (even the closers Nathan and Ryan) have projected ERAs below 3.00, despite excellent peripherals.  The projections for Burnett, Thomson and Towers seem more reasonable to me than the ZIPS projections.  In general terms, the projections for starters seem to me to be very accurate.  Randy Johnson's projection is generous with regard to innings, but seems to be to be dead on otherwise.  Perhaps Tango has made some changes to the formula since 2004. 

Anyways, the Jays appear to be set to allow about 4.5 runs per game, or 730 in a season.  They look to have a comparable club to the Tigers in terms of overall quality, with somewhat more offence and somewhat less run prevention.

Gerry - Tuesday, January 16 2007 @ 01:09 PM EST (#162016) #
You didn't follow through Mike.  842 runs scored and 730 allowed translates into 92 wins.  A few career years and the Jays could make the playoffs, a few down years or injuries and they are in the mid 80's in wins.
Pistol - Tuesday, January 16 2007 @ 01:11 PM EST (#162017) #
842 runs scored, 730 runs allowed = 92.6 wins.
Mike Green - Tuesday, January 16 2007 @ 02:29 PM EST (#162023) #
Yep.  92.6 wins.  The Marcels, like ZIPS, have the Jays as playoff contenders.  It might be a slight overestimate because there is an unmeasured modest defensive decline from McDonald to Clayton.  Still, 92 wins will usually place a team in the playoff hunt.
Pistol - Tuesday, January 16 2007 @ 02:47 PM EST (#162025) #
Well, I would hope if the Jays are in contention they would go out and get a good shortstop which presumably would help their projection as it is now.
Mike Green - Wednesday, January 17 2007 @ 10:19 AM EST (#162071) #
Jeff Sackmann has published interesting data on #1-#6 starter performance in today's THT. Average ERAs (across both leagues) for the #3-#6 starters are 4.74 through 4.96.  An adjustment upwards must be made for the Toronto context both for park and league.  To succeed, of course, you want to be better than average, but it seems that the projections for Marcum and Janssen (and Thomson) are indeed better than average.  The objective projections do not account for the fact that Thomson's lost time last year was due to a shoulder injury, rather than some less worrisome cause.
ayjackson - Wednesday, January 17 2007 @ 12:05 PM EST (#162092) #

Also of note in the Sackmann article is this comment in relation to the jump in average ZiPS ERA from the Ace to #2 starter:

The biggest difference is between the #1 and #2 starters, suggesting why having two aces catapults a team so far ahead of the pack.

This would suggest that the focus of our concerns should not be on the #3-6 starters whom, by projections, look like they'll fit the roles adequately, but on the entrenchment of AJ as a 180 inning-sub 4 ERA guy.  This would give us two front end starters who could shave over a run-and-a-half off the average every time through the rotation.


 

Chuck - Wednesday, January 17 2007 @ 01:25 PM EST (#162103) #

This would suggest that the focus of our concerns should not be on the #3-6 starters whom, by projections, look like they'll fit the roles adequately, but on the entrenchment of AJ as a 180 inning-sub 4 ERA guy. 

Except that Burnett is already entrenched at #2. No amount of focusing figures to change that, or to necessarily get him to pitch like the #1B we hope he can become to Halladay's #1A.

On the other hand, while the #3-6 spots could well be average within the context of #3-6 pitchers, the organization does have work to do, namely figuring out who specifically these people are. While Chacin seems to be the de facto #3, I'd imagine that the focus at spring training will be on sorting out which of Thomson, Marcum, Towers and Janssen start the year at #4 and #5.

ayjackson - Wednesday, January 17 2007 @ 03:43 PM EST (#162117) #
I didn't say entrenched as a #2.
ayjackson - Wednesday, January 17 2007 @ 03:50 PM EST (#162120) #

while the #3-6 spots could well be average within the context of #3-6 pitchers, the organization does have work to do, namely figuring out who specifically these people are

My point is, that the bigger gauge of our success will be the front two's ability to give us 400 innings at 3.3 ERA+.  I think we have enough competition for the back end to get vanilla production there.  But, as the article suggests, the difference maker in a rotation is having the two studs at the beginning.

Mike Green - Wednesday, January 17 2007 @ 04:04 PM EST (#162124) #
That's not a realistic expectation, given their histories.  360 innings at 3.5 ERA would be the optimistic account, and that's ZIPS'  view.  330 innings at 3.7 would be the pessimistic view, and that's MARCEL's.  It is the unusual pitcher who can be reasonably projected to throw 200 innings in a season now. 
Chuck - Wednesday, January 17 2007 @ 04:43 PM EST (#162135) #

I didn't say entrenched as a #2.

My mistake. I guess I misunderstood "This would suggest that the focus of our concerns should [be] on the entrenchment of AJ as a 180 inning-sub 4 ERA guy." I guess I didn't understand what that meant.

I think we were talking at cross-purposes. It sounds like you are assuming that #3-6 will provide average (for #3-6) performance, and huge potential gains therefore rest with Burnett, a #2, hopefully performing at a #1 level.

I meant to suggest that Burnett is the #2, as we all know, and nothing is going to change that, be it an ERA of 2.50 or an ERA of 5.00. My concern, from a roster management perspective (as opposed to yours, from a roster performance perspective) is on figuring out who makes up the balance of the rotation. While many of the projections have all of Chacin/Thomson/Marcum/Janssen/Towers as indistinguishable league averagish types (presumably leading to your confidence about #3-6), my concern is that this may not be the case at all. There may well be 2 or 3 pitchers in that group who can deliver a 4.50-5.00 ERA, but I'm skeptical that all five can. My concerns lie with the organization being able to identify the 2 or 3 who can. Not that that I believe management to be an inept group of decision-makers, but rather because of the innate difficulty in forecasting young pitchers (or any pitchers for that matter). I do know that none of these pitchers will be given as long a leash as Towers was in 2006, so I would expect to see a lot of aggressive flip-flopping.

ayjackson - Wednesday, January 17 2007 @ 04:50 PM EST (#162136) #

That's not a realistic expectation, given their histories.  360 innings at 3.5 ERA would be the optimistic account, and that's ZIPS'  view.  330 innings at 3.7 would be the pessimistic view, and that's MARCEL's.

That's the point.  The Jay's success this year could bear more on how close Doc and AJ get to 400 IP and 3.3 ERA+ (ZiPS 3.5 prediction is given that they play the home games in Rogers Center, is it not?), rather than how their back end performs.

Granted, if the #5 guy had an ERA+ of 5.75 that may offset Ace performance out of the top two, but is that reasonable?

As for the unreasonableness of 400 IP out Doc and AJ, it should not be expected, but it would not be unreasonable.  Doc has pitched over 200 IP three of the last five years, while averaging exactly 200 inning over that period.  AJ has exceeded 200 IP twice in those five years.  So a crude analysis would suggest that there is a one in four chance that both exceed 200 innings, one would assume a slightly greater chance that they would combine for over 400 innings.

ayjackson - Wednesday, January 17 2007 @ 05:07 PM EST (#162141) #

potential gains therefore rest with Burnett, a #2, hopefully performing at a #1 level

My point of emphasis with Burnett is how much he pitches.  If he peforms at a level like last year (<4.00 ERA) for 180+ innings, then that would mark a significant improvement over the average #2 starter and, as the article suggests, significantly affect the success of the club.

 

My concern, from a roster management perspective (as opposed to yours, from a roster performance perspective) is on figuring out who makes up the balance of the rotation.

I see your point here and it is a concern.  I am concerned that Janssen and/or Marcum (or McGowan/Rosario for that matter), could be pitching better than one or more of Chacin/Thomson/Towers, but not get a starting role due to 'other' factors (salaries, waiver risks etc).  I wonder if Towers pitched average during ST and Janssen (for example) was excellent, could we sneak Mr. Towers through waivers again.  Or could we trade him. 

Mike Green - Thursday, January 18 2007 @ 11:50 AM EST (#162182) #
Want some fun?  Visit Tangotiger's wonderful site, and check out the runs scored calculator.  The calculator does, of course, much more than that.

So, if you take the ZIPS or Marcel projection for Aaron Hill, make changes in the calculator to the baserunning assumptions to reflect Hill's skill in that department, you can get your own homemade run estimator of Hill's projected offensive value.  I plugged in the Marcel figures and adjusted Hill's baserunning figures and ended up with 5.6 runs per game.  That's probably a touch high, as it does not account for outs on the bases.

2007 Blue Jays ZIPS projections | 51 comments | Create New Account
The following comments are owned by whomever posted them. This site is not responsible for what they say.