Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Friday, August 11, 2017

FiveThirtyEight: Baseball’s ‘Hot Hand’ Is Real

Every starting pitcher in our data shows a noticeable pattern of switching between hot and cold states.5 Some pitchers’ streakiness manifests in a pronounced downside when they’re cold, like what happens to Texas Rangers lefty Cole Hamels. When he’s on, Hamels’ fastball is just a tick faster than average for him. But when he’s off, he loses about two and a half miles per hour compared to his average fastball. The impact is massive: That almost 4 mph difference in heat translates to a 1.03-run difference in projected runs allowed per nine. If you apply those numbers to Hamels’ 2016 season, he had ace-level stats when hot (3.41 RA/9, 18th among 73 qualified starters), and mediocre ones when off (4.44 RA/9, 44th).

What’s more, the differences in Hamels’ performance seem to be steady from year to year. Running the same analysis on his 2014 and 2015 seasons shows that Hamels always fluctuates between about 2.5 mph down when cold and 1 mph up when hot. He’s not alone: Players who appeared in all three seasons we studied tended to show the same hot and cold effects from year to year, suggesting that we picked up on some of each pitcher’s true characteristics rather than just noise.

RoyalsRetro (AG#1F) Posted: August 11, 2017 at 02:03 PM | 20 comment(s) Login to Bookmark
  Tags: fivethirtyeight, hot hand

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. Pat Rapper's Delight (as quoted on MLB Network) Posted: August 11, 2017 at 02:15 PM (#5511345)
Glad to see the data confirm that pitchers pitch better when they pitch better.
   2. TDF, FCL Posted: August 11, 2017 at 02:26 PM (#5511358)
Glad to see the data confirm that pitchers pitch better when they pitch better.
Ya know, I've been less than enthused about 538's baseball analysis.
   3. Walt Davis Posted: August 11, 2017 at 06:40 PM (#5511573)
Haven't read it but unless they show that "hot" in today's start is a reasonable predictor of "hot" in the next one, I wouldn't call this "hot hand." That said, if 1st inning (or bullpen) fastball velocity is indicative of what Hamels is going to do today, that should be quite useful to teams although in today's game they have limited options to deal with it.
   4. Bug Selig Posted: August 11, 2017 at 06:44 PM (#5511576)
So "hot hand" means "pretty much healthy". Thanks.
   5. cardsfanboy Posted: August 11, 2017 at 07:38 PM (#5511596)
I'm loving the early snark on this thread already. I read the excerpt and was wondering how a guy who is pitching good today(by measurable amount) is the hot hand.

People think of hot hand as a predictor before the result. Sure in basketball maybe the hot hand is the guy having a good 5 minutes of play, but in baseball it's almost always been referred to a guy on a good streak(as in multiple consecutive games of playing above his established ability)
   6. BDC Posted: August 11, 2017 at 07:40 PM (#5511599)
if 1st inning (or bullpen) fastball velocity is indicative of what Hamels is going to do today, that should be quite useful to teams

Yes, though I'd imagine coaches and catchers have known about this for a while. Pitchers talk about it, and old-timey memoirs talk about not having it on a given day and needing to improvise, etc.

We found that the typical pitcher goes through 57 streaks in a season, jumping between hot and cold every 24 pitches


That's not very many pitches.
   7. cardsfanboy Posted: August 11, 2017 at 07:54 PM (#5511605)
That's not very many pitches.


True, but it does show that a pitchers fortune can change and it can be quantified(assuming this thing survives peer review) How many times have you seen a guy who looked like he didn't have it in the first inning, allowing a run or two and getting lucky to get out of that inning with only a couple of runs, and then you look up in the 6th inning and he's allowed one base runner after that, and the opposing team hasn't even had a hard hit ball.

the argument that these streaks exist, and that they can change on a dime, is definitely something that it would be nice to be able to quantify and predict. (or at least see it happening---of course great pitching coaches and managers have instintively known this and could perform probably on par with a decent computer program at these predictions)
   8. PASTE, Now with Extra Pitch and Extra Stamina Posted: August 11, 2017 at 08:01 PM (#5511608)


True, but it does show that a pitchers fortune can change and it can be quantified(assuming this thing survives peer review) How many times have you seen a guy who looked like he didn't have it in the first inning, allowing a run or two and getting lucky to get out of that inning with only a couple of runs, and then you look up in the 6th inning and he's allowed one base runner after that, and the opposing team hasn't even had a hard hit ball.


At a guess, about one-fourth as often as you see a guy who looks like he doesn't have it in the first inning, allowing a run or two and getting lucky to get out of that inning with only a couple of runs, and then you look up in the 3rd inning and he's been KO'd with 6 or 8 runs allowed.

It's undeniable that some days a pitcher has nothing. The ##### is that, thus far, no one has figured out any way to know in advance that the pitcher is going to have nothing tonight.

On the other hand... the scenario you described DOES happen sometimes, and my feeling has always been that a few hitters having really good at-bats bunched together just happens sometimes even when the pitcher's stuff is fine. And of course, that's most prone to happen in the first inning, the only inning in which a manager gets to pick which guys hit.

Sometimes a three-run first inning happens because the pitcher sucks tonight. Sometimes it happens because four of the first six hitters happened to have really good at-bats. It's an important part of the pitching coach's job, and the manager's job, to be able to tell the difference, but I don't know how good major league coaches are at that skill, or how wide the spread on that skill is among the 30 major league clubs.
   9. cardsfanboy Posted: August 11, 2017 at 08:57 PM (#5511628)
Don't disagree with anything you said in post 8. But there is still data that is able to be gathered, and this article suggest that at least there is the possibility of a short term prediction in the mess of data. Most important about the entire article though is that it's arguing that early performance within a game can be projected for later performance.... Which of course seems obvious to everyone, but having quantifiable data means that it's going to be potentially possible to slice the data-live-to determine what is illusion vs what is real.
   10. PreservedFish Posted: August 11, 2017 at 09:02 PM (#5511634)
Sometimes a three-run first inning happens because the pitcher sucks tonight. Sometimes it happens because four of the first six hitters happened to have really good at-bats. It's an important part of the pitching coach's job, and the manager's job, to be able to tell the difference


I don't know if the study is even remotely valid, but this is contradicted by the "every 24 pitches" idea reported above. A guy with authentically crappy stuff in the first inning may well recover his good stuff in the second inning.
   11. Pat Rapper's Delight (as quoted on MLB Network) Posted: August 12, 2017 at 10:48 AM (#5511830)
Sometimes a three-run first inning happens because the pitcher sucks tonight. Sometimes it happens because four of the first six hitters happened to have really good at-bats. It's an important part of the pitching coach's job, and the manager's job, to be able to tell the difference, but I don't know how good major league coaches are at that skill, or how wide the spread on that skill is among the 30 major league clubs.

I don't know how important that skill really would be these days because if the starter doesn't have it, his "role" is now to throw around 100 pitches of whatever quality they are so he can "save the (8-man) bullpen," even if it means giving up 8 runs on 13 hits in 3+ innings and letting the game get completely out of hand since no one in the bullpen has the role of "mop-up long relief."
   12. cmd600 Posted: August 12, 2017 at 12:07 PM (#5511850)
I don't know how important that skill really would be these days because if the starter doesn't have it, his "role" is now to throw around 100 pitches of whatever quality they are so he can "save the (8-man) bullpen," even if it means giving up 8 runs on 13 hits in 3+ innings and letting the game get completely out of hand since no one in the bullpen has the role of "mop-up long relief.


Which is the right way to go. Where exactly is the line between a game getting out of hand and you need a reliever to give yourself a chance and the game already being out of hand and might as well let the starter eat innings as he isnt coming back for another five days anyway? And, of course, the guy whose job would be mop-up long reliever is not likely to be very good at preventing runs anyway. If he was, he would be in the rotation or pitching late innings. So when the choice is between near-replacement level performance in blowouts or another short-stint hard thrower, its clear which of the two can actually make a difference between winning and losing games.
   13. Dr. Vaux Posted: August 12, 2017 at 12:40 PM (#5511860)
It's clear that a significant factor in the most recent round of increased offense is managers letting starters pitch more when they don't have it and after they've ceased to have it, because of the "must throw 100 pitches" and "save the bullpen" mentality. Inning splits show the 4th, 5th, and 6th with more scoring than any others but the 1st. Concurrently, they pitch less when they do have it and might not become ineffective in 7th or 8th innings that they dno't pitch in, though I think that's much less of a problem, since any reliever will usually be more effective than any starter in a 7th or 8th inning.
   14. cardsfanboy Posted: August 12, 2017 at 02:22 PM (#5511890)
Since we are talking about pitchers not having it, Wainwright last night felt tingling in his arm during warmups, and could tell immediately he wasn't able to throw hard for a full game, so he intentionally started to drop the speed on his changeup/curve to maintain the mph difference between that and his under 90mph fastball. The first inning he got lucky with a bases loaded double play to only allow one run, over the rest of the game (he pitched 5 innings total) he allowed no more runs, but he did experience troubles but not to the same degree that he had in the first inning. He was still feeling out how best to throw the ball at a reduced speed, he had multiple pitches that didn't register on the radar gun(supposedly, I wasn't really paying attention to whether it was true or not during the game, but that is what the stltoday article said)
   15. Walt Davis Posted: August 12, 2017 at 08:37 PM (#5512069)
#11 hits on what I was hinting at. Hamels is "hot" at 94 MPH (say) and "cold" at 90. But if he's throwing 90 in the pen, you're not going to go to an emergency starter or declare bullpen day. If he's sitting at 90

But re-reading the intro, they're also talking about a difference of just 1 run per 9 innings. That's big for sure but that means Hamels will give up an extra 2/3 of a run over his 6 innings today. And that's assuming the effect is the same on a very good pitcher as it is on a 6th starter. Moreover Hamels 3.30 ERA is a mix of hot/cold so cold is, what, a 4 ERA. A 4 ERA is still no worse than the Ranger's #2 starter (now that Darvish is gone). Hamels starts no matter what. And at the other end of the pitching universe, presumably the only reason AJ Griffin is warming up to start in the first place is because you have no other option.

But that "typical" pitcher stat -- 57 streaks, 24 pitches -- makes this pretty much meaningless. First "typical" is doing a lot of work there. A "typical" starter probably is only healthy enough to make 25 starts (92 pitchers made 25+ starts last year), so we seem to be talking about more than 2 hot/cold streaks per start (presumably there are average streaks too). 24 pitches is 1.5 innings when you're "hot" and probably just one inning when you're "cold."

So this doesn't seem to be telling us that if a guy is dominant in his first 2 innings, you know he's "hot" today -- it's telling us that dominance/suckitude over the first 1-2 innings tells us squat about how the next 1-2 innings are going to go. That further suggests that Hamels throwing 90 or 94 in the pen doesn't tell us anything either.

Dominant start -- 3-4 hot streaks, up to 1 average/cold streak
Quality start -- 2 hot streaks, 1-2 average, 0-1 cold
Poor start -- 1 hot streak, 1-2 average, 1-2 cold
Crap start -- 2-3 cold, 1 average, 0-1 hot (possibly pulled after 72-80 pitches)

At the end, this looks like little more than the statement that "clutch hits exist" or "batted balls with high exit velocity are more likely to be hits."

I will speculate that Arrieta leads the league with an average of 5 streaks per start.
   16. cardsfanboy Posted: August 12, 2017 at 09:09 PM (#5512100)
Let's be honest, this study doesn't really say much, other than it's potentially possible that in the future we'll be able to work with this type of data to make predictions over the course of a game while it's in action, even if it's just short term predictions.

The pitch framing studies were met with similar issues, in that they haven't really passed any type of "peer review." but the fact is that the fact that they are able to say "anything" is a start at improving the discussion and analytics.
   17. Sunday silence Posted: August 13, 2017 at 01:29 PM (#5512338)
the problem with all these sports studies is that you cant get the data to output in a controlled way. You cant get 100 pitches of "hot" Arrietta to compare to 100 pitches of "cold" Arrietta etc. Because obviously in a competitive sports environment there are humanoids/managers who are pulling the plug on such guys when they get cold. Or they should...Anyhow its not controlled scientifically.

Same for hot pitchers. If they always left them out there for 100 pitches OK but obviously sometimes they are ridden too far as well.

The most recent study having to do with "hotness" that I recall that was well thought of was of basetball players (It was mentioned somewhere in our site). CLaiming the hot hand didnt exist. However, again, if Michael Jordan is hotthey are going to double team him. and if someone is cold presumably he wont getthe ball. So I dont how they controlled for that or if they could.

The 24 pitch thing is also kind of bizarre. You usually dont want to leave a pitcher out in one inning for more than 25 pitches or so. So I wonder if that sort of thing is also playing into it. That you have a human factor that is bound to start pulling pitcher at about 24 pitches anyhow, so that may be throwing the data as well.
   18. Greg Pope Posted: August 13, 2017 at 02:33 PM (#5512361)
I don't know how important that skill really would be these days because if the starter doesn't have it, his "role" is now to throw around 100 pitches of whatever quality they are so he can "save the (8-man) bullpen," even if it means giving up 8 runs on 13 hits in 3+ innings and letting the game get completely out of hand since no one in the bullpen has the role of "mop-up long relief."

Which really doesn't make sense. Let your valuable SP take a pounding in order to save the arms of the last 2 guys in your bullpen. Why would you waste the arm of one of your valuable assets to save the arms of guys who are probably literally replacement level?
   19. Sunday silence Posted: August 13, 2017 at 03:14 PM (#5512376)
I agree with Gregg, with the advent of Win Expectation that sort of statement he quotes out to be backed up with some evidence.
   20. cmd600 Posted: August 13, 2017 at 05:42 PM (#5512456)
Which really doesn't make sense. Let your valuable SP take a pounding in order to save the arms of the last 2 guys in your bullpen. Why would you waste the arm of one of your valuable assets to save the arms of guys who are probably literally replacement level?


Its not saving the last two arms. Those guys still have to finish the game after the starter cant get out of the fifth in under 100 pitches. You're saving your top three or four arms. And that starter isn't coming back sooner or pitching all that much better the next time out whether he goes 60 or 100.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

News

All News | Prime News

Old-School Newsstand


BBTF Partner

Support BBTF

donate

Thanks to
Mike Emeigh
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogOT: New Season August 2017 Soccer Thread
(720 - 7:09pm, Sep 21)
Last: Baldrick

NewsblogOTP 18 September 2017: Ex-Baseball Star Darryl Strawberry Criticizes Jemele Hill, Praises POTUS: Trump is ‘A Great Man’
(998 - 7:06pm, Sep 21)
Last: Fancy Crazy Town Banana Pants Handle

NewsblogMLB, Rob Manfred must expand protective netting | SI.com
(53 - 6:58pm, Sep 21)
Last: Ulysses S. Fairsmith

NewsblogThis is baseball’s home run breaking point | New York Post
(36 - 6:55pm, Sep 21)
Last: Rockwell

Newsblog‘Friends,’ the Sitcom That’s Still a Hit in Major League Baseball
(388 - 6:21pm, Sep 21)
Last: Jay Z

NewsblogPat Neshek rips Zack Greinke for not signing autograph | SI.com
(36 - 6:09pm, Sep 21)
Last: Baldrick

NewsblogOT-NBA off season thread
(2428 - 5:56pm, Sep 21)
Last: PASTE, Now with Extra Pitch and Extra Stamina

NewsblogRaul Mondesi, former Yankees outfielder, sentenced to eight years in prison for corruption while mayor of San Cristobal
(35 - 5:55pm, Sep 21)
Last: bfan

NewsblogDayton Moore on why Royals did not sell | MLB.com
(17 - 5:53pm, Sep 21)
Last: Walt Davis

Newsblog‘Highly unlikely’ Mets risk Harvey burning them — with another team | New York Post
(12 - 5:35pm, Sep 21)
Last: Walt Davis

NewsblogInside Baseball AL Notes | Young Yankees Re-Writing Record Books
(8 - 5:18pm, Sep 21)
Last: Walt Davis

NewsblogNever tell OMNICHATTER the odds! for September 21, 2017
(15 - 4:48pm, Sep 21)
Last: Joyful Calculus Instructor

Gonfalon Cubs15 To Go
(72 - 3:57pm, Sep 21)
Last: Moses Taylor, aka Hambone Fakenameington

Sox TherapyStep One Complete
(11 - 3:53pm, Sep 21)
Last: villageidiom

NewsblogPrimer Dugout (and link of the day) 9-21-2017
(20 - 3:30pm, Sep 21)
Last: Wahoo Sam

Page rendered in 0.4850 seconds
47 querie(s) executed