hey everyone,
so back in october, before my son was born, i had grand plans to do lots of cool stuff. that never materialized because he was born and i don't know why i thought i'd have to time to do anything and frankly i don't even know how people raise multiple kids and still have any hobbies at all.
i put in some late nights the last couple of days to get back up to speed and to deliver part of what i promised, and all of what you guys support me for (i.e., pay me for). truthfully i am stoked about a couple of these updates so i hope you are too.
the leaderboard now includes spring training and postseason data. you can access it normally using date filters, but there is now a dropdown above the date filters that is currently toggled to "regular season." i have to do this so the japan games don't get mischaracterized as spring training given their timing. you can look at (for example) spring training by itself, or you can string the regular season and the playoffs together, etc.
the 'specs' tab has been split into 'release' and 'shape.' there are so many pitch specs to track, and i didn't want to have to artificially constrain myself to a certain number of them, nor did i want to have to change the dimensions of the leaderboard because i'm lazy, nor did i want have to introduce some weird sideways scrolling that nobody would've realized was there. this is, in my opinion, the best solution currently.
arm angle! it's in there now, on the 'release' tab.
there are now "dynamic dead zone" metrics. this is maybe my favorite new item.
if you pay any attention to pitching nerds you'll hear them talk a lot about "IVB." IVB is induced vertical break, aka the amount of lift (or drop) a pitcher imparts on the baseball in the absence of gravity. generally, more IVB is better. but a pitcher's ability to generate IVB is naturally constrained by his arm angle. so the objective should not be "generate as much IVB as possible" as much as it should be "generate as much IVB as possible within the constraints of your delivery." in other words, you cannot tell me a fastball with 19" of IVB is better than a fastball with 13" of IVB—not in a vacuum, at least; not without telling me the mechanics of the pitchers' deliveries.
(this is one of my biggest gripes in baseball analysis currently. more IVB in absolute terms is not always better, and anyone who presents IVB information to you this way should be met with skepticism until they provide you more information; namely, arm angle and velocity.)
additionally, IVB is a flawed metric. two pitchers can have identical release speeds, arm angles, etc. and have wildly different IVBs. how? one way is if they have profoundly different extensions. pitchers with longer extensions allow the pitch less distance to move. so if two pitchers impart identical accelerative forces on their pitches but one has a shorter extension, the pitcher with the shorter extension will "create more movement" because the pitch has more distance to cover and thus will have more time to move—which "creates" more IVB but is ultimately an artifice of the data.
all this leads me to the premise of "expected" movement. i've long held the theory that hitters have a memory bank of visualizations of tens or hundreds of thousands of pitches they've seen in their lifetimes. they have subconscious expectations about movement at most arm angles and velocities. this ties back to the previous premise, although it is not necessarily a truism: a pitcher wants to create movement that is substantially different from what a hitter can expect. this is why a 100 mph fastball is not necessarily good if its movement profile is not exceptional in anyway, but a 92 mph can be lethal. that 100 mph fastball lives in the "dead zone."
if anyone tells you there are certain coordinates (e.g., 16" of IVB and 16" of horizontal break) that define the dead zone, they are wrong. the dead zone is dynamic; it depends hugely on arm slot. max bay—formerly of astros R&D, currently of dodgers R&D, creator of stuff+, one of the smartest people that public analytics had to offer—created an app for this to visualize a pitcher's proximity to his slot-specific dead zone: https://dynamic-dead-zone.streamlit.app/
i have buried the lede, but: i now have dynamic dead zone on the leaderboard, in the 'shape' tab. the calculation is not identical to max's; i would not ask a master to reveal his secrets. but i developed my own, and i spot-checked it against max's, and i can tell you pretty confidently that it's very close to his, in terms of both magnitude and order/rank. you can do your own spot-checking and tell me if you find any big discrepancies; i'd be curious to see them. the metrics are called "V Acc OE (ft/sec²)," "H Acc OE (ft/sec²)," and "Pythag OE (ft/sec²)." V = vertical, Acc = acceleration, OE = over expected. Pythag = the hypotenuse of V and H. again, we're using acceleration here and not raw IVB because IVB is misleading.
i'm stoked because it will be super cool to (1) compare across players and (2) be able to filter by all the usual filters, like date, but also by batter handedness, pitch types (although this should probably only be used for fastballs), etc. etc.
there are a couple of other new things; there's a uniqueness score, which basically compares a pitch's specs to others and creates a score for how unique it is. although it is scored as "0-100," 50 is actually mundane, and 0 or 100 is unique. i should probably make that clearer somehow. i'll work on that.
somewhere down the line i would like to fix my "location runs" metric but it's pissing me off so i removed it for now. TBD.
i have talked your ear off but i hope i have made things up to you by basically giving you a free fangraphs post about dead zones.
one last thing: you may notice that the password field is currently populated. i am allowing non-subscribers to view all the stuff they can't normally see without a password to entice them to perhaps subscribe after they go back behind the paywall. it's called a marketing gimmick. bear with me while i allow people to see what they're missing.
thanks everyone for being so patient this offseason. i really, really wanted to have the leaderboard live for spring training so folks could do crucial fantasy research, but it just didn't happen, and i'm sorry. amazingly i didn't lose too many of you through canceled subscriptions, and i hope these new additions make the very long wait worthwhile.
if you are seeing this email and you are a one-time donor or a former subscriber, maybe this entices you to come back.
happy baseball day to you.
love,
alex