I just don't follow your logic here, Sir Charles. You're re-writing the historic baseball philosophy with that one.
I just don't follow your logic here, Sir Charles. You're re-writing the historic baseball philosophy with that one.
Rounding third and heading for home...
No, what that lineup maximizer is telling you is what your odds are for producing runs with a reorganized lineup IF EVERYONE PRODUCES EXACTLY AS THEY HAVE BEEN. Once you reorganize them...you don't know how they'll produce as the situations have changed. If this game was as easy as the computer models made it out to be, everyone would be following them. It's not. Some hitters handle pressure situations better than others, some batting slots induce more pressure situations than others. Handedness does matter, probably not as much as many think, but it does and that also plays a role whether we want it to or not.
I'm not against the use of computer models and such...but I am against it when people come out and state that by following such a model our run output would increase by such and such per game...because that's just crap and a flat out guess based on speculation. It's just more information. Not a bad thing, but also not something to take as gospel on how things should be done.
oneupper (07-08-2013)
Barry On Baseball Also blogging at Banished to the Pen.
Let me be clear, I'm not saying I want Cozart in the 2 hole. I'm simply saying that you can't just look at the top 4 hitters and put all your focus on that portion of the lineup without realizing what the impact will be on the REST of the lineup. If I was a pitcher facing the Reds, I'd much rather face those 4 guys in a row and then all the weak hitters in a row. I think it would make my job MUCH easier.
You might want to sleep on this on Sir.
Some of your subpoints make sense but the whole not wanting to stack your best hitters together is quite puzzling.
You are conflating actual production with projections. Of course it isn't a guarantee. Neither are the odds in Vegas, but people use them because they have a certain % likelihood of success.
These are projections based on statistical probabilities. I prefer to put my money on a system that calculates run production based on the career norms of players and what they will do when put together in a certain order. That system says that 1000 games of a Choo-Votto-Bruce-Phillips lineup will outproduce 1000 games of a Choo-Cozart-Votto-Phillips lineup.
The following lineup would be above average at every lineup spot except 8 (based on NL ops)
Lineup spot ops NL average
1 Frazier 744 727
2 Paul 727 709
3 Votto 943 823
4 Choo 872 792
5 Bruce 821 750
6 Phillips 735 724
7 Rocco 698 683
8 Cozart 631 659
I do like stacking our best hitters together...but not if that means stacking your worst hitters together and possibly making the WHOLE worse than the sum of the parts.
When our good hitters are only getting on around 35% of the time and our bad hitters are getting on around 25% of the time...is that difference really going to make a positive difference over the course of a game when it allows a pitcher to regroup as I described? I'm not so sure.
Yes it is. And he still makes an out like around 70% of the time. Hitting is hard. So yes it does allow for a chance to score. But having someone else in that spot (no I'm not suggesting hitting someone in Bruce's spot exactly), the odds of making an out increase by what...5%? 10? But those odds depend upon sooooooo many different factors that to just state a percentage is kinda nuts. How is the pitcher throwing that day? How does that batter fare against that particular pitcher? Handedness differences? Batter on a hot streak or struggling? Day or night game? How long since a day off? etc, etc, etc, etc. A lineup optimizer doesn't take all that stuff into consideration.
And after all that is said and done, the odds of scoring there still aren't great. But the odds of us NOT scoring the following inning are VERY good. And the next inning.
My point is that it's not as simple as many here are making it out to be. And Dusty may not be as wrong as many here are making him out to be either.
I was saving this because I know the discussion angst it causes but it's interesting. Note, I wrote this after yesterday's game so I don't believe it includes that. Nevertheless:
I know it's a curiosity but I plugged Dusty's most common lineup into the Baseball Musing's lineup tool and it said that lineup should product 391.587 runs through 87 games.
The Reds have produced 384 runs through 87 games.
The lineup tool said the most optimal lineup would produce 4.825 runs per game or 419.775 runs to this point: 35 more than we have which is about 3.5 wins.
(Note, we are 3.5 games back in the standings.)
If Dusty moved Cozart to 7th and everyone else up one spot, it predicts 4.630 runs/game or 402.81 runs up to this point. 18 runs or almost 2 wins more.
Givens:
- I used this year's stats so far
- It's a curiosity, yes, I understand this
While I'm all for getting better players, I think there's a (some superlative between nominal and significant) improvement gained by moving Cozart down in the lineup.
Mock if you must. Discuss if you can.
No, I'm not confusing them. It's projections based upon actual production. There's the problem. Depending on situations (batting order) the production is certain to change. You can't just juggle things around and expect the same results.
If people would say that there's a CHANCE of a 20 run per year increase...no big deal. But that doesn't happen here. There's just as equal of a chance that we score 20 FEWER runs a year with whatever proposed lineup is thrown out there. It's a coin flip.
And the problem again is with what you just typed. You only listed the top 4. Of course those 4 would produce better than the version with Cozart in there all things being equal. But when taken as a whole...it's MUCH less certain because you simply don't know how certain players will produce in certain spots in the bating order.
I'd want Cozart moved even if it meant scoring one more run per season. I know the numbers aren't everything, but they agree that this change is almost 100% likely to increase scoring -- and probably by much more than that. I think the Baseball Musings projection of 35 more runs is a bit too optimistic, but I think most of the players would get behind the idea of changing lineup "roles" if it meant more chances to win.
I find the counter-arguments don't have much to say other than some vague notion of how a change would affect the team psychologically. Well, knowing that your manager has studied run production in every way possible would be comforting to many players, too. I can't imagine someone like Votto is unaware of the probabilities. He's just keeping his mouth shut because it is bad form to break the chain of command.
“Every level he goes to, he is going to compete. They will know who he is at every level he goes to.” -- ED on EDLC
SC, these lineups are computed using the career stats of players. The slot a player hits in the order has minimal effect on his own stats, which usually will trend toward those career norms. What it does do is change the order in which those stats take place. That changes the situations in which your best hitters come to the plate.
The most painful example of this every day for me is when Cozart comes up with two out and Votto in the on-deck circle.
“Every level he goes to, he is going to compete. They will know who he is at every level he goes to.” -- ED on EDLC
Board Moderators may, at their discretion and judgment, delete and/or edit any messages that violate any of the following guidelines: 1. Explicit references to alleged illegal or unlawful acts. 2. Graphic sexual descriptions. 3. Racial or ethnic slurs. 4. Use of edgy language (including masked profanity). 5. Direct personal attacks, flames, fights, trolling, baiting, name-calling, general nuisance, excessive player criticism or anything along those lines. 6. Posting spam. 7. Each person may have only one user account. It is fine to be critical here - that's what this board is for. But let's not beat a subject or a player to death, please. |