Arm Slots / Sides / Pitches

[Coriolanus]

There's a bit of old baseball wisdom that facing a bullpen of pitches with different arm slots, pitches and sides (r/l) makes it harder for hitters to get comfortable.

Has anyone found more than anecdotal evidence that varying arm slots/sides and relievers' best pitches impacts outcomes? What about for starters? If you throw all 3/4s but one, does that one OTT get a benefit because of novelty/unfamiliarity, as is the case on some level with knuckleballers?

[Argonaut]

I've done some testing on this, so you can look through my post history on threads I've started.

The short version of what I've learned in response to your questions... arm slots affect the splits you'll see between RHBs/LHBs that the pitcher faces. The more extreme armslot, the more pronounced splits will be in general. So a submariner LHP makes a great LOOGY. Over-the-top is the most neutral arm slot.

Pitch types matter, but maybe not in the way you'd expect. Most of the time it's safe to just look at the Stuff rating, but if a pitcher has a Knuckleball (and not a Fastball), then he will get an overall decrease in his BABIP allowed.

[Gruber_Tagged_Him]

Originally Posted by Argonaut

I've done some testing on this <snip>

SOME testing, eh? Please warn us all when you are about to do a LOT of testing, as we'll want to avoid being within a 5 km radius of your computer when it goes critical...

[Coriolanus]

Originally Posted by Argonaut

I've done some testing on this, so you can look through my post history on threads I've started.

The short version of what I've learned in response to your questions... arm slots affect the splits you'll see between RHBs/LHBs that the pitcher faces. The more extreme armslot, the more pronounced splits will be in general. So a submariner LHP makes a great LOOGY. Over-the-top is the most neutral arm slot.

Pitch types matter, but maybe not in the way you'd expect. Most of the time it's safe to just look at the Stuff rating, but if a pitcher has a Knuckleball (and not a Fastball), then he will get an overall decrease in his BABIP allowed.

I was aware of your studying this in individuals pitchers etc., I was actually wondering if anyone had looked at sequencing, i.e. does it matter if you follow an over-the-top pitcher with a sidearmer? A lefty with a righty?


Because common sense dictates that if my entire squad is 3/4s righties with FB, Curve, and a Slider or Changeup that there would be a certain "familiarity" on the batters part compared to when the ball is coming from many different arm "holes" and with a different menu of pitch movements



I was wondering if anyone had seen evidence this is modeled into the game or if the devs had said as much.


anecdotally on a tangental subject it seems that fastball/changeup is the better of the offspeed combinations if not taking the batter's handedness into account; it feels like you suggested this as well, but I can't remember

[Argonaut]

My gut feeling is that the sequencing that you're talking about isn't reflected in the game. I think it'd also be kind of hard to test quickly and/or properly.

The Fastball/Changeup thing might be an interesting one to test. I'm fairly confident these pitch types affect stolen bases, but as for them being a better combination than "expected", I couldn't say.

[sabrtoothtiger]

small sample size but I have noticed Kazuhisa Makita (a submarine delivery) has had increased strikoeut numbers while playing for my team. It may be slight but I believe compiling a pen of multiple deliveries may help extract the most out of each players skills. The impact on others cannot be easily quantified.

[Dyzalot]

From my experience OOTP seems to only incorporate those things that have empirical, statistical analysis and numbers behind them. So unless someone like Bill James has done an in depth analysis of this particular phenomenon and proved its existence, I doubt it is in the game.

[Coriolanus]

What makes you think it ISN'T something that is discussed and studied in SABR circles?


I took about 20 seconds to find this article from September from Baseball Prospectus, the spiritual kin of the ole Baseball Abstract.


here's the article I found. it's part of a series apparently.


https://www.baseballprospectus.com/n...er-sequencing/

[Dyzalot]

Originally Posted by Coriolanus

What makes you think it ISN'T something that is discussed and studied in SABR circles?


I took about 20 seconds to find this article from September from Baseball Prospectus, the spiritual kin of the ole Baseball Abstract.


here's the article I found. it's part of a series apparently.


https://www.baseballprospectus.com/n...er-sequencing/

Never said that it wasn't discussed. I just don't believe we have any strong statistical samples that show any kind of strong correlation. Certainly the link you gave me didn't since...

To close this portion out, here are a couple of things to think about, based upon these two simple scatterplot charts. They show a mediocre, at-best, correlation between winning percentage and whiff rate and chase rate in different relief sequences.