PDA

View Full Version : Mackanin's lineup construction:



texasdave
07-19-2007, 07:05 PM
Here is a post on today's game thread by Oneupper, a person whose posts I greatly respect. However, I have to take issue with this one.



On the lineup:

MacNarron insists in splitting Griff and Dunn with Phillips. It was and continues to be a bad idea.


At the heart of this post is the concept of 'protection'. I am on the fence on whether 'protection' exists or not, but for the sake of this discussion let's assume that it does.

His contention, if I am reading his post correctly, is that by splitting up Griffey and Dunn they have no protection and can therefore be pitched around. Essentially he posits that by doing this, Mackanin is allowing the other team to take the bats out of their hands. This forces Phillips and Hatteberg (among others) to win the game instead.

I would like to make several points:

1st) Mackanin has managed 13 games. He has 'split' the lineup in 9 of these. In these 9 games the Reds have averaged 5.4 runs. In the 4 'unsplit' games the Reds have averaged 5.0 runs. I am aware that these numbers are close and it is a very small sample size. But it is also seems to indicate that by 'splitting' Griffey and Dunn the Reds' offense has not suffered.
2nd) Only one of the two players - Griffey or Dunn - can be 'protected. They can not hit behind one another. There will always be one or the other that can be pitched around. And another player besides these two will always be in the position to have to step up.
3rd) Is any loss in 'protection' suffered by Griffey or Dunn by being split up offset by the additional 'protection' now offered to Phillips (or whomever else is placed between them? IOW, who needs to be 'protected' more - a seasoned veteran like Griffey, an extremely patient hitter like Dunn, or a youngster like Phillips?
4th)I see no advantage being transferred to the other team by constantly and consistently putting Griffey and/or Dunn on base by pitching around them. This seems to me to be a recipe for disaster for the opposing squad.

Of course this is in the very early stages of experiment and perhaps over the long run either the 'split' or 'unsplit' lineup will prove more beneficial. I have a sneaking suspicion that it really is not going to matter all that much.

PS I think it is somewhat unfair to address another person's post without them being able to respond. Therefore if Oneupper wants to respond to this he can PM me with his thoughts and I will post them on Sundeck.

ChatterRed
07-20-2007, 02:24 AM
The results speak for themselves. I have been saying for about a month that Phillips is the RH power bat that needed to be put in between Junior and Dunn. It is working out, whether people admit it or not.

durl
07-20-2007, 11:32 AM
I think it's strange that Phillips is taking on the role that Encarnacion should be doing. I'd really like to see Edwin get back on track offensively.

foltza
07-20-2007, 01:03 PM
all of these are valid points, but i'd like to interject that griffey and dunn are by far our biggest power threats. it just seems logical that they should hit 3-4, no matter that they are both left-handed. isn't that the way line-ups are supposed to be constructed?
1. lead-off
2. get em over guy
3. best hitter
4. clean-up guy -- big power, drive in runs
5. next best hitter, and so on

CySeymour
07-20-2007, 01:09 PM
For the most part, I still think Bill James' mantra is right: Line-up construction really doesn't mean that much. The most important thing is to have the right players playing.

JLB5
07-20-2007, 02:15 PM
For the most part, I still think Bill James' mantra is right: Line-up construction really doesn't mean that much. The most important thing is to have the right players playing.

James has also noted that this assumes that the best hitters are at the top of the lineup. Hitting your .163 batting pitcher lead-off or your .242 hitting SS clean-up would be just plain stupid. James says to forget so much about the traditional roles and get the best .OBP guys at the top of the order. Also, his statistics have demonstrated that the #3 hitter is the least important of the top 5. Of course, the Reds have their best hitter in the #3 spot. By James' theory, you would probably end up with a lineup of:

Hatte
Hamilton
Edwin
Griffey
Dunn
Phillips
Gonzo
Ross

Or with Hamilton out:

Hatte
Dunn
Phillips
Griffey
Edwin
Freel/Hopper
Gonzo
Ross

Against lefties, something like:

Hamilton
Phillips
Dunn
Griffey
Edwin
Conine
Gonzo
Ross

texasdave
07-20-2007, 03:09 PM
Oneupper sent this reply via PM. He makes a number of good points:



Your post addresses mine, so we’ll do this runaround to get into this discussion on the Sun Deck.

The issue of “protection” is a tough one. I’ve been reading “The Book” by tangotiger, et al. where they address the issue and conclude (with a large sample size) that unprotected hitters tend to walk more AND strikeout more. Overall production doesn’t seem to be affected that much.

Their analysis (I believe they admit it also), is somehow constrained by the fact that its hard to define “protection” and is also limited by the fact they work with situations where a IBB could be warranted. It’s a tough issue.

My analysis on it is anecdotal and observational, and given the multitude of variables involved, there isn’t much more to go on.

Here’s the issue:

If you lineup Griffey/Phillips/Dunn/RH as opposed to let’s say Griffey/Dunn/Hatteberg, against a RH starter., it makes for quite a simple gameplan: pitch “around” Griffey and Dunn…go after Phillps and RH (usually EE, could be someone else -AG).
Pitching “around” is not walking. It’s pitching carefully…going for the corners, breaking balls on hitters counts, etc. Best case scenario, the batter chases a bad one, or you get an ump call on a borderline pitch. Worst case scenario, batter walks, bloops or bleeds one somewhere. Phillips or RH may beat you. But that’s the plan.
Here’s the OPS vs RH

Griffey .985, Phillips ,729 Dunn 1.027 EE .754 Hat .887

Anectodally. This situation looked quite clear in the July 8 vs. the D’backs. Young RH on the mound for Arziona. Griffey was 0 for 3 with 2 walks, Dunn 0 for 2, 2 walks. Great .444 OBP, but not much to show.
Reds actually one that one in 11 innings..4-3.

Against LH, there could be a case for the “split”

Griffey . 863 Phillips 1.003 Dunn .684

Unfortunately, Narron usually had Conine hitting 4th (.752 vs LH). Mac changed that. Good for him.

Once again anectdotally…another game. Sunday vs. Atlanta.

Braves with one run lead, two out eighth inning. Soriano faces Griffey with Dunn on deck. Nursing the lead…Griffey walks. Dunn up, with Hat on deck. Soriano, trying to avoid putting the tying run in scoring position with a good hitter on deck (Hat), gets a bit too much plate to a OPS 1.000 hitter. You know the results.

Anectodes, yes. But it’s foolish to think that opposing managers won’t try to take your best hitters out of the game.

The other argument for “splitting the lefties” is to nullify the LOOGY. Fine, the opposing manager will do a couple of switches and you get one chance (perhaps) for a favorable matchup (Like Dunn vs. Miller last night).
However, compare that to three times of having your best hitters together. It’s a matter of trying to win the game out of the gate or (maybe) getting a favorable matchup later on.
CYA managing.

As for your points

1st) Your sample size is WAY too small. Too many variables.
2nd) True…only one player can be protected. Dunn or Griffey. But BOTH can be left unprotected. Actually, I’d say Hatteberg can offer some protection to another lefty or Hamilton when he comes back.
3rd) Absolutely, Phillips is benefiting from hitting between Griff and Dunn, much the same way as Aurilia did last year. However, management is making a mistake if they consider BP a RH power hitter. His .729 OPS vs. RH (with Dunn or Griffey behind HIM!) tells you that. BP has one more homer than Carlos Lee. But frankly Carlos Lee is a RH power hitter (OPS .857 vs RH) and BP is NOT.
Opposing managers are NOT confused on this issue.
4th) I think I addressed this above. Pitching “around” is not IBB.

I’ve got more on this subject, but this post is long already.

thanks for the chance to reply.


One thing is certain, the sample size so far is small. It will be interesting to follow the numbers if this pattern continues for the rest of the season. Over 15 games the numbers may not tell much; but over 80 plus games it may be another story altogether. Or they may not.