"Stats" or "Baseball Savvy",

[Doctor Gast]

In the beginning of this off season, I thought I'd start a debate between "Stats" and "Baseball Savvy", but I've been too busy. Now I'm going to give it a shot.

I remember when 2 baseball movies that came out about the same time, "Moneyball" and "Trouble with the Curve". Both were great movies but they had  contrary themes. "Moneyball" portrayed analytics as everything and scouting evaluations and baseball savvy as worthless, and contrary "Trouble with the Curve" portrayed scouting evaluations and baseball savvy as everything and analytics as worthless.

In real life both are important. Analytics has revolutionized Baseball, any team that ignore analytics will be left behind. At the same time teams that only look at stats only and not have proper evaluations and baseball savvy do not know how to properly utilize the stats that are available. Some cherry picks certain stats to present a certain player over another also IMO some stats are developed to favor certain players over another. Also general stats don't indicate factors and conditions that were present. Factors and conditions need to be considered to adequately evaluate players and prospects. For example, the short 2020 season, I don't really  put a lot of stock in (whether a player had great production or poor) because some had difficulty adjusting to it and other did well because it was a short season. The real evaluation should be on regular full seasons. Other considerations are ""live/ dead ball", stadiums and teams when looking for trades.

Again both are needed but which one do you think is more important?

 

[In My La Z boy]

Stats are more important for the front office. Baseball savvy is more important for the field manager. 

[cHawk]

For me analytics. Deemphasizing analytics is thinking backwards in the modern era. 

[Unwinder]

Baseball savvy is just analytics we haven't figured out how to measure yet.

[Doctor Gast]

1 minute ago, Unwinder said:

Baseball savvy is just analytics we haven't figured out how to measure yet.

I like that

[Nine of twelve]

Analytics is simply baseball savvy quantified.

[tony&rodney]

4 hours ago, Unwinder said:

Baseball savvy is just analytics we haven't figured out how to measure yet.

This is absolutely true. The best baseball minds have always intuitively used math to make decisions. It wasn't necessarily written down very much, but there were numerous discussions that could be labeled savvy mixes with analytics. 

I'm rereading Page Smith's "Our Nation Comes Of Age"(pub. 1981). It is pretty funny to see the constant parallels between the U.S. from today against two hundred years ago. The book is straight history (1,200 pages) and does not make the connections but the author writes in a fashion that you are meant to see the futility of thinking society (or baseball for that matter) changes dramatically. It is a constant ruse to think the minds operating today have better and more useful information. Sorry for too many words - I agree with Unwinder (great name).

[Richie the Rally Goat]

Channeling my inner Chief here:

 

 

yes

[Sousy]

I guess I view the use of 'analytics' and 'baseball saavy' in a different way.

Analytics (statistics) is the definitive method for describing things as they are.  A baseball player will have a particular game history, or a particular set of data that can be used to describe how they play baseball in the past and right now. 

I think of good baseball saavy as a way of determining what a player can be given the player's past and present.  Maybe a certain player has a lower wOBA than another, but a good scout should be able to look at the two players and evaluate what a potential upside of a particular player might be.  (Would a change in technique change things entirely for the player? Does the player have a personality and ability to adapt?)

To put this another way, I view analytics for player evaluation and player development as related - but that latter skill requires a bit more intuition ("saavy") on how well a player can adapt and/or be coached into using analytics to improve their game.

[chpettit19]

The interesting third part that I don't know is always lumped into 1 side or the other is the technology and data we get on the actual physical movements of players, the ball, etc. Spin rates, exit velo, sprint speed. I see those as a marrying of the 2 sides to some degree. They're quantifying what the scouts are trying to do in deciding if a pitch has good break or a guy hits the ball hard or hits the ball with loft or is fast, but also putting context to the analytics of the stats that a player produces.

Maybe people see that as analytics, and if that's the case I'd say analytics are much more important. I see it as a 3rd side to help tie the other 2 together. How players get to their stats is important. Hitting .320 in AAA looks nice, but doesn't mean you're a MLer depending how you go there. Hitting .320 with top exit velos, launch angles, and contact rates means you have a much better chance of surviving the next step. The scouts also provide context in how the player is reaching their numbers. The data companies are gathering now with defensive positioning on every ball in play for every level of the minors and college adds even more context.

[wsnydes]

2 hours ago, Sousy said:

To put this another way, I view analytics for player evaluation and player development as related - but that latter skill requires a bit more intuition ("saavy") on how well a player can adapt and/or be coached into using analytics to improve their game.

I like how you put this.  There are intangibles like situational awareness on the fly that has the element of instinct that can't always be taught can't be quantified.  These are oftentimes things that make lesser skilled players hang around longer and/or increase their value.

[jorgenswest]

Analytics are critical and I am confident that Falvey has built a solid analytics staff from almost nothing before him,

One of the most important things about understanding data is to know when to ignore it. I am guessing the Twins do that pretty well. The media on the other hand needs something to say or write and are constantly making assertions using data that are truly not supported. 

 

 

[kenbuddha]

I have a hard time believing that anyone still thinks analytics is controversial. After all, baseball has always had more statistics than any sport. As a kid if you were into numbers baseball is what you gravitated to. Analytics is just diving in deeper.

With Moneyball it was a realization that as an organization they couldn't just keep doing the same thing they've always done. They had to figure out a different way since the teams with money could hire the best scouts, pay the highest bonuses, sign the expensive free agents and retain their top talent. They had to find an edge other than "baseball savvy" because that's what they had always done and they couldn't compete any more.

In today's game, analytics is finally the norm but there's lots of unexplored space in analytics and the today's teams continuing to explore are the equivalent of Moneyball's A's. Finding an edge.

Baseball savvy of course still has a place, just not anywhere close to the start-middle-and-end of each organizational decision that it once was.

And in my opinion, "Trouble With The Curve" was the equivalent of an old man yelling at kids to get off his lawn. Just a big giant dated cliche. And I'm an old man.

[Squirrel]

34 minutes ago, kenbuddha said:

And in my opinion, "Trouble With The Curve" was the equivalent of an old man yelling at kids to get off his lawn. Just a big giant dated cliche. And I'm an old man.

Lol ... that was hilariously so accurate a description

[ashbury]

6 hours ago, Sousy said:

Analytics (statistics) is the definitive method for describing things as they are.

That is unnecessarily limiting.

A description from the business world I like is that the first level is Descriptive Analytics, which is what you are saying.  Once you have that as a solid foundation, you can move to Predictive Analytics, which uses additional tools and which as the name implies has a forward-looking perspective.  Once you have that, you can practice Prescriptive Analytics, which involves other techniques to figure out what is the best course of action among many options, given the information provided by the other forms of analytics. 

It's certainly true that you can't try to optimize anything until you first have a handle on things as they are. But you need insight as to where things are probably going if you don't take action yourself, and analytic tools are available for that; and then to decide on what to do to change the future requires some more methods. 

None of this means you "let the computer decide" - the art is in using the tools meaningfully.  Call it baseball savvy, if you like, when that's the industry doing it.

To use the word statistics as a synonym for Analytics, or to equate them, misunderstands what businesses for decades, and baseball teams at long last, mean by the word.  A baseball team in the 21st century doing only Descriptive Analytics would be left behind in the dust; indeed I think it largely is why the forays into Analytics under Terry Ryan didn't amount to much.

 

[Trov]

I had a very similar argument years ago, when at all-star break Twins were around .500, but run differential was not in line with record.  The all antilytic people said the Twins will have a bad second half because run differential says so.  I argued, run differential is not a predictive stat, but just a number.  People would push back and say history shows the record cannot be sustained with that run differential.  I replied again, you are looking at what happened, not what will happen.  The number only says it is unlikely if the differential stays the same the record will remain well, but what is to say the run differential will not change, the difference only says it is likely they two will not stay in line as they were, but nothing says the record would drop.  What happened was the run differential got better. 

Where I am going is, stats are only numbers of the past, they do not predict the future.  They can be used to help go hand in hand with ones eyes to determine what is most likely to happen, but not set in stone.  In Moneyball they used example of Kevin Youkillis who had an odd stance and scouts said he will need to change that to do well at MLB, but the antilytic person pointed out he was doing just fine how he was doing. 

Just looking at numbers will never tell the full story.  I have commented about Sabato and his numbers in Fort Myers.  I wanted to know what led to the numbers.  He had huge walk rate, and K rate, with terrible BABIP, low sluggling and average.  The question I had, was he getting pitches to hit and missing or was he being pitched around?  Was he hitting wall scrapers or popping up when he did put ball in play? The numbers raised an eye brow, to look deeper.  

Many like to use analytics for Keplar in saying his numbers should have gone up over his career, because hard hit rate was not in line with BABIP.  However, year after year, numbers were the same.  I said at some point you just have to accept he is the exception. 

There is no single number that can predict future output.  You cannot decided who to have on a roster just based on numbers.  Take Trevor Plouffe as an example.  If you looked at his HR you would say he had a couple of good years, but when you look at fact most were solo, you would find out, he would hit so many HR when huge leads or behind huge and lower talent pitchers would serve up fastballs, but when guys were on base and pitcher bared down Plouffe would fail more often than not. 

When past front office and mangers hated new analytics, I would get so mad, because they were given data to find edges, but they would dismiss it and go with old school ways, never looking at reverse splits in some pitchers or hitters.  I remember Joe Madden with Rays would have his switch hitters bat same side for a few pitchers because they were reverse split pitchers due to type of pitches they used.  That is where analyitcs can be helpful.  

All one way or the other is bad, you need to merge the two in my opinion.  It is important to remember not all the numbers take contex into account.  They try to narrow it down, but looking at inning, score, runners on base, counts, pitcher, all those things are important to get the contex of where the numbers came from.  

[lukeduke1980]

Savvy is hearing hitters talk about setting up pitchers to get them to throw a certain pitch later in the game then crushing it.  Analytics try to reduce those scenarios 

[Doctor Gast]

I'd like to thank everyone who has participated in this discussion. It has been both interesting and enlightening. I liked what was mentioned about the degree of importance depending on the level, for example analytics aren't as important to the owners as it is to the assistant GM or coaches.

I like what was said about Analytics that it be can split up in 2 area  #1 as a baseball science for developing players #2 stats that can be used w/ baseball savvy which both are important to evaluate players. 

I also liked what was mentioned "Baseball savvy is just analytics we haven't figured out how to measure yet." Although thinking about it I don't know if much of baseball savvy could ever be measured or quantified. 

It'd be difficult to summarize all that has been discussed but I'd like to share one conclusion. Although IMO owners aren't necessarily obligated to know analytics, I do believe they should have some baseball savvy. I don't believe Carl Polhad was lucky when he selected Andy Macphail, I believe he had enough baseball savvy to do so. While looking at the Rockie's, I scratch my head on how they completely mismanaged a team with a great core of players. By getting players that might've looked good on paper but couldn't produce on the field at the same time way overpaying them and the way the they are letting go there main core. Right now the Rockies are hopeless.

[mikelink45]

I am late to the discussion, but I want to add another perspective.  Baseball is a game.  All games have analytics - even Monopoly and scrabble, but a game that consists of people, objects that are not responsive to the toss of the dice needs psychology as much as statistics.

To say the old game did not have analytics would be wrong, they just did not have as many.

But the problem that is inherent in all these discussions is the fact that as analytics continue to get more and more complex the game gets longer and more boring and the reaction of the general population is a lower level of respect and interest.

Baseball has to figure out how to put excitement back in the game.  If analytics say a SB is not a good play, I disagree. Give me Jackie Robinson stealing home, or Rickey Henderson stealing everything.  Put the bunt back in the game - challenge the shift don't outlaw it. Speed up the game - all those Ks and Walks take more pitches and make the game longer.  

If all the teams are using the same analysis then they are equal.  Give me an aggressive team like the Billy Martin Twins, the Fox and Aparicio GO - GO White Sox, or the Whitey Herzog Cardinals.  Shake things up.  I want a game not a mathematical formula.  I do not care how far or fast a HR goes, I want a runner on base so we get more runs.   

Launch angle is not my concern - did the player get on base?  Yes the team needs to evaluate and use all these stats, but they also need to think about the fan and enjoyment.  I do not enjoy an opener?  Are they really better than the starter?  I do not like BP games or starters going 4 innings or less.  They are taking the enjoyment out of the game for a person who has been listening and caring for 70 years (I had to have a few years of childhood before finding the game) and that should be a very important factor because when the baby boomers are gone they will need the soccer and lacrosse generations.  

The statistics I am most interested in are the ones that measure fan interest and commitment to the game.  I have found myself yawning through too many long games and it has taken away my attendance dates.  

[dxpavelka]

BOTH.  But I would point out that the last time the Oakland Athletics won a championship Billy Beane batted .241 for them.

[USAFChief]

On 1/26/2022 at 5:45 PM, kenbuddha said:

And in my opinion, "Trouble With The Curve" was the equivalent of an old man yelling at kids to get off his lawn. 

I know, right??!! GREAT movie!

[Rigby]

On 1/30/2022 at 10:54 AM, USAFChief said:

I know, right??!! GREAT movie!

Yeah, it was pretty Rowdy.....

[Mike Sixel]

On 1/26/2022 at 5:19 PM, Squirrel said:

Lol ... that was hilariously so accurate a description

Truth. That movie was terrible. Moneyball was meh at best.