Batter's Box Interactive Magazine Batter's Box Interactive Magazine Batter's Box Interactive Magazine
Three games in, and already grumbling about the Jays batting order has begun. Mercy! I would say that the John Gibbons has been pretty conventional is his deployment of the troops, but of course conventional doesn’t always mean right. Everyone has their own ideas, naturally, but lets take a look at the current nine and a more statistically oriented bunch and see what we can come up with.


Lineup construction has advance from simply balancing lefties and righties, and is being recognized as an increasingly important strategy decision (depending on who you ask anyway.) We’ll start with a review of the 2007 season. Reed Johnson, Lyle Overbay and Troy Glaus all missing substantial time due to injury, but everyone else was mostly healthy.

On average, the Jays starting lineup usually looked something like this last year (OBP/SLG plus number of games started in parenthesis.)

1. Alex Rios .354/.498 (139)
2. Lyle Overbay .315/.391 (107)
3. Vernon Wells .304/.402 (143)
4. Troy Glaus .366/.473 (110)
5. Frank Thomas .377/.480 (147)
6. Aaron Hill .333/.459 (158)
7. Gregg Zaun .341/.411 (93)
8. Adam Lind .278/.400 (72)
9. John McDonald .279/.333 (94)

Of course, there were variations – Vernon and Reed also spent some time leading off, with Rios dropping to the 3 hole; Matt Stairs played in a bunch of games and generally hit 2nd; Frank Thomas and Glaus alternated in the 4 and 5 holes; Aaron Hill hit anywhere from 5th to 7th; and Royce Clayton somehow managed to start 55 games... The results were, on the whole, pretty underwhelming, as the Jays finished 10th in the league in runs scored, 12 in batting average, 12th in OBP, 8th in SLG and 9th in OPS.

Broken down by batting order position the Jays hit thusly (OBP/SLG):

1. .321/.407
2. .319/.420
3. .326/.403
4. .342/.444
5. .391/.547
6. .334/.445
7. .324/.423
8. .274/.319
9. .301/.370

I wont get into the (numerous) failures of the 2007 Jays’ hitters, but clearly it was not a strong campaign on the whole. The team’s struggles started at the top of lineup, as a past strength (whither Freed Johnsononatto?) became a huge weakness. In a weird quirk, Frank Thomas and Troy Glaus each hit better batting 5th than 4th. So what could have been done better last year?

The epic baseball book, The Book, covers lineup construction with a great deal of insight, and I am going to borrow greatly from it here (for a primer, check out Dave Studeman’s piece on the always excellent Hardball Times.

The Coles Notes version:
Your best hitters should bat 1st, 2nd and 4th, with the 1 and 2 guys being big OBP guys. After that, you want your next two best hitters in the 3 and 5 slots, and then everyone else in descending order of non-suckitude.

There are a bunch of reasons for this. You want your best hitters hitting near the top of the lineup where they are more likely to get that extra AB at the end of a close game and extra AB’s over the course of a season, and you want OBP guys to set the table. The number 3 hitter is less likely to lead off an inning (as he almost never leads off the 2nd) and thus is less in need of OBP skills, and the number 4 hitter is generally a pretty likely candidate to lead off the second inning, as well as bat with men on base in the first, so he should be a good hitter all around. Looking back at the Jays 2007 lineup with this focus, a more ideal lineup would have looked something like this

1. Alex Rios .354/.498 (R)
2. Frank Thomas .377/.480 (R)
3. Aaron Hill .333/.459 (R)
4. Troy Glaus .366/.473 (R)
5. Gregg Zaun .341/.411 (S)
6. Vernon Wells .304/.402 (R)
7. Lyle Overbay .315/.391 (L)
8. Adam Lind .278/.400 (L)
9. John McDonald .279/.333 (R)

I think we can all agree that Thomas, Rios and Glaus were the Jays three best hitters last year (maybe not in that order) and here we have them batting 1st, 2nd and 4th. Having Glaus and Thomas hit back to back leading off might have been a bit much, so Glaus gets dropped to 4th, and Rios could be dropped to second so that DP machine Aaron Hill doesn’t bat right behind him, but now we’re quibbling. I have Hill batting 3rd and Zaun 5th because Zaun has better OBP skills, and the rest of the Jays follow in descending order of quality. It’s not for me to say that this lineup, featuring the Jays making the most starts, would have done better, but it’s not outside the realm of possibility. Hold your arguments ‘til the end, please.

Lets look at 2008. David Eckstein replaces John McDonald for our purposes, Troy Glaus gets replaced with Scott Rolen, Shannon Stewart replaces Reed/Lind in the right field platoon, and Rod Barajas gets some at bats backing up Zaunie. The rest of the Jays lineup is the same, for all intents and purposes (though Scutaro will bat for Rolen for the first month of the season or so.) Here’s what John Gibbons has thrown out the first two games, with three year batting splits of OBP/SLG included, along with handedness.

1. David Eckstein, SS .357/.375 (R)
2. Shannon Stewart, LF .336/.388 (R)
3. Alex Rios, RF .339/.472 (R)
4. Vernon Wells, CF .327/.470 (R)
5. Frank Thomas, DH .373/.518 (R)
6. Lyle Overbay, 1B .355/.455 (L)
7. Aaron Hill, 2B .341/.415 (R)
8. Marco Scutaro, 3B .331/.384 (R)
9. Zaun/Barajas, C .353/.409 (S)/.310/.434 (R)

There are a couple of problems here, I think. Firstly, despite David Eckstein’s intangibles, he is simply not good enough to be batting leadoff, compared to some of the other Jays hitters. I don’t think it would be inaccurate to say that he is one of the worst two hitters in this lineup as currently constructed, though the team is good enough that this isn’t a slap in the face. Vernon also is a poor candidate to be hitting cleanup this point – unless he shows that he is the hitter he was in 2006. I think Aaron could be moved up the lineup – he had a good 2007 and figures to improve. Also the Zaun/Barajas combo figures to be better than Scutaro, who could be dropped in the order.

Keeping this in mind, as well as our earlier defined goals for lineup construction, here is one line up the Jays could field.

Leading Off: Lyle Overbay .355/.455 (L)
This is a pretty unorthodox choice – first basemen aren’t lead off hitters. However, Overbay is arguably the team’s best on base threat, given Frank Thomas’ recent woes. 2008 promises to be a make or break year for Lyle Overbay. He is hitting the presumed downside of his career at 31, and was injured for a good part of last season, none of which portends well. However, he hit .312/.372/.508 in his last full season, which I think the Jays would happily take. Eckstein obviously figures to get the majority of the at bats here, which is probably not a great thing

Batting Second: Matt Stairs/Shannon Stewart 358/.470 .336/.388 (L)/(R)
The Stairs/Stewart platoon (though not a strict platoon) hits second, though Stairs is a better choice here than Stewart. Stairs had a very good year last year, and posted a very good OBP (well, for the Jays) of just under .370, and was clearly one of the Jays best hitters in limited duty. He gets spelled by Stewart, who has also shown very good OBP skills during his career (.361 career average) and is more of a traditional top of the order player. I think this is really the Jays most logical choice, and Gibbons seems to agree. Other candidates here would be Zaun because of his good on base skills, though he is probably not a good enough hitter overall, and Thomas, which would be rather unorthodox.

Batting Third: Aaron Hill .341/.415 (R)
I think this is really the most logical place in the order for Hill. He has to be a cinch to hit better than his three year average, and he is exactly the type of hitter you want batting third. He has pretty good power but only average plate discipline, and I think that a .300/.350/.450 year would not be a surprise. Rios figures to be the guy hitting here all year, and while I think Rios should be batting cleanup, neither he nor Hill is a bad choice given their skill set (assuming that Wells returns somewhat to form.)

Batting Cleanup: Alex Rios .339/.472 (R)
While there can be arguments over whether Rios was the Jays best hitter in 2007, he sure looks like he will be their best hitter in 2008. This is pretty traditional, but makes sense. Of course, Vernon will hit fourth until he struggles – let’s hope he makes this all moot by having a fantastic year.

Batting Fifth: Frank Thomas .373/.518 (R)
A conventional choice - the Big Hurt seems to be on the downside of his career, but still figures to be a sufficiently above average hitter, as well as an OBP machine. Additionally, he hit .285/.391/.544 in almost 200 ABs from the 5 slot last year, which while perhaps meaningless, is production any team would take in a heartbeat

Batting Sixth: Vernon Wells .327/.470 (R)
Boy, $126 million doesn’t buy what it used to. Vernon had a bad 2007, and almost has to rebound in 2008 (right? Right?!) I think that the drop would take some pressure off of him, and until he hits better there is not much reason to have him take up valuable space at the top of the lineup. Lyle figures to get the majority of plate appearances here, which his unfortunate given is on base skills. Scott Rolen could figure too.

Batting Seventh: Gregg Zaun/Rod Barajas .353/.409 (S)/.310/.434 (R)
Gregg Zaun is getting pretty old, but again possess good on base skills, while Barajas mixes in the power. At this point, they are better than the remaining options. I guess Rolen might hit here when he returns, depending on Overbay.

Batting Eight: David Eckstein .357/.375 (R)
Eckstein does a lot of things you would want from your leadoff hitter – draw walks, gut out at bats, see a lot of pitches. He just doesn’t do them well enough to justify being the everyday leadoff hitter on a team with playoff aspirations.

Batting Ninth: Marco Scutaro .331/.384 (R)
Nothing against Scutaro, but he is not an everyday player at this point, and its not unreasonable to drop him to ninth. The catchers figure to see regular duty here once Rolen returns


Optimal(er) lineup?
1. Overbay L
2. Stairs L/Stewart R
3. Hill R
4. Rios R
5. Thomas R
6. Wells R
7. Zaun S/Barajas R
8. Eckstein R
9. Scutaro R

So in keeping with the theories we touched on earlier, the Jays best hitters at this point (Overbay, Stairs, Rios, Thomas and Hill) get the bulk of the at bats hitting 1 through 5, with Wells sixth and then the rest of the hitters in order of goodness. If Vernon picks up and Thomas struggles the two of them could be flipped. When Rolen gets back I think he slots in the seventh hole in this arrangement.

The chief difference between this lineup and the lineup the Jays have actually used is basically just flipping Eckstein and Overbay and flipping Vernon and Aaron, along with a few minor shuffles up or down. I think that Overbay and Aaron are likely to do better than the people they replace, so I think it makes sense. If one were to stay more conventional, two important changes could be moving Rios to the four slot and batting Wells third, and dropping Eckstein to the bottom of the order. On the whole, this lineup might be five runs better over the course of a season – not even a win. That’s not really the point though – the point is to come up with a different lineup.

Alright Bauxites – how could this, or the real Jays lineup, be improved.


The 2008 Blue Jays Lineup | 21 comments | Create New Account
The following comments are owned by whomever posted them. This site is not responsible for what they say.
ChicagoJaysFan - Friday, April 04 2008 @ 10:55 AM EDT (#182088) #
The challenge of it all is picking / determining which players are the best hitters. 

For instance, if Rolen is the Rolen of not 2006 or 2004, then he's the best hitter in this line-up.  Having him 7th is a mistake.  If he's the other Rolen, 7th could even be too high in the order.

Batting order construction is incredibly difficult because with people like Rolen you won't know how to construct your line-up optimally until it's almost too late to do so.  And wwith Wells and Thomas, the Jays have 3 guys who could each be either the best or worst hitter on the team.

Mike Green - Friday, April 04 2008 @ 11:30 AM EDT (#182091) #
Fangraphs.com has 5 different projection systems which taken together give, I think, reasonable projections for each of the Jays. Here is Lyle Overbay's page. Aaron Hill's average projection comes out at .340/.422; personally, I think that's a little low, but I concede that I may be biased.  Vernon Wells' projection comes out at .331/.460.  Wells is a reasonable #3 hitter if he meets the projection, but a poor choice for a cleanup hitter. 

To see the value of averaging the projections, check out Scott Rolen's.  There is one ridiculously optimistic projection (Bill James) at .363/.509 and one ridiculously pessimistic projection (ZiPS) at .319/.392.  The other three are squarely in the middle and reasonable. 

Speed and baserunning ability enter into the calculus, as well.  It's not enough to offset 40 points of on-base percentage for a leadoff hitter, of course.

In any event, it is clear that the best offensive player on the club right now taking into account all aspects is Alex Rios.  My ideal lineup right now would probably be Hill, Stairs/Stewart, Wells, Rios, Overbay, Thomas, Zaun/Barajas, Scutaro, Eckstein.

Ryan Day - Friday, April 04 2008 @ 11:39 AM EDT (#182092) #
In Rolen's case, I don't think averaging is very useful at all. The three mid-ground projections seem to be averaging out his injured and healthy years, which isn't terribly relevant - either Rolen's been healthy and productive or he's been hurt and sucky, with no middle ground. James assumes he'll recover - his projection is a slight decline from 2006, the year between Rolen's injuries - and ZIPS thinks injured Rolen is the new standard.

Of course, as far as lineups go, I'm not sure it really matters. If Wells and Rolen repeat their 2007 seasons, it doesn't matter where they hit - the team is pretty much sunk offensively.

ChicagoJaysFan - Friday, April 04 2008 @ 11:50 AM EDT (#182093) #
To see the value of averaging the projections, check out Scott Rolen's.  There is one ridiculously optimistic projection (Bill James) at .363/.509 and one ridiculously pessimistic projection (ZiPS) at .319/.392.  The other three are squarely in the middle and reasonable.

I think we'll have to agree to disagree - those middle projections are the only ones that I think are unreasonable.  It wouldn't surprise me at all if Rolen was healthy and put up .872 OPS (or higher) and it wouldn't surprise me at all if he was unhealthy and put up .711 (or even worse actually).  What would stun me is if he did something in the middle.

In his career Rolen has had years below .730 OPS and years above .860 - he's never had an OPS in the middle of those two numbers.  In those two bad years, his struggles were apparently due to health issues, so I don't see .802, .777, and .749 being reasonable.
CaramonLS - Friday, April 04 2008 @ 12:07 PM EDT (#182095) #
I'd switch Stairs/Rios vs. RHP, not sure about having my 2 lefties back to back to lead off the game.  But it still just doesn't look right having Overbay lead off, infact, I'd almost be more comfortable with Zaun leading off vs. Righties.

I'd almost rather do this, especially if you want Hill as a #3:

1) Eckstein
2) Overbay
3) Hill
4) Stairs
5) Rios
6) Thomas
7) Wells
8) Zaun
9) Scutaro

Someone like Reyes or Ichiro would just be perfect for the leadoff slot on this team (or every team for that matter).
Geoff - Friday, April 04 2008 @ 12:24 PM EDT (#182097) #
Rather than leading off, I'd sooner see Overbay batting cleanup, with Stairs/Stewart leading off, followed by Hill and Rios.

If Overbay hits like he can, he's fine as a cleanup hitter and probably more comfortable in that role than leading off the game.

Problem is, the Jays don't really have a go-to guy to bat cleanup outside of Rios. Wells and Thomas are the only other 'sluggers' on the team and neither looks ready to walk the walk. I understand the Jays are pinning hopes on seeing Vernon return to the form he is being paid for, and giving Thomas a chance to put together a good April and May, but Lyle could do just as well.

I expect Wells will remain #4 until he proves he can't cut it (despite some bad ABs and missed RBIs, Vernon's big Wednesday reflects well.)

Lyle's a good hitter, but he should be hitting behind Rios to drive him (and/or Wells, Hill, Stewart, ideally) in, not vice versa.

If the Jays are to succeed, they must count on Wells and Thomas to be run producers and the return of Rolen to give them a shot in the arm. If Wells and Thomas need a month to get into overdrive, they should be batting 5th or lower both until the time is ready.

I'd prefer to see Rios, Hill, Overbay, Wells, Stairs, Thomas, Rolen, Zaun, Eckstein when everyone is healthy. If Wells bats cleanup worse than Benjie Molina this year, well the Jays have no hope of playoffs and Wells will have to wear that weight.

ChicagoJaysFan - Friday, April 04 2008 @ 12:43 PM EDT (#182099) #
I don't think people really understand what Rolen's like as a hitter.

If he's worth playing (i.e. not the 2005 and 2007 Rolen), he's pretty much guaranteed to slug over .500.

That's your clean-up guy right there.  As it stands right now, including Rios, we don't have a player that has the capability of hitting as well as Rolen.

At the same time, his performance of 2005 and 2007 is not #7 worthy - it's #9 or on the bench worthy.  Think Mark Scutaro #'s if you want to know how bad.  And by the way, Scutaro actually posted almost identical OPS+ numbers to Rolen in 2005 and 2007.

Mike Green - Friday, April 04 2008 @ 12:44 PM EDT (#182100) #
The assumption that a player who has had hot and cold seasons in four years running due to health issues will have either a hot or cold season in a following season is dubious.  Players often have lukewarm seasons (as a whole) due to being injured in the middle of the year, or taking time to recover after an early injury. 

To put it simply, the three middle OPS projections for Rolen are between .749 and .802. The high is .872; the low is .711.  I am pretty sure that he'll be between .720 and .850.  You can have the over .850 and the under .720, CJF, and I'll give you 3-1 odds.  But, since you are in Chicago, it'll have to be 3 virtual no-prizes to my 1. 

ChicagoJaysFan - Friday, April 04 2008 @ 12:54 PM EDT (#182102) #
MG - I'm curious why you say it's dubious that Rolen will not perform this year in a way that he has never performed previously in his entire career.  We'll find out in 5-6 months or so though.
uglyone - Friday, April 04 2008 @ 01:05 PM EDT (#182106) #

It's especially tough this year to order our batters from "best" to "worst".  Especially given the large question marks around so many hitters like Wells, Overbay, and Rolen - all of whom could realistically be .850-.900ops hitters, but at the same time could also realistically be .700-.750ops hitters.

IMHO, we have such a balance throughout the lineup, that we should probably determine the lineup in more the traditional way of "style" of player, rather than on "ranking" the hitters from best to worst. (i.e. speedy high obp guys at the top, power hitters in the middle).

That is, of course, unless the hitters end up separating themselves out over the course of the season with their performances, so that the difference between our "best" and "worst" hitters becomes much more significant.

 

Barry Bonnell - Friday, April 04 2008 @ 02:01 PM EDT (#182110) #

The Jays have called a press conference for 3PM today presumably to announce the contract for Rios.

http://blog.rogersbroadcasting.com/mikewilner/

The Bone - Friday, April 04 2008 @ 03:02 PM EDT (#182113) #

Here's my (somewhat off the wall) suggestion:

Vs. Righties

1. Rios (RF), 2. Overbay (1B), 3. Rolen (3b), 4. Stairs (DH), 5. Wells (CF), 6. Zaun (C), 7. Hill (2b), Stewart (LF), Eckstein (SS).

Vs. Lefties (the off the wall part)

1. Rios (RF), 2. Hill (2B), 3. Wells (CF), 4. Thomas (DH), 5. Rolen (3b), 6. Zaun (1b), 7. Barajas (C), 8. Scutaro (LF), 9. McDonald (SS)

 

 

 

ChicagoJaysFan - Friday, April 04 2008 @ 03:11 PM EDT (#182114) #

Vs. Lefties (the off the wall part)

1. Rios (RF), 2. Hill (2B), 3. Wells (CF), 4. Thomas (DH), 5. Rolen (3b), 6. Zaun (1b), 7. Barajas (C), 8. Scutaro (LF), 9. McDonald (SS)

Why Zaun over Overbay at first?

Overbay is better defensively (his range factor is usually way ahead of the league average and Zaun's never played there) and his career OPS vs. Lefties is better than Zaun's.

Also, when looking at Zaun specifically - he seems to break down towards the end of each year, so I think he needs more rest, not less.
Barry Bonnell - Friday, April 04 2008 @ 03:16 PM EDT (#182116) #

Wow, Hill has been extended as well. 4 years, 12 Million with up to a 3 year option!

DRising - Friday, April 04 2008 @ 03:19 PM EDT (#182118) #
Not just three year options, three year CLUB options -- for 2012-14 -- according to Jays site. Great news, though if I were Hill I'd be looking for a new agent...



timpinder - Friday, April 04 2008 @ 03:24 PM EDT (#182119) #

I'd platoon DH Stairs and Thomas, but that's not likely to happen, at least for a while.  I'd probably go with the following lineup against RHP:

Hill
Overbay
Rios
Wells
Rolen
Stairs
Thomas
Zaun
Eckstein

I think Overbay's struggles last year had much more to do with his injury than batting 2nd, and I just don't think Thomas will be any better against RHP than he was in 2007, in fact he may get worse.  I'm assuming Wells, Rolen and Overbay are going to hit like their career norms, but I feel like Wells' next few years are going to be closer to 2003 and 2006.  Just a feeling.

On a side note, there have been a lot of comments about Lind's future with Toronto.  Blair made it abundantly clear in his blog today what the plan was with Lind.  "Adam Lind will be the Opening Day left-fielder in 2009".  Blair's been very reliable in the past.  Lind has showed a lot of power so far this spring and the first day into the season.  I have to think that if he's mashing in AAA come June, and Stairs' hip is still bothering him, Lind might become the regular left-fielder sooner rather than later.  The Jays would have more options in the lineup with four left-handers (Lind LF, Overbay 1B, Zaun C, Stairs DH).

christaylor - Friday, April 04 2008 @ 03:30 PM EDT (#182120) #
I question averaging projections. First, each projection predicts different amounts of playing time (this could be taken into account with a weighted average)... but more damning is that while the numbers look the same, I think like averaging apples and oranges. Different projection systems have different built in assumptions an methodologies. In a glance I took at the errors two projections systems (James/BIS and PECOTA) a couple things were revealed to me that speak against just averaging the two. First the two projection systems are correlated (which make sense) but this tells us that knowing one tells us something about this other, so when you add another projection you're not getting another new data point, but somewhat less than one data point. The second thing I learned from a brief poke around the numbers I did a few weeks back is that the errors the two systems make (comparing them to real data)... this point is related to the first but it makes it easy to see why averaging doesn't tell us what one might think it does. For example with averaging you'd expect that the average of the two would better than either one... this is not the case, because the errors are correlated both projections tend to under-project or over-project, this tends to make the average projection worse than one or the other projection.

This doesn't really say anything about Rolen's projections for this year and I need a deeper look at the data I have (at sometime other than at 4am taking a break from writing my PhD thesis...) and I'm not criticizing you and I myself had the same thought about averaging projections (that's why I looked at this)... but as I've tried to explain (badly) here averaging projection carries some assumptions, which are violated when look at the data.

If anyone has studied this issue in more detail, I'd love to hear about it.
Shane - Friday, April 04 2008 @ 03:33 PM EDT (#182121) #

Not just three year options, three year CLUB options -- for 2012-14 -- according to Jays site. Great news, though if I were Hill I'd be looking for a new agent...

According to an article this past week, Hill has been very involved in his contract deal all the way through. So if he has a problem with his deal from 2012-2014 he might only have himself to blame.

christaylor - Friday, April 04 2008 @ 03:33 PM EDT (#182122) #
The agent might be the problem (bad advice)... but the paranoid guy who lives inside my head thinks that Hill might know that he's going to have to have knee surgery or something nasty like that soon... and took the sure money.

I have no evidence, just mere paranoia (but just because you're paranoid don't mean they're not after you).
ChicagoJaysFan - Friday, April 04 2008 @ 03:35 PM EDT (#182123) #
Wow, Hill has been extended as well. 4 years, 12 Million with up to a 3 year option!

Even for someone that is so far from free agency, that seems very low.  3 mil per year for top-5 defense (if not better) at a key defensive position and 105+ OPS+ at the same time. 

What's really scary about our division though is that Hill is probably the 4th best 2B in the division during those 4 years (Cano, Roberts, and Pedroia). 
Wildrose - Saturday, April 05 2008 @ 11:04 AM EDT (#182190) #
Well done Anders, this is the best sabermetric analysis I've read all year on the Box. Nothing much to add either than to  keep it coming.
The 2008 Blue Jays Lineup | 21 comments | Create New Account
The following comments are owned by whomever posted them. This site is not responsible for what they say.