Filter posts by...
Batter_v_Pitcher
Thursday, February 19, 2009
Colin gives it his all:
...but assume a league average batter faces a league average pitcher (using 2008 numbers, an OBP of .333). If a batter has been hit previously in the inning, the expected OBP for that plate appearance is .334. Not a huge effect. Apparently the effect of being fired up is an extra walk/base hit every thousand plate appearances.
But what of the hit batsman himself? ...Psychologically, it makes sense that the batter would be more likely to be scared if he were facing the same pitcher, rather than a new one, and the memory will be freshest on the same day. The effect was not significant, but favored the pitcher, and dropped the batter to a .326 expected OBP.
Probably would have been better to figured his wOBA, as his power might be the thing that would drop. Anyway, when a batter faces the same pitcher each time through the order, he gains 8 points in wOBA. So, the pitcher gets quite an advantage here. However, Colin reports the result as not statistically significant. (How much is 1 SD Colin?) Nonetheless, a great idea to research.
Tuesday, December 16, 2008
He said:
If I did happen to move throughout the lineup, my approach would only change depending on the situation that time at-bat. When I’m leading off, I only lead off once a game. The rest of my at-bats deal with the score, outs, and if there are runners on base.
Exactly the way I think. If Granderson is typical of his peers, this whole “I’m a leadoff hitter” cr-p is, well, cr-p. As he correctly surmised, it’s not important where in the batting order you are, but what the game state you are faced with. (Though, I will include that you’d like to know who is on deck and how fast the runners are, two things that can be inferred by your batting slot.)
In any case, this is what baseball is about. Figure out the game state, figure out how much impact a strikeout, walk, single and HR have relative to each other for this game state (including the count) and adjust accordingly (while your opponent also adjusts, and you adjust to his adjustments as he does as well).
Tuesday, November 25, 2008
Cool article by Pizza. Here are my comments:
My recommended method is what Pizza is doing (Odds Ratio). As Pizza said, in the “something better than nothing” category, you can use the differential method.
Since we know that there is an in-game advantage for the pitcher the first time meeting, it is not a surprise that the first time they meet ever happens to also match the fist time meeting in any game.
So, I’d like to see a control for the meeting within game, and for career. That is, if you look at the 23rd time meeting in a career, and it was the 1st, 2nd or 3rd time meeting in a particular game, how does the graph look? Is the function almost entirely an in-game effect?
Finally, sample size almost certainly accounts for the up-and-down at the end of your graph. Perhaps you can show what the +/- 2 SD range is, based on the binomial, for the number of PAs you have at each level.
Friday, November 21, 2008
Nice.
However, more important than the “at” pitch count is the “through” pitch count. The “at” pitch count tells you what the guy did at that count, if you knew the PA ended there. So, for example, is it any surprise that at the starting count of 3-0 (shown as ending count of 4-0 in the linked chart), and you knew the pitcher made just one more pitch to end the PA, that almost all the PA were walks? The “through” count tells you that starting at 3-0, this guy walked say 40 times, but he also went to 3-1 xx number of times. And then, in those xx times, he yada yada yada. That’s why I prefer looking at the “through” pitch counts, shown as “after” on baseball-reference.
Paraphrasing Homer’s love for donuts.... “Mmmmm… is there anything b-r.com can’t do?”
Tuesday, September 09, 2008
Good post by Peter:
My preliminary studies have shown that if I define three categories: intentional walk, strategic walk, and non-strategic walk, then I get more accurate results with run values set as .10 for IBB, .27 for SBB and .33 for NSBB with about 25% of the non-intentional walks falling in the SBB category.
The idea here is accurate, and really represented by win values of the walk.
We can see a more basic version here:
http://tangotiger.net/RE9902event.html
Where we only consider the base/out state (we should also consider inning/score).
On average, the run value of the walk is roughly 10% more extreme than the out, and of the opposite sign. So, if the run value of the out is -.30 runs, then the run value of the walk would be some +.33 runs.
If we go to the above link, we can see that the run value of the walk with bases empty and less than 2 outs is some 50% higher than the run value of the out. And with 2 outs, it’s some 10% higher. So, we can say that a walk in the less than 2 outs, bases empty situation that those are not strategic walks.
The more basic situation is the one with 1b open (runner on 2b) with 1 or 2 outs. In those cases, we see that the run value of the walk is some 40 to 50% LOWER than the run value of the out. These we could classify as “strategic” walks, especially if they were given to good hitters.
So, I agree that there are nuances to the run value of the walk, and calling them as Peter is doing is a great way to make the point, and to show which pitchers/batters are using the strategy. But, overall, the run value of the walk remains what it is.
Monday, August 25, 2008
Similar to my “Fixing VORP” blog post, treat this post as a “Sabermetric Announcements of Public Service” (with the unfortunate acronym of SAPS). John Dewan talks about long and short atbats, and the profile report is now part of Bill James’ site, to which I replied:
John, Bill:
I’m not sure that your readers appreciate the bias in the long and short AB data. What an AB of 1-pitch represents is: “What happens when the player makes contact on the first pitch” (*).
If the batter swings and misses at the first pitch, it does not count toward the 1-pitch data, and instead will now be part of the 2-pitch data. (Similarly, a called ball on the first pitch will remove it from the 1-pitch data.)
As the league data shows on the 1-pitch at bat (very high) it makes it seem like everyone should be swinging on the first pitch. But, the data itself is the *result* not the *approach*.
If on the other hand you asked: “What happens when the batter swings at the first pitch”, you will get different data, because now this will include the swing&miss.
You guys are probably aware of all this. But, seeing posts on outside forums, and lots of readers simply are not thinking of it in those terms.
You might like this chart:
http://tangotiger.net/halejon/allcounts.html
And, Linear Weights by the Pitch Count provides a great basis for figuring out the pitch count approach. But, that’s topic for another discussion.
(*) Foul balls that are not caught provide a tiny wrinkle to my statement. The general idea holds.
Tom
Tuesday, May 13, 2008
Some great data by Pizza, on the relationship between pitch counts and performance. If the numbers look low he notes: “Again, these numbers are lower than might be expected due to some of the methodological problems I ran into. If I have a moment I might try to correct for it.”
Regardless, the pattern is fairly plain to see. Roughly speaking, it looks to be almost 2 wOBA points per 10 pitches thrown. There are roughly 33 pitches thrown per time through the order, so that gives us an average change of roughly 6.5 wOBA points, each time through the order. In The Book, table 82, I show that each time through the order shows a difference of 8 wOBA points. So, fairly close.
Pizza: can you add a parameter for “time through the order”? Table 80 makes it seem like there is a definite jump each time. Perhaps your results are smoothed out what may be a staggered effect.
Friday, May 09, 2008
This is where game theory and PITCHf/x will collide.
Suppose: if you know the batter knows the data, then you make a change to your approach. But, since the batter knows you know he knows, and he knows you’ll change your approach, the batter’ll change his approach. But, what if you don’t know that the batter knows the data? Do you presume the batter doesn’t know and keep pitching the same way? But, if the batter actually does know the data, but the batter knows that you don’t know that he knows, then he’ll cream you.
Question: are you better off if everyone knows, or are you better off taking the chance that he might not really know? That is, might it be to your benefit to know 100% that everyone has the data and compiled in the same way you have it, or is it to your benefit to have that data well-compiled, while the other guy may or may not have it, and you have no way to know whether he has it well compiled?
Someone can insert the Princess Bride youtube clip right about.... now!
By , 04:07 AM
Wed. night in the 9th inning of the Tex/Sea game (not that anyone would be watching that game), C.J. Wilson, the Rangers’ regular closer, came in in the 9th inning to preserve a 2-run lead. (Padilla had pitched a very good game, throwing 96 and 97 mph gas!)
Wilson pitches to the first batter and goes 1-0. Remember that Texas is leading 2-0, there are no outs and no one is on base.
Read More
Thursday, April 17, 2008
A great idea by Jeff, at looking at hitters with a pitcher on deck and not, with dramatic differences. (Jeff should have noted the IBB, but that’s the least of the impact.) Great stuff, and time to look at multi-year data.
As you know, we looked at “protection” in The Book, and we do see changes in hitting/pitching approach, but the overall production didn’t really change.
One thing I noticed in The Book is how the #5 hitter hits better than he does in other spots, presumably because the #3 and #4 hitters wore him out. There’s lots of great studies to be done in terms of the on-deck impact.
Great job Jeff.
Thursday, March 13, 2008
That seems to be what the data is showing here. It is of course a silly thing to “never” do something, as that’ll get exploited in short order. You’d hope anyway. I’m not sure the last time that Mike Piazza has swung at a 3-0 count. Must be back in the 90s. But, pitchers were scared of him the whole time. Anyway, the A’s “pass through” performance once they reached a 3-0 count shows that they are as good or better than the league average. It seems that, for now, no one’s been calling the A’s on their “bluff”.
Do you PITCHf/x-ers want more work? Look at hitters who almost always swing (break them up into power hitters and not). Where in the strike zone do the pitchers pitch to them? Now, look at hitters who swing ALOT on 3-0 (again, break them up into power hitters and not). Where in the strike zone do pitchers pitch to them? (Compare that to some neutral count, like 0-0 or 1-1.) Are the pitchers smart enough to realize that the 3-0 always-take hitters are in fact always taking and therefore are pitching them more down the middle?
Of course, a player is not a robot, and will start to adapt if pitchers throw too much down the middle. So, what we have to do is identify the hitters as of today, let the world know about it. And let’s see who starts to adapt in 2008.
THIS is what baseball is all about. Thank you PITCHf/x.
Tuesday, September 04, 2007
More great stuff from John Walsh.
The pitch-by-pitch approach is one of those projects that I’ve had on the backburner for a long time. It becomes an exercise in how a batter and pitcher should approach each count, based on the skill ranges of himself and his opponent, and game theory (or expectations that everyone has, and how to leverage that). If you focus on what John quoted Ted as saying, “the single most important aspect of hitting was getting a good pitch to hit.” That’s the simple version. But, if you try to expand on it, you might have threshhold levels of say +.01 runs at the 0-0 count, +.10 at the 2-0 count, and -.07 at the 0-2 count (all numbers for illustration only). That is, your strike zone expands and contracts after every pitch. And, like I said, you’d modify all that based on the particular parameters we have in hand. It’s all a matter of when you should swing, and how quick you should swing.
On a related topic would be when to bring in your ace. Perhaps in the 7th inning, you’d need an LI over 4 to make you think to bring him in. Maybe in the 8th inning, that goes down to 3. With 5 outs to go, you might want 2.7, with 4 outs, you might want 2.1. In the 9th, maybe 1.8. If he didn’t pitch in two days, maybe you drop all those numbers by 25%, etc. (All numbers for illustration only.)
It’s all a question of optimization, with a whole set of dynamic variables, which themselves may be hard to quantify.
Thursday, May 31, 2007
Pizza Cutter offers a study in By The Numbers on measuring plate discipline. I’m afraid it’s going to take me several reads to understand it. It looks cool. You can get a link to his article, and talk about it on his blog:
http://mvn.com/mlb-stats/2007/05/24/the-adam-dunn-debate-defining-plate-discipline/
(Comments closed.)
Tuesday, April 17, 2007
By , 09:57 PM
According to Jeff Brantley during the CIN/MIL game on Tuesday, “smart baseball” is to take until you get a strike. Not withstanding the fact that you probably don’t want to be asking advice on strategy from Brantley, is this true?
Read More
Tuesday, January 16, 2007
George Brett said this about Blyleven:
Bert is up there with the toughest four or five guys I faced in my career.
A look at Baseball-Reference tells us that:
Read More
Monday, June 26, 2006
An NY Times article on Jason Giambi’s approach to hitting.
A while ago, I did (ubpublished) research on hitting at different counts. Mike Piazza, at the time and from the years I had data, never swung at a 3-0 count. Never. And yet, pitchers still threw him a ball 40% of the time. One of the pots of gold in research is how to hit and pitch at each count, given the strengths and weaknesses of the participants, as well as the game context.
from Amazon
Recent comments
Older comments
Page 5 of 342 pages « First < 3 4 5 6 7 > Last »Complete Archive – By Category
Complete Archive – By Date