Fascinating article in FG by Jeff Sullivan about the way Sox hitters are going about their business differently this year. Short version: they're handling pitches out of the zone the same as ever -- i.e., well -- but they're being much more aggressive with pitches in the zone, especially early in the count.
There's a graph of their early-count Z-Swing rate since 2008 that looks like the famous climate change hockey stick: puttering along with minor variations till 2018, then shooting up like a rocket.
I agree. This has some very interesting analysis that's potentially impactful.
More importantly, his arguments for this new approach being better than the previous more patient one are also very convincing since the pitching (and I would argue defensive) environment has evolved.
Sullivan concludes:
"Again, in many seasons, the more patient approach worked just fine. The Red Sox have fielded some extremely productive lineups. This looks like it’s going to be a differently-productive lineup, and perhaps one that’s more suitable for the era in which it plays. Driven by J.D. Martinez, this lineup has a new feel to it. Let this be a word of warning to future get-me-over strikes."
Agreed - I think this is all flowing from Cora's stated philosophy of "hunting strikes," and being aggressive rather than passive early in the count - as he told Mookie, being passive led to Houston's gameplan to have their pitchers throw him a first-pitch strike every time because he was so likely to take it.
As I noted in the game thread, Mookie is swinging at just 20% of early (0-0, 1-0, 0-1) pitches this season, compared to 21% last year and 29% in 2016. He is swinging at more first-pitch strikes (40.4%) than last year (33.4%), but at an identical rate to 2016 (40.4%).
So it really doesn't look like Mookie's being more aggressive early in the count—he's actually swinging slightly less often early on. But he is identifying the strike zone better than last year (and essentially on par with 2016). Or something. The samples are pretty small, and once you start slicing them down it's not clear how much significance they really carry.
I'd be very interested in comparing player-to-player the stats Fangraphs discusses (as I did for Mookie above). Nuñez's early swing% is about 7 points higher than Pedroia's was last year, and Martinez has an extremely high 47% early swing%; those two personnel changes may account for most of the difference, with little change in aggressiveness by holdover players. That's perhaps a shift in organizational philosophy in terms of personnel selection, but it doesn't illustrate that last year's players are taking more hacks early in the count.
Just picking 2 more players who had a significant number of ABs both years: Hanley is swinging at 38% of early pitches vs. 40% last year. JBJ is 36% this year to 37% last year.
You'd have to run through the holdover players and aggregate their comparative swing% weighted by PA (and maybe by balls vs. strikes thrown) to start being able to say anything conclusive, obviously, but I'm not seeing anything obvious to support the idea that players are swinging the bats more early on.
(You might also want to control for month—at least eyeball April vs. whole-season numbers across the league for a couple of years to see if there's seasonal variance).
(Also, the fact that out-of-zone swinging is essentially unchanged across the team suggests there's something more at play than increased aggression, unless you think we suddenly have perfect strike zone recognition.)
https://baseballsavant.mlb.com/statcast_search has the ability to find these numbers, but it's a bit of work. Set the season, set group by “league”, and set Count to 0-0/0-1/1-0. Select a batter and run the query to get the denominator (total number of early count pitches), then hit back and set pitch result to “All Swings” to get the numerator. That'll give you the early swing %; add in Gameday Zones->Pitches in Zone if you want the strike-only numbers.
EDIT: I'm looking at these numbers more, I think something may be screwy in my math above.
EDIT 2: Nope, looks good.
As an opposing pitcher, how would you attack this? It doesn't seem to have any obvious weaknesses if the hitters can execute the plan. Swing hard at strikes, no matter the count. Don't swing at balls, no matter the count.
I remember my little league coach telling me to do just that. He also had us take until we got a strike when the guy in front of us got walked. Maybe less applicable at the ML level.
As an opposing pitcher, how would you attack this? It doesn't seem to have any obvious weaknesses if the hitters can execute the plan. Swing hard at strikes, no matter the count. Don't swing at balls, no matter the count.
I remember my little league coach telling me to do just that. He also had us take until we got a strike when the guy in front of us got walked. Maybe less applicable at the ML level.
That's kind of the point. It forces pitchers to throw pitches that are hard to hit but be called strokes and not just once or twice in an at bat, but several times.
Pitch recognition is the key. If you cant recognize what will be a hittable pitch before you are committed to swinging the approach fails.
I imagine the league will adjust by throwing fewer fast balls earlier in the count. Most hitters look FB early in the count.
Not sure we have faced many tough starting pitchers or bullpens though. Ohtani was hurt. Did pretty well against Severino but he may have been tipping pitches
Well, maybe more like however you’ve been attacking every other team in the majors for the last 10 years—all those teams that have been sporting a higher z-swing% than the Sox.
It’s all a game of adjustments. These articles suggest BOS caught the rest of the league flat-footed by changing their approach. This would certainly help explain the success so far— we all hoped they’d be more potent than last year— but: leading MLB in runs? Will be interesting to see how opposIng pitchers adjust.
Okay, so if you actually pull all the players in then they are being a little more aggressive this season. Indeed, the first 3 I picked are the only ones who are down in early swing%. It's a significantly smaller difference than the article implies, but there's something there.
Pulling all of last year's starters who are still on the team, early swing% (total weighted properly by early-count pitches seen):
As I noted in the game thread, Mookie is swinging at just 20% of early (0-0, 1-0, 0-1) pitches this season, compared to 21% last year and 29% in 2016. He is swinging at more first-pitch strikes (40.4%) than last year (33.4%), but at an identical rate to 2016 (40.4%).
So it really doesn't look like Mookie's being more aggressive early in the count—he's actually swinging slightly less often early on. But he is identifying the strike zone better than last year (and essentially on par with 2016). Or something. The samples are pretty small, and once you start slicing them down it's not clear how much significance they really carry.
One of the things FanGraphs noted was that Mookie was also hitting the ball in the air (launch angle increased by 10.7%) and has also been pulling the ball more, which is relevant because his wRC when he pulls the ball is 242. When he goes oppo? It's about 1/10 that---25. They also note he's becoming more selective about avoiding swinging at pitches in the outer part of the zone---which are the pitches that he is going to have the most trouble pulling. So maybe its less about being "aggressive" early in the count but recognizing that if he gets something he can pull and in the air.....swing.
I do think the key concept here is being selectively aggressive. It's not simply a matter of going up there and swinging at strikes early in the count. It's looking for a pitch you can drive. If it's a strike but a pitcher's pitch - nasty break or on the black - let it go. But if you get your pitch, jump on it.
I think its a bit asinine to say look for a pitch you can drive if its a strike but leave it if its a nasty pitch. It just aint that easy, and if it was, that approach would be relevant throughout history. History shows that batters lose the battle on average to pitchers, and thats for a reason
my take on this new approach might be related to things like using analysis to identify situations where pitchers tend to fall into predictable patterns and groove fastballs over the plate, like when theyre behind in the count for example, I've thought for a long time that it is interesting that the standard strategy on 3-0 counts is to take, after all, pitcher is struggling to find the plate, you have a higher chance of a walk and you might just make an out if you make bad contact, but on the other hand, it's interesting how often batters just take and then look at a meatball in the centre of the plate for example. Probably a lot of pitchers have strong tendencies to try and get strikes over in certain situations, early in the count etc
I do wonder if this is one of those "metagame" things where we had a long period of hitters just taking pitches, and driving pitch counts up, seeing lots of pitches as a residual benefit on top of pushing for a high OBP type strategy, and there was a correction towards grooving more strikes, and now theres a counter correction to starting to pre select specific pitches to expect balls in good hitting spots. It would probably have been a more interesting post if I'd the time or skill to look at some analysis of pitch fx data to see if the evidence is there for a noticeable change in strike zone strategy related to the "working the pitch count" strategies that were in vogue a few years back during the moneyball boom.
I think its a bit asinine to say look for a pitch you can drive if its a strike but leave it if its a nasty pitch. It just aint that easy, and if it was, that approach would be relevant throughout history. History shows that batters lose the battle on average to pitchers, and thats for a reason
my take on this new approach might be related to things like using analysis to identify situations where pitchers tend to fall into predictable patterns and groove fastballs over the plate, like when theyre behind in the count for example, I've thought for a long time that it is interesting that the standard strategy on 3-0 counts is to take, after all, pitcher is struggling to find the plate, you have a higher chance of a walk and you might just make an out if you make bad contact, but on the other hand, it's interesting how often batters just take and then look at a meatball in the centre of the plate for example. Probably a lot of pitchers have strong tendencies to try and get strikes over in certain situations, early in the count etc
I do wonder if this is one of those "metagame" things where we had a long period of hitters just taking pitches, and driving pitch counts up, seeing lots of pitches as a residual benefit on top of pushing for a high OBP type strategy, and there was a correction towards grooving more strikes, and now theres a counter correction to starting to pre select specific pitches to expect balls in good hitting spots. It would probably have been a more interesting post if I'd the time or skill to look at some analysis of pitch fx data to see if the evidence is there for a noticeable change in strike zone strategy related to the "working the pitch count" strategies that were in vogue a few years back during the moneyball boom.
Well, to build on and respond to this, some people are theorizing that working up pitch counts on starters could becoming less beneficial as mid-relief gets better. Not as much advantage to chasing the starter out; still some to be sure, but not as much. So the key is relative efficiencies and inefficiencies--getting to the relievers to take your shots becomes less valuable relative to trying to hit starters, so you adjust hitting strategy.
Gomes and O'Brien were talking the other night about the idea that launch angle up is a response to the shift working. These two things strike me as similar approaches, and suggests that a lot of the analytics are leading to real "chess matches" across time in figuring out the next response.
Which is cool. Like, how often do we watch baseball strategy evolve like this in front of us?
Oh, thats a great point I'd not even thought about (expansion of the middle reliever roles), I think you're right, it's an amazing thing to have a sport where there is a real evolution of thinking as the impact of "big data" analysis and number crunching has become the norm. All kinds of things that the average fan might not necessarily think about, it's such a shame that in all the countless web blogs and articles, and even in the wider media, there's not more of this, or at least room for a few examples, like imagine watching a game with someone like Rob Neyer or Joe Posnanski alongside a more traditional game commenter, able to bring up stuff like this to do with fielding zones, launch angles etc etc. Its fascinating for sure, it probably happens in all sports if you are into it enough to look deeper but something about baseball and its isolation of the "one on one" aspects really make it shine, I think
Simmer down, cowboy. I didn't mean to suggest that looking for a pitch to drive early in the count, and laying off the nasty stuff, is "easy." But the early results suggest that they are having improved results, based in part on improved plate discipline/selectivity, and perhaps less of an emphasis on driving up pitch counts (which really used to be their thing). And I've said elsewhere, I think Cora and his staff have the analytical data and know how to use it, so I don't doubt that some of this includes data on tendencies/preferences so that batters go up their with a better idea of what to look for in various counts, which of course could help them in hunting a particular pitch early in the count.
Going from the days of "The Greek Gods of Walks" of Moneyball lore, to the current reign of "The Greek Gods of Pounding the Crap out of first-pitch Fastballs" is definitely an improvement from the fan's perspective.
Hi! It looks like you are not logged in. SoSH has a bunch of content only available to users who have registered so we suggest creating one.
If you are seeing this because of a supposed issue using the site, all site outage notices will be posted to our official Twitter account at
SoSH on Twitter.
Thanks!
