Photo

Notes from Astroball


  • Please log in to reply
26 replies to this topic

#1 BSLMikeRandall

BSLMikeRandall

    Sr. Ravens Analyst

  • Members
  • PipPipPipPipPip
  • 19,024 posts

Posted 29 January 2019 - 10:51 AM

Sorry if this is TLDR. I finished the book Astroball a few weeks back and compiled some notes that I found interesting. Some things related to the O's, some things we can maybe expect from the new regime led by Mike Elias and Sig Mejdal. Some of this might not be news, but enjoy it anyway.

 

------

 

Sig Mejdal (pronounced my-dell) has two masters degrees, was a NASA rocket scientist, but applied what he studied in psychology of astronauts to his dream job, in baseball. He got his start after reading Moneyball in 2003, and sending every team an analytics forward resume, pitching that he could make teams twice as good at drafting by coming up with a way to translate amateur players stats to professional projections. He also went to the winter meetings with brochures, trying to sell himself to anyone whose face he recognized from Baseball America’s photos of front  office employees. Knowing most of his brochures ended up in the garbage, one person called him up. Jeff Lunhow, who was just promoted to scouting director of the St. Louis Cardinals in 2005. He sought Mejdal to become a scout for him. Then Mike Elias, 23, the following year.

 

Mejdal found that the best college player in 2005 was Jed Lowrie at Stanford. Mejdal feared he’d be laughed out of the room being the new guy and suggesting a short, skinny, second baseman would have a better career than a big, imposing, son of a major leaguer, Lowrie’s teammate, John Mayberry Jr.  The Cardinals also liked SS Tyler Greene from GA Tech who “looked like a major leaguer.” The Cardinals ended up selecting Greene at pick number 30, while Theo Epstien with the Red Sox scooped up Lowrie at pick number 45.

 

Mejdal was right. Lowrie still going strong, earning his first ASG selection in 2018. Greene and Mayberry Jr. never amounted to much.

 

Lunhow and Mejdal in the wake of Moneyball felt the story showed the scouts to be antagonists. They didn’t want that to be the case. They concluded the human element can’t be eliminated. So in 2006, Mejdal created a metric called STOUTS. Half his performance based algorithms and stats, half scouts analysis. From that, they had some really successful, drafts with the Cardinals, including finding some talent deep into the double digit rounds.

 

Lunhow eradicated previous biases after so many teams passed on Mike Trout. Mainly because only one of ten New Jersey high schoolers drafted in the first round ever became anything remotely noteworthy, and that was in 1965. Players size, where they are from, what level they play on, none of that matters to them anymore. 

 

In 2010, Elias was sent to Puerto Rico to scout Carlos Correa, a 15-year-old playing ball with men much older than him. Elias noted that he saw Correa hit a line drive off the opposite field (right field) foul pole, and the ball never got more than 10 feet off the ground. He'd never seen that speed off the bat before. 

 

When Lunhow became GM of the Astros in 2012, he brought Mike Elias, then 29, to be his special assistant for scouting. Sig Mejdal came along and was baseballs first Director of Decision Sciences.

 

They believe a lot in high breaking ball spin rates. They picked up Colin McHugh off waivers despite awful traditional stats. 8.94 ERA over 15 outings, and turned him around. Same goes for Charlie Morton who closed out the final innings of the World Series. They simply had those guys rely on their better pitches more often.

 

Turned around George Springer. Extremely high K/rate in single-A of once every 3.7 ABs, from swinging too hard. Adam Dunn, striking out the most in MLB in four of his 13 seasons, did so once in every 5.5 ABs at single-A. Springer didn’t swing at bad pitches though, drawing a walk every two games. Just tried too hard on good ones. I'm not sure this is Chris Davis problem because he swings at alot of bad pitches. But it's good to see that it can be fixed with the right coaching. 

 

Turned around Dallas Keuchel, taught him to pitch backwards. His FB topping out at 89mph would appear faster after starting hitters with breaking stuff. They also introduced a lot of video work to the Astros staff, combined with where to attack every hitter, and where to stay away from. This helped Keuchel go from a name that away teams PA announcers always butchered, to Cy Young winner.

 

In their third year in a row picking number one overall, they narrowed down their options to Brady Aiken, Tyler Kolek, Carlos Rodon, Aaron Nola, Nick Gordon, and Alex Jackson. They took Aiken based on all the information they had compiled over the past three years, based on all  the data, all the human elements, and disregarded the fact that he was a high schooler. Turns out upon his physical, Aiken’s UCL was congenitally small, making Tommy John more likely. The Astros lowered their offer to $3.1M. Aiken declined, and eventually was the first number one overall pick in the last 31 years not to sign.  But the Astros knew the rules that they would receive compensation as long as they offered 40% of the slot value. The lowball offer was exactly the 40% of the $7.9M slot value for the 1:1 pick. Compensation was the number two pick in the next years draft.

 

Aiken tore his UCL after 13 pitches into his first competitive action for IMG Academy in Florida.

 

With the compensation pick the next year, the Astros selected Alex Bregman number two overall. Even if he was undersized by some major league execs. "Allegedly six feet tall? We don't give a shit."

 

To recap, since the O’s have the 1:1 pick this year, Lunhow, Elias, Mejdal’s three 1:1s were Carlos Correa, Mark Appel, and Brady Aiken. Appel never got out of single-A, went to the Phillies, retired after three years of never getting to the majors. Aiken hasn't gotten out of Cleveland's A-ball in 178 IP with an ERA well over 5.

 

They let J.D. Martinez get away. A hitting coach with no answers for a player OPSing sub-.700 suggested he “change his swing”. To which Martinez said, “so they can send me to Triple-A after I go 0-for-20 while changing my swing?” He found the stroke on his own when he went to play winter ball in Venezuela. Returning to camp next spring, he begged them to give him a real chance, that he had changed things for the better, he was ready to prove it. The Astros didn’t give him that chance. He wanted 40+ at bats, and only got 15, and then DFA’d. 

 

Lunhow had fired manager Bo Porter after the 2014 season, and knew the next manager had to have worked in many facets of an organization so they know how they all interact. Had to understand the power of analytics, had to be data driven. They interviewed 10 candidates and asked each one if it would be ok if the front office and analytics department sent down the lineup card every day. Many said “yes” which was the wrong answer. The front office wants to give the manager all the data possible, but leave the human element to them. Only they know if someone was battling a cold, didn’t sleep well, had sore legs, didn’t trust a pitch that day, anything. They hired A.J. Hinch. 

 

The Orioles hiring of Brandon Hyde doesn't check all the boxes Hinch did for Lunhow. But maybe Hyde checks enough of them and really buys into what Elias and Mejdal want to do here.

 

Mike Elias became Jeff Lunhow’s assistant when former assistant David Stearns took over the GM role for the Brewers at age 30, in 2015. The Brewers are a team on the rise. We can only hope Elias has the Orioles going in the same direction soon.

 

Lunhow and Mejdal believed in chemistry. Just because you can’t quantify it, doesn’t mean it doesn’t exist. Someone or something in the locker room is the glue that holds it all together. For the Cubs it was a guy like David Ross. The Astros went out and brought back Carlos Beltran for the same reasons.

 

I would not be surprised one bit if we see Adam Jones back in an Oriole uniform before he hangs it up.

 

Beltran knew it was tough for young players to approach veterans for help. Remembered how scared he was to ask Barry Bonds for advice when the two shared a locker room at the All-Star Game. Bonds obliged the young Beltran and took him to the cage and taught him a few things. Bonds would turn the pitching machine all the way up (95+mph). Then starting in the batters box, would move closer and closer to the machine until you physically could not catch up to the ball. Seeing 95mph at 45 feet away makes seeing it at 60 feet away look a lot slower.  

 

Beltran was at least seven years older than anyone else on the team when he came back to Houston in 2017.

 

Through a complicated study, it was determined a team could win or lose three games based on “in-groups” in clubhouses. If the young guys and old guys were separated, the Americans and the Latinos had their own groups, pitchers and hitters…the divisions, or “fault lines” in locker rooms could be a six game swing because someone’s mind might not be on the task of winning, but on the task of learning a new language, wondering if the old guys despise a youngster. A once poor English speaking Beltran was taken under the wing of an English speaking teammate when he made it to the big leagues. It put him at ease that someone would cross that “fault line”.

 

Beltran noticed when he got to the Yankees in 2014 that they had three Japanese players, each with their own translator. However, they had five times as many Latino players, many who spoke no English, and they had no translators, so Beltran played that role for them. He lobbied MLB to ensure each team hired a full time Spanish-translator, and two years later they did. But today, walk into almost any MLB clubhouse, and you’ll see a dividing line between the in-groups, the English speakers, and Spanish speakers. Beltran sought to break that barrier.

 

Beltran united the Astros clubhouse by making things fun. Getting rid of those fault lines. Alex Bregman helped Cuban defector Yuli Gueriel, ten years older than him, with English, and vice versa with Spanish. Just like the help Beltran got as a rookie. Brian McCann held an elaborate funeral for Beltran’s glove after he had DH’d for two months straight.  They got replica boxing title belts, and after each win the pitcher of the game and hitter of the game would be awarded the belts. Next win, the belt holders would award it to the next guy.

 

They had fun as a team. The Orioles version of this was the post-game pie in the face, which was banned by the front office. Somewhere along the way it stopped being fun. Perhaps the Angelos brothers will allow fun things besides locker room ping pong and corn hole. 

 

Beltran was also addicted to the video room not just for himself, but watching for his teammates so he could help them out. He was the one who found Yu Darvish tipping his pitches, exposed that to his teammates who ripped the ball off Darvish in his World Series starts.

 

A flaw in Mejdal’s algorithms are that they don’t like deadline trades, where one team is giving up lots of potential future value, for a little bit a present value. The Astros trading for Verlander would have been vetoed by Mejdal if they relied solely on math to make their decisions. But they also felt like having the Tigers eat $16M over the remaining two years of Verlanders deal, meant that getting a pitcher like Verlander at essentially a 2/$40M deal, would be impossible on the open market. Even at age 35, a 4/$100M+ for future HOFer would be more likely.

 

The Orioles pulled the plug at the last minute on trading Zach Britton to the Astros, but no reason why was given. Perhaps Elias won’t be a non-waiver deadline dealer, but they know there are two deadlines. The 2017 Astros were very quiet at the non-waiver deadline while other teams loaded up. They were patient and landed Verlander a month later.

 

Mejdal doesn’t just hide in the “nerd cave”. He spent three months in the summer of 2017 coaching first base for the Valleycats. Houston’s A-ball equivalent to the Ironbirds. He was there to make sure that even at the lowest level of the system, they are incorporating data into their infield positioning and pitch selections. Among other secret tasks.

 

Of the 25 players on the Astros World Champion roster, three were inherited by Lunhow when he took over in 2012. Four were draft picks of his. Nine were players traded for, and nine were free agents they had signed. 

 

Five years from now, will three players in the Orioles organization still be here? 

 

Final words of the book: “As the data landscape flattened—as every organization gained access to the same set of numbers and technologies—the new edge might come from those sources that were virtually impossible to quantify and to incorporate into statistical models. There would, in other words, always be a place for human intelligence alongside the artificial kind, and not just in baseball. There would always be a role for gut feels.”


  • BSLRobShields, BSLSethBondroff, Ricker Says and 11 others like this
@BSLMikeRandall

#2 BSLMikeLowe

BSLMikeLowe

    CFB Analyst

  • Moderators
  • 12,796 posts
  • LocationPortland, Oregon

Posted 29 January 2019 - 10:54 PM

Turns out upon his physical, Aiken’s UCL was congenitally small, making Tommy John more likely. The Astros lowered their offer to $3.1M.

 

Now we know why the Angelos' wanted someone from Houston.


  • DJ MC, BSLMikeRandall and FFH like this

#3 dude

dude

    HOF

  • Members
  • PipPipPipPipPip
  • 6,251 posts
  • LocationColumbus, GA

Posted 29 January 2019 - 11:35 PM

Mike, great stuff.  Thanks for taking the time to do that.

 

I haven't read it other than what's been shared.  My comments would be on it being contextually accurate (not saying it's not at all).

 

------------

 

I have some other comments to make about it, but my overall comment would be it's encouraging. 

 

Glad they got them and I hope we see some tangible results from the things they do as a group. 


  • BSLBobPhelan and BSLMikeRandall like this

#4 BSLMikeRandall

BSLMikeRandall

    Sr. Ravens Analyst

  • Members
  • PipPipPipPipPip
  • 19,024 posts

Posted 29 January 2019 - 11:56 PM

Now we know why the Angelos' wanted someone from Houston.


Could certainly get burned if you don’t cover all the bases, no pun intended.

Wish like football we would get a combine and players would have physicals and interviews there. So you don’t take your first overall pick, only to later find out some imperfection that make it a terrible signing if you make it.

Hard to do with a 50 round draft. Which is also ridiculous. Can probably cap it at 15 rounds and make the rest of the world signable by any team in an undrafted free agent frenzy. Like football. Most guys after round 30 aren’t signing anyway. Most guys after round 20 are novelty anyway and not serious ball players. Buck Showalters daughters best friends boyfriends roommate or something.
  • sandiegosean likes this
@BSLMikeRandall

#5 weird-O

weird-O

    All Star

  • Members
  • PipPipPip
  • 1,025 posts
  • LocationI'm here from downtown, I'm here from Mitch and Murray.

Posted 30 January 2019 - 08:34 AM

Could certainly get burned if you don’t cover all the bases, no pun intended.

Wish like football we would get a combine and players would have physicals and interviews there. So you don’t take your first overall pick, only to later find out some imperfection that make it a terrible signing if you make it.

Hard to do with a 50 round draft. Which is also ridiculous. Can probably cap it at 15 rounds and make the rest of the world signable by any team in an undrafted free agent frenzy. Like football. Most guys after round 30 aren’t signing anyway. Most guys after round 20 are novelty anyway and not serious ball players. Buck Showalters daughters best friends boyfriends roommate or something.

The major difference is that baseball franchises have multiple levels of teams, whereas football teams just have a practice squad. MLB needs all those players. And while it's true that most of them will never sniff the majors, they need them, if only to have enough players to fields teams for the legit talent to get their seasoning. I'm sure the owners much prefer the slotting system to an undrafted FA class.   


  • BSLMikeRandall likes this

Good news! I saw a dog today.


#6 BSLMikeRandall

BSLMikeRandall

    Sr. Ravens Analyst

  • Members
  • PipPipPipPipPip
  • 19,024 posts

Posted 30 January 2019 - 01:18 PM

The major difference is that baseball franchises have multiple levels of teams, whereas football teams just have a practice squad. MLB needs all those players. And while it's true that most of them will never sniff the majors, they need them, if only to have enough players to fields teams for the legit talent to get their seasoning. I'm sure the owners much prefer the slotting system to an undrafted FA class.   

I get that. You might be right about the slotting system. 

 

I just did a quick look at the Yankees drafts. The team where every kiss ass dreams of wearing pinstripes. 40 rounds.

 

2016, pick 16, 22, 26, 29, 31-35, 37-38, and 40, all didn't sign.

2017, picks 21, and 23-40 all didn't sign. Nearly half the draft wasn't kept. 

2018, pick 31, and 35-39 didn't sign. So they kept most.

 

Looks like in about 40 rounds the O's don't sign about nine or so guys. 

 

Still think you could get away with a 20 round draft for MLB and then teams can sign anyone they want to fill in the Gulf Coast League and NY/Penn league rosters as needed. At the time of the NFL merger they had a 17 round draft. That's been cut by more than half.


  • Ricker Says and weird-O like this
@BSLMikeRandall

#7 Ricker Says

Ricker Says

    HOF

  • Members
  • PipPipPipPipPip
  • 43,163 posts
  • LocationGermantown, MD

Posted 30 January 2019 - 01:21 PM

Mike - really appreciate you putting this together. 


  • BSLMikeRandall likes this
@0TheRick0 (AKA The Rick)
"You can't sit on a lead and run a few plays into the line and just kill the clock. You've got to throw the ball over the damn plate and give the other man his chance. That's why baseball is the greatest game of them all." ~ The Earl of Baltimore

#8 BSLMikeRandall

BSLMikeRandall

    Sr. Ravens Analyst

  • Members
  • PipPipPipPipPip
  • 19,024 posts

Posted 30 January 2019 - 02:56 PM

Mike - really appreciate you putting this together.


Thanks man. A lot of good info in the book. Little bit of a look into the personal life of some guys. It’s a good read.
@BSLMikeRandall

#9 BSLBobPhelan

BSLBobPhelan

    Movie & MMA Analyst

  • Moderators
  • 8,453 posts
  • LocationBel Air, MD

Posted 30 January 2019 - 03:54 PM

Thanks man. A lot of good info in the book. Little bit of a look into the personal life of some guys. It’s a good read.


Co-sign. Highly recommended for all O’s fans heading into the season. But if you can’t, the OP is a nice synopsis.
  • BSLMikeRandall likes this

#10 dude

dude

    HOF

  • Members
  • PipPipPipPipPip
  • 6,251 posts
  • LocationColumbus, GA

Posted 02 February 2019 - 09:23 AM

I wanted to wait until others had made their comments on Mike's effort here before commenting on a number of the points.

 

Final words of the book: “As the data landscape flattened—as every organization gained access to the same set of numbers and technologies—the new edge might come from those sources that were virtually impossible to quantify and to incorporate into statistical models. There would, in other words, always be a place for human intelligence alongside the artificial kind, and not just in baseball. There would always be a role for gut feels.”  

 

I fervently agree with this.  That sounds weird, but I don't have the right word(s) to describe how much I agree.

 

I've worked Technology, I've worked Metrics, I've worked Human and I've done the integration between Tech and Operational application (a lot) and you never get away from the consequence of Human. Analytics gives you information - an understanding - you may not have had before and the success is figuring out how to leverage that value meaningfully to get to your endstate....but execution will still always be humans.


  • BSLMikeRandall likes this

#11 dude

dude

    HOF

  • Members
  • PipPipPipPipPip
  • 6,251 posts
  • LocationColumbus, GA

Posted 02 February 2019 - 09:54 AM

Lunhow had fired manager Bo Porter after the 2014 season, and knew the next manager had to have worked in many facets of an organization so they know how they all interact. Had to understand the power of analytics, had to be data driven. They interviewed 10 candidates and asked each one if it would be ok if the front office and analytics department sent down the lineup card every day. Many said “yes” which was the wrong answer. The front office wants to give the manager all the data possible, but leave the human element to them. Only they know if someone was battling a cold, didn’t sleep well, had sore legs, didn’t trust a pitch that day, anything. They hired A.J. Hinch.

 

 

Again, hard to frame how much I agree with this.  This is, and always has been, my point with application.

 

Build the best tools and understanding you can.  There was another article maybe a year or so ago that basically said that with the proliferation of data and analysis and the transience of people across FOs, everyone eventually has similar information.  What will make the bigger difference is figuring out how to get the buy in between the information and the people executing it.  

 

A challenge of data use can be that it assumes the conditions that the analysis was developed under are the same as conditions as you have in front of you.  

 

One of the articles I haven't written yet is titled, "We're playing Poker, not Blackjack".    

When you play Blackjack, there's no emotion, no gut, you just straight play the numbers because the cards aren't humans.

When you play Poker, you better understand the numbers but there's a layer of human that drives success or failure on every single hand. 

Baseball is more like Poker.


  • Ricker Says likes this

#12 dude

dude

    HOF

  • Members
  • PipPipPipPipPip
  • 6,251 posts
  • LocationColumbus, GA

Posted 02 February 2019 - 10:38 AM

Lunhow and Mejdal believed in chemistry. Just because you can’t quantify it, doesn’t mean it doesn’t exist. Someone or something in the locker room is the glue that holds it all together.

 

The acknowledgement of Chemistry is outstanding.  I think it's a real challenge to define it.  They've taken their shot along things like Clubhouse 'in groups' and 'fault lines' and all of that (some definition) is critical to try and create the database you'll use for any analysis.

 

Through a complicated study, it was determined a team could win or lose three games based on “in-groups” in clubhouses.

 

...but I'll push back on this a little.  First, any time you start with "it's complicated" it always feels like you're saying...'don't bother, just trust me' which is a tool that avoids scrutiny...but we'll get past that :)

 

...but my larger issue is that I wouldn't put a number on it.  For me it comes down to an understanding of 'how you win'.  It's not a roll up of WAR numbers (like some want to do with MWAR - Manager WAR) and the notion that you can "have chemistry" and add 3 wins (or lose 3) seems to avoid how teams actually win.

 

The analysis of Talent creates some generic baseline for a team's opportunity (win total).  Analytics application can actually impact the Talent factor because it (endstate) may change the quality of the individual(s) considered for that analysis.  When you get to things like Attitude, Chemistry and most importantly, Leadership (layers against all 3)...then you are moving up or down against some baseline opportunity. 

 

You also have to layer it against the other teams around you.  We know sayings like "it's not the size of the dog in the fight, it's the size of the fight in the dog"....but what happens when the big dog knows how to fight just as well?  Size matters, it's why we have weight classes in Fight Sports and use the term "pound for pound".  Personally, I like to use the NCAA Basketball Tournament (greatest cumulative sporting event every year) as the framework for understanding applications for winning.

 

Attitude matters

Talent matters

Chemistry matters

Leadership matters


  • Ricker Says likes this

#13 BSLChrisStoner

BSLChrisStoner

    Owner

  • Administrators
  • 110,228 posts

Posted 01 April 2019 - 08:14 PM

@BenReiter
BALTIMORE! Thrilled to announce I’m visiting this Friday for a really exciting reason. I’ll be discussing and signing ASTROBALL at the Inner Harbor B&N at 6 PM ... and I’ll be doing so with Sig Mejdal, the Astros’ former data guru and the Orioles’ new Assistant GM. Come meet us!

 

D3FCoRFW0AEBzGv.jpg


  • BSLMikeRandall likes this

#14 Coker

Coker

    HOF

  • Members
  • PipPipPipPipPip
  • 4,510 posts

Posted 01 April 2019 - 09:38 PM

Cool, I will try to make my way over to catch some of this.

Thanks for posting about it, Chris. 


  • sandiegosean likes this

#15 BSLMikeRandall

BSLMikeRandall

    Sr. Ravens Analyst

  • Members
  • PipPipPipPipPip
  • 19,024 posts

Posted 02 April 2019 - 09:07 AM

Fantastic book.
@BSLMikeRandall

#16 ivanbalt

ivanbalt
  • Members
  • 332 posts

Posted 02 April 2019 - 10:38 AM

Fantastic book.

 

Definitely.  I'd like to see an additional chapter about the Orioles "analytics" department that Sig found when he came here.  :lol:


  • BSLMikeRandall likes this

#17 BSLMikeRandall

BSLMikeRandall

    Sr. Ravens Analyst

  • Members
  • PipPipPipPipPip
  • 19,024 posts

Posted 02 April 2019 - 10:50 AM


Definitely. I'd like to see an additional chapter about the Orioles "analytics" department that Sig found when he came here. :lol:


It’ll be a real short chapter.
@BSLMikeRandall

#18 ivanbalt

ivanbalt
  • Members
  • 332 posts

Posted 03 April 2019 - 06:10 AM

It’ll be a real short chapter.

 

"We found an old Apple II computer that was mainly used for playing Oregon Trail."


  • BSLMikeRandall likes this

#19 Nigel Tufnel

Nigel Tufnel

    HOF

  • Members
  • PipPipPipPipPip
  • 4,189 posts

Posted 03 April 2019 - 08:13 AM

"We found an old Apple II computer that was mainly used for playing Oregon Trail."

 

"They died of dysentery."


  • BSLMikeRandall likes this

#20 BSLMikeRandall

BSLMikeRandall

    Sr. Ravens Analyst

  • Members
  • PipPipPipPipPip
  • 19,024 posts

Posted 03 April 2019 - 08:29 AM

"We found an old Apple II computer that was mainly used for playing Oregon Trail."

You mean I could have worked in their analytics department as a 5th grader? Ugh. Missed opportunity there. 


@BSLMikeRandall




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

Partners