Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > SABR - BBTF Chapter > Discussion
SABR - BBTF Chapter
— SABR's First Virtual Chapter

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. Mark Armour Posted: August 08, 2016 at 04:41 PM (#5280749)
Chris Dial gave a talk on Thursday morning on the past and future of defensive statistics. He had only a few hours notice (replacing someone who had to cancel), and he pulled it off well in my opinion. He did not get a big audience because it was not in the schedule.

I believe that a missing piece in SABR (for many years) has been the lack of overviews/primers on statistical analysis. I gave a talk at a local chapter meeting many years ago that attempted to explain the concept of linear weights -- I got sick of hearing talk after talk that insisted that nerds were valuing walks just because they liked that kind of baseball -- not because of any evidence that they had any value. Like Chris's talk, I did not employ "math", other than using words like "correlation" and "best fit" which everyone seemed to understand.

I believe that there are other primer/Primer talks that could be given that might help people to keep up. Maybe even a panel.

   2. MHS Posted: August 08, 2016 at 08:27 PM (#5280881)
Imagine how great that talk would have been if Dial was actually an expert on defensive statistics...
   3. TomH Posted: August 09, 2016 at 09:38 AM (#5281066)
Mark - I agree in principle with what you said; there should be a place at SABR conventions for primers. Perhaps it would be best for convention planners to consider 1 or 2 of these during the presentation sessions, but not call them "research presentations" that must be vetted ahead of time; it's not so much research as it is explanation. A bit of a separate category that a SABR member or team could volunteer to work up.

I think it would be well received. I mean, even if someone read Win Shares cover to cover, wouldn't they likely still look forward to listening to Bill James verbally explain how it works?
   4. villageidiom Posted: August 09, 2016 at 10:25 AM (#5281112)
Not a SABR member, but I did set up the thread, so...

Most primers on sabermetrics can be summarized as follows.

1. Basic stats tell you what someone did.

2. That is meaningless unless compared to some kind of benchmark. Is it more than anyone has ever done? Is it less than the player did last year? How long ago was it that someone did the same thing? How often does it happen?

3. The benchmark to use depends on what you're trying to understand. If "how rare is it?" you want to know how many times it has happened before. But for a novice even that needs context: how many years has MLB operated? How many games have been played? Or maybe you want to know how rare it is today. Then you might want to know how long since the last time it happened, or just tighten up the range of years you're looking at when counting the number of times it has happened. If "what's the impact of it" then you want to know how it affected the outcome - of the PA, of the inning, of the game, of the season.

4. Most advanced stats are all about trying to line up the right benchmarks to answer a specific question. Coming into advanced stats without knowing the questions is like hearing the punchline without the joke. So let's revisit the questions.

I don't get the sense that sabermetrics primers actually go that way, but IMO they should. They should emphasize that benchmarks/context/expectations are vital for any stat, and then get into what the right benchmark is for each purpose by going back to what the purpose was.
   5. Hans Van Slooten Posted: August 09, 2016 at 10:37 AM (#5281125)
I completely agree that most sabermetric primers jump right to saying, "Here's linear weights... Here's Runs Created... Here's WAR..." Without understanding why these were created, they will lead the average baseball fans to just think, "Damn nerds with their fancy stats."

Primers should first start with what lead us to where we are today. Why are Batting Average and RBIs a poor tool for analyzing baseball players? How can we determine the value of a PA? How can we predict the future value of a player?

I guess this is just a restatement of #4 that we need to be better at explaining the "mindset" of sabermetrics which will go a lot further than just spouting stats at people that already think the stats are just meaningless numbers that nerds came up with to look smart. You could literally have a primer on advanced statistics without discussing ANY advanced stats and just reviewing the method by which you should analyze baseball and I think it could be useful (obviously I think you should still provide a backgrounder on the main ones).
   6. villageidiom Posted: August 09, 2016 at 01:57 PM (#5281292)
Primers should first start with what lead us to where we are today. Why are Batting Average and RBIs a poor tool for analyzing baseball players? How can we determine the value of a PA? How can we predict the future value of a player?
If you really want someone who likes BA to adopt other metrics, start walking them there using hits. Hit totals can be used as a measure of quality. But they are improved as a measure of quality when you compare hits to opportunities for hits. Thus, dividing by AB gives a better answer.

There are two ways you could go from there. I think the tendency is to go with the first one below, but I'd say the second one is the way to go.

1. But even though BA is an improvement on hit totals, it still has flaws. Walks are quality. Triples are higher quality than singles. Reaching on error is part skill, as speed has an influence. So you have OBP, and you have SLG, and you have other metrics that look at everything a player does. Those are better.

2. Everything that has been built from there is the same concept: use a benchmark to get a better answer to a question. It's just a matter of what the right benchmark is. All of that depends on what you're trying to determine.
   7. Fred Garvin is dead to Mug Posted: August 09, 2016 at 10:29 PM (#5281728)
I completely agree that most sabermetric primers jump right to saying, "Here's linear weights... Here's Runs Created... Here's WAR..." Without understanding why these were created, they will lead the average baseball fans to just think, "Damn nerds with their fancy stats."

It will also lead to other basic (and good) questions: Why do we need all these different stats if they essentially measure the same thing? Why should one be used rather than the other?

Answering those questions will help diffuse the "nerds with their fancy stats" feelings even more.
   8. Steve Treder Posted: August 09, 2016 at 11:47 PM (#5281755)
Not a SABR member


Why not? Please explain.
   9. Mike Webber Posted: September 14, 2016 at 02:25 PM (#5300303)
Wednesday June 28, 2017
Yankees at White Sox
Mets at Marlins

Thursday June 29
Yankees at White Sox
Mets at Marlins

Friday June 30
Yankees @ Astros
Phillies @ Mets

Saturday July 1
Yankees @ Astros
Phillies @ Mets

Sunday July 2
Yankees @ Astros
Phillies @ Mets

I guess SABR goes to Queens!
   10. villageidiom Posted: September 14, 2016 at 04:37 PM (#5300438)
Why not? Please explain.

Because this site exists, and for more than a decade has been the only outlet for baseball discussion that I have any measure of time for.

I guess SABR goes to Queens!

I'd noted previously in the 2016 BTF NYC Softball thread that we should time the annual BTF softball game to coincide with the SABR convention. If you think a softball game in Central Park would appeal to SABR membership it might be worth going over to that thread and asking local folk to book another field, preferably adjoining, for 7/1.
   11. Mike Webber Posted: September 17, 2016 at 09:45 AM (#5302034)
Some of these biographies don't really give all the details Chris Dial passed on to us at the convention...
1934 Cardinals Biographies now on-line
In SABR's The 1934 St. Louis Cardinals: The World Champion Gas House Gang, edited by Charles F. Faber and published in 2014, we highlighted one of the most colorful crews ever to play the National Pastime.

What a cast of characters it was! None was more picturesque than Pepper Martin, the “Wild Horse of the Osage,” who ran the bases with reckless abandon. The irrepressible Dizzy Dean delivered on his boasts by winning 30 games in 1934, the last National League hurler to achieve that feat. Dizzy and his brother Paul Dean accounted for all of the Cardinal victories in the 1934 World Series.

The club’s hitters were led by the New Jersey strong boy, Joe “Ducky” Medwick, who hated the nickname, preferring to be called “Muscles.” Presiding over this aggregation was the “Fordham Flash,” Frankie Frisch. Rounding out the club were worthies bearing such nicknames as Ripper, “Leo the Lip,” Spud, Kiddo, Pop, Dazzy, Ol’ Stubblebeard, Wild Bill, Buster, Chick, Red, and Tex. Some of these were aging stars, past their prime, and others were youngsters, on their way up. Together they comprised a championship ball club.
   12. Mark Armour Posted: September 19, 2016 at 11:47 PM (#5303044)
For those of you who went to the the convention in Miami, you might want to check out the video of Giancarlo Stanton's home run tonight. It traveled over the bar we were all drinking at pre-game.

Impressive Video.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

 

Page rendered in 0.3469 seconds
59 querie(s) executed