CSS Button No Image Css3Menu.com
New! Search comments:
(NOTE: Relevance, Author, and Article are not applicable for comment searches)
Based on your final paragraph about management, you will surely be interested in this working paper from some close colleagues of mine:
In thinking a little more here--and hopefully Guy has some thoughts on this--it's important to note that the data on the number of In-Zone vs. Out-of-Zone pitches are specific to pitches called by the umpire.
This may or may not matter. Since the umpire doesn't need a prior if the batter swings, the prior that he uses based only on calls that need to be made would make sense, but I'm curious what Guy has to say.
Hopefully he and I didn't miscommunicate on what the data were made up of.
Umpires have been monitored for years and it's making a difference on accuracy:
I understood your point, and I was providing further information. Much of what you say here was the entire point of our retort in JSE.
I'll also add a bit to what our original paper reported. I fit a number of models with substantial control variables (including location of pitches) to evaluate discrimination that were not included in the published paper or its appendix. I found no statistically significant effects of race matching in the FX data, either (data through either 2010 or 2011, I forget specifically). I even fit these with separate race match variables for each race classification combination.
And Michael explains why it might be troublesome to look there (including my own estimates): all pitches are closely monitored after the implementation of F/X. We also make this note in our paper (IIRC, maybe it was removed) that if the interest is the differential effect of monitoring, then it seems strange to use a sample where all pitches are monitored very closely (the data are publicly available) to make this evaluation.
There is much more to do here with such great data. I continue to try and work with it and wish I could do it all!
Based on my own study of modeling a strike zone, 2,500 pitches tends to not be nearly enough called pitches to have a good idea of what that strike zone looks like even for a single umpire with less variability than an aggregate of multiple umpire zones.
No need to guess. It's not many minority umpires in general--and as we show in the appendices, the effect estimates are strongly affected by changing just one umpire race/ethnicity categorization. Here's a direct passage from our paper:
"Also of note is that even though the sample has nearly 8.3 million pitches, the number is reduced greatly when examining some combinations of umpires and pitchers only for pitches subject to judgment by umpires. Indeed, from 2004-2008, there were only around 2,550 pitches thrown by Black pitchers requiring the judgment of Black umpires."
Sorry. I clicked around the provided links and CTRL+F'ed him before asking, but apparently missed this. Thanks!
Steven Souza, oh my!
Any thoughts on this projection? 25-25 season with reasonable average. That's quite a monster expectation. Also found the play time projection interesting. No platoon expectation?
Give me a break, just having a little fun. Just ignore.
So, you're saying that I'm in good shape in my 20-team keeper league since I have 8 guys on this list (including 4 of the top 6) in addition to Mookie Betts, Steven Souza, and Wil Myers, yes?
Still going to take some patience...
I agree here on treating the annuity like a lump sum in 2035 is wrong. But your analysis is also incorrect.
You must realize that Bonilla waited 11 years before beginning to receive his 10.8 million in PV. So you need to backdate the 10.8 million PV of what I am assuming to be 2011 dollars, to 2000.
Basically, the annuity is equivalent to a lump sum in 2011 of $10.8 million is what you're saying with the 10% annual nominal rate (and again, we're for some reason leaving aside the real rate here)
So you could treat your 10.8 million PV of the annuity as a lump sum in 2011, backdate it 11 years to 2000, and we end up at
(10.8)/(1.1^11) = $3.79 million
But we know that the real interest rate is probably lower, which means this was much closer to the $5.9 million in 2000 dollars that he was owed and everything seems to work out fine for both sides (leaving aside the Madoff investment issue, of course).
I'll note that maybe your implication is that the real return on S&P has been 10%, but that seems awfully high to me.
Thank you so much for putting this out there. I get rather annoyed with the nominal value thrown around all over the place.
One thing that seems strange to me is using 10% here, but not noting that it's the nominal rate. Why not adjust that for the inflation rate, and get the real valued PV? After all, even if we expect 10% yearly return on the S&P over the next 15 years, we also expect inflation to be above zero. Even the quick and dirty (Interest - Inflation) calculation would be helpful here.
Also, I'm glad you mentioned Bonilla. I go through this example in my Sport Finance course (with slightly different paramters).
No worries, I see that. Just trying to help with the communication of the really interesting result. I think maybe doing less distinct shades within-group would help is all, and ensuring color progression is consistent within each group.
With no balls, you go from red to yellow (dark to light).
With 2 balls you go from light blue to dark blue (light to dark).
I like what you did with 2 balls, keeping the same general color and using a scale. But the consistency would help.
As I said, though, this is really interesting stuff.
Also, as to the substance, the Good vs. Bad stuff is really interesting! There is some fascinating game theory application here, and it would be interesting to see this broken down by pitcher quality as well.
Thanks for the insight!
Pet peeve complaint here: no reason to have so many colors on your bar plots. If you instead made different colors based on groups of counts with the same number of balls, it would be much easier to see the effect of the upward swing likelihood trend as there are more strikes on the batter.
I'm glad you bring up the point that avoiding elbow surgery, irrespective of MLB prospects, is an important question to be framed.
My own experience with UCL injury has been that while it feels like a knife stabbing my elbow every time I throw a ball, it honestly does not hurt to do any other movement or motion in my everyday life.
Maybe I'm unique in that respect, but when I got my elbow checked out by the orthopedic surgeon, his recommendation was to leave it alone unless I was super serious about playing adult league baseball. I wasn't, and I didn't get surgery. But he did note that my UCL wasn't just torn, but was completely gone (along with some bone chips floating around in there).
I continue to play slow pitch softball, which hurts like hell, but at least I can give excuses for poor throws. Within about a week, nothing is sore anymore as long as I don't throw. And I can hop over to the gym and lift weights, do pushups, etc. with no issue.
Just a perspective on the everyday life of a UCL tear. I guess there could be a chance of longer-term damage if I continue to play slow-pitch. But thus far I haven't had any further issues (it's been about 8 years now since the initial injury).
In thinking about this further, in fact using the context dependent outcomes may actually increase value calculations of the better catchers disproportionately. Why?
Well, if better catchers are getting more strikes, on average, before that count occurs, then they are more likely to see those high-run-impact counts in the first place.
Since I don't have the data, I don't know how much effect they would have on the final tallies. But definitely something to think about.
As JRoegle says, base state and out state are also important from what I've seen.
Finally, thanks for the shout out, guys. I'm glad you found the mgcv package to be a big benefit. This is awesome stuff.
Good inquiry here, Zachary. I have a recent paper that suggests catchers get more favorable calls from the umpire (for whatever reason), and perhaps some evidence that it is less favorable when they aren't behind the plate.
I didn't have enough data to go very far on the front of how that actually affects performance (or whether the effect I saw was persistent once one leaves catching), but it might lead to one suggestion about what might be happening.
We like to think that performance increases because of *insert reason here*. Maybe the overall bat does. But, on the other side, perhaps the less favorable umpire calls ultimately make it a wash (but perhaps make careers longer, and allow solid performance at the plate for a longer period).
In any case, I have no idea if this is happening, but it's just another piece of evidence I found interesting. Thanks for sharing!
I had this ready to go this past year with my $9 Reddick-Gomes-Smith combo and then Reddick just decided to mash everything so I wasn't willing to take him out most of the year. Still might have been a decent rule for rotating these guys that outpaced Reddick alone, though (particularly since it has both AVG and OBP as categories).
(8x8, 20-team, 30 man roster, 3 DL slot, with 3 to 8 minor league slots separate from the 30, so Ryan Braun is valued at about $85 here).
"But we don’t currently have strike probability calculated on a pitch-by-pitch basis"
I don't have the data past 2010, but have most of this put together at a general level using my semi-parametric models I have talked about before (pitch type, count, location, right/left handed batter/pitcher, pitcher fixed effects) if you guys are interested. Easy to go through 2012 if someone has the data easily accessible and is willing to share.
Always good to note the league structure influences on player values. There are a couple things I believe that no valuation system (even very nice customized ones like the PFM here) out there handles well that I did not see here:
1) Daily lineup changes
2) Large benches
The first can be very important, because predictable LHB/RHB platoons instantly become more important. The second then comes into play, as you won't be able to find those platoons on the waiver wire.
I find these two issues are exacerbated by having H2H competition. Flexibility to stick in Juan Pierre for the last 3 days of a period when 1 SB behind (and 5 HR ahead) can be the difference of an extra category every period or two. The fantasy valuation systems generally have difficulty putting value on that flexibility.
Really cool stuff. Thinking a bit more from our contact before, I think the next step here would be to look at within-pitcher variation in hSOB as well. Across pitcher variability gives more of an idea of the spread in talent, while within pitcher variation (with significant regression to the mean) might give more insight to ability to control the outcome. I think that you do address this with the split-half correlations, but I'd love to see Observed and Regressed standard deviations for the players in the tables as well to get an idea of the spread as you do at the aggregate level.
Really awesome stuff.
Hey Mike, glad you highlighted this here!
I think the key to understanding the link between the 4 or 9 'zones' and a heatmap is this: a heatmap essentially does the same thing. The main difference is that the 'zones' are continuous throughout the entire strike zone, which requires the heat map to use a weighted average of adjacent data. This is essentially lots of tiny blocks like above, but with a weighted averaging between blocks.
But things can really, really break down when there is not adjacent data (edge of where the batter swings). This is why I have stuck to using heat maps exclusively for density of pitches thrown or for umpire calls. Sample sizes are sufficient and there are generally not artificial boundaries to the smooth. Even then, using a heat map for umpire calls should be 1) Cross-validated and 2) Use a lot lot lot of data. Without CV, comparison between umpires with different sample sizes will be difficult upon visual inspection (the entire point of a heatmap). Without lots of data, CV breaks down with certain types of analyses (especially in the binomial and ordinal realm), and in that case you have no idea if you are smoothing optimally.
In the end, pitch data is just extremely noisy. Location and pitch type matters less than knowing where the pitch will end up (a hypothesis by me). As a professional hitter, if you're a crappy outside pitch hitter, but you know it's coming, you can probably hit a fastball out of the park. We don't know "when" the batter guesses right on a pitch and even then, it's difficult to hit any pitch, which creates this large amount of noise.
Lastly: thank you for pointing out the color palette issue! While I often use a Blue-to-Yellow-to-Red palette (most natural interpretation), I have noticed that using a single color presents the best representation when attempting to interpret the smooth at a more granular level visually. In this latter case, there isn't a "breakpoint" of interpretation of the color scheme.
Really enjoyed this. I think there is a lot of work to do here both from the perspective of the sabermetrician as well as applications of this in catcher training and practices. As usual, Mike presents stuff that is really pertinent to those both in the front office as well as on the field. Keep them coming.
Not often I don't zoom to the end and just read a paragraph about the results of an analysis. Checked this one out from beginning to end...very nice.
I agree there is a lot to learn regarding batter approach.
I find this nearly physically impossible:
"However, if, with runners on first and third, the pitcher, while in contact with the rubber, steps toward third and then immediately and in practically the same motion “wheels” and throws to first base, it is obviously an attempt to deceive the runner at first base, and in such a move it is practically impossible to step directly toward first base before the throw to first base, and such a move shall be called a balk."
Who pulled that one off that they had to make a rule about it?
This would also be impressive:
"(f) The pitcher delivers the ball to the batter while he is not facing the batter;"
Lots of fun to read! Not much else to say. I look forward to the next installment.
Thanks for the comments! Definitely think there's added value in looking player by player, and certainly at pitch types. Hoping to check these things out in the near future. Definitely other angles to take at the 'approach to rookies' question.
Thanks for the comments. I agree with all of your points. I was hoping to go a bit further later on and look at individual player trajectories. And I generally look at location, rather than pitch selection here. There are absolutely more things to look at. One of the reasons I have focused on location is that I don't have pitch types clustered throughout the data outside of the Gameday stuff.
I am hoping to get into looking more specifically at individual player trajectories and rather than 'grooving it' vs. 'not grooving it', look more specifically at changes in average location (inside-to-outside, up-to-down). Going through each player that way is a pretty big project.
Thanks, Tango. Appreciate the comment.
A quick note: While Pitchers remain in the table at the bottom, I DO remove pitchers from the original analysis in the article.
I did it with them in as well, but it didn't really change anything (given the 'non-pitcher' result, this should be expected, as I doubt the pitching approach to batting pitchers is really going to change: just let them make the out).
CORRECTION: Did not mean to say "business taxes from owner income", just meant the owner's private income.
Just wanted to say this is a fantastic read, and quite fascinating. As someone interested in the economics of sport, I'd love to see more stuff like this.
Here is a question for you:
What about team owners' private income? Given that the firm is operating out of it's own state roughly half the time, have states looked to collect business taxes from owner income itself? Most owners have other streams of revenue and private LLCs to put the operating losses on the books with respect to the baseball team and get around taxes to begin with. Do states look to tax their general income (i.e. income from being Chairman of Starbucks) and attempt to specifically trace a portion of that to owning the sports team and operating in other states?