//
you're reading...
Uncategorized

Don’t Look Behind the Curtain!

If you’re not taking advantage of the “level the playing field” nature of Twitter…you’re missing out. This fast-rising online tool brings celebrities, bloggers, athletes, and even fans together in one massive chat room to “tweet” ’til they just can’t “tweet” no more. Sure, sometimes you get the irritating inspirational messages at random by one-time celebrities like Kirstie Alley (good lord, woman…please STOP being relevant), but for the most part, the “pointless babble” (as USA Today called it) on Twitter is full of casual interaction with otherwise out-of-reach personalities (see, Matthew Leach, Derrick Goold, and even the ever-cynical Joe Strauss – love ya’, Joe!). 

But every once in a while…you get wrangled into a twee-bate (that’s Debate on Twitter for the un-twitterized among you).

Perhaps a bit more information is necessary. On Twitter, a user is limited to statements and/or replies of 140 characters or less (known as “updates”). Imagine your frustration with such a limitation as you try to single-handedly turn back the tide of Statistical Snobbery in Major League Baseball (heroism, it appears, is severely under-appreciated). In short, it just don’t work.  On to the blogs!

This morning, I found myself in a somewhat spirited twee-bate with @fungoes (that’s a user handle on Twitter, kids) about the importance of OBP (on-base percentage) vs. BA (batting average). If you read his account HERE (I am, sadly, identified as “among others”…oh, the shame), you’ll have a head start. In fact, I would go so far as to say it is necessary before continuing with this post…seeing as this post is somewhat of a reply to his post. Go ahead…read it.

Done? Okay…let’s continue.

Let me start by saying I love OBP. I really do. It’s an outstanding advancement in statistical analysis for baseball (or rather, it’s recent applications are advancements). I do not, however, believe it is a replacement for the traditional (Fungoes used “antiquated”) stats. Instead, OBP adds to the 100 plus year old knowledge bank of baseball statistics. Make no mistake…it adds to it in a major way…but it is not the beginning and end all in baseball statistics. It is a way for us to see a more-detailed image of a baseball player. It is a significant component in attempting to predict a baseball player’s success and a team’s win percentage and run totals.

But it is not the Holy Grail of baseball.

Okay…enough about that…let’s get onto what really matters: Replying to Fungoes’ post. On his blog, Fungoes puts forth two separate fantasy teams made up of 9 players each. On one team, the top-ranked players by BA. On the other, the top-ranked players by OBP. He then plugs the statistics of each player into a “lineup analysis tool” apparently supported by a fantasy baseball site. In the end, his conclusion is that the OBP team scores just over half a run (.533) more than the BA team. Then, he continues by predicting the OBP team would win, approximately, 8.6 more games than the BA team.

Okay…all well and good. I was actually getting ready to accept the conclusion and move on…but wait! What if we look behind the curtain? What if we question a bit of the data we’re being fed? Hmmmm…well, then…we notice a few problems with the analysis. So…after that incredibly long-winded introduction, I present 3 primary flaws with Fungoes’ post and results. Keep in mind, I’m not saying Fungoes is not right…but I am saying this calculation and the resulting conclusions are more than flawed.

1. Unrealistic Lineup Composition for Both Teams

This is really a minor problem as it doesn’t quite taint the results as much as it simply calls for more realistic test subjects. It seems clear that Fungoes gave very little (if any) thought to the positions his team members would play in a real setting. For example, his OBP team alone has 5 first basemen (out of 9 players) and no player capable of playing center field. At this rate, the runs this team would give up in the field may well be over its projected offensive production of 7.553 runs per game! I jest…but my point is this…if you’re going to use OBP to construct a team, you can’t use it in a vacuum. This is one of the dangers I see with many baseball statisticians crying “Savior!” at the so-called “advanced” or “sophisticated” stats. They tend to see things in a vacuum. In reality, it just doesn’t work that way. OBP is helpful in constructing a lineup…but it is by far absolutely NOT the only thing to consider. RBIs, Slugging Percentage (SLG), Defensive ability, etc. are all important. If we continue to try and simplify baseball into 1 singular “keystone” stat, we take away much of what I love about this game – its multi-faceted nature. Okay…enough about that…let’s move on.

2. Wrong Tool for the Job

In Fungoes’ blog post, he is the picture of transparency! For each team, he lists the lineup along with the “AVG.” or “OBP” he used – well…sort of. You see, the “lineup analysis tool” Fungoes used…it kind of works like this (in @fungoes words): “The lineup tool uses whatever OBP and SLG values the user types in and generates run totals based on how much each impacts runs.”

“Sooo…you used a tool that doesn’t even use BA as a sep. stat to compare a BA vs. OBP team? tool itself leans towards OBP over BA?” I asked (innocently, of course).

“It doesn’t ‘lean towards OBP’; it uses OBP and SLG to calculate runs,” he responded.

Okay…anyone else see the problem here? We’re supposed to be testing BA vs. OBP to see which team scores more runs…and yet, the tool used to calculate runs only uses OBP…not BA. I call foul!!! Alright, kids…if you’re going to see which stat has a greater impact on runs scored, doesn’t it make sense to actually USE the stats you’re testing? I would think so. Here’s the basic problem…the tool itself is built on the premise that OBP and SLG impact runs…but BA does not (otherwise it would be included in the calculations). If this doesn’t make sense at this point, I can’t help you. Let’s move on…

3. Where Did SLG Go?

Oh…what? What’s that you say? Where did the SLG go? ……..HEY! You’re right! When I go to Fungoes’ blog post, I don’t see the SLG numbers he entered. Well…let’s just check that out using the links he provided us. Ah…there they are…I think…um…hey…wait a minute. Alright, you gotta’ see this:

BA Team Stats:

 BA Team Stats

 OBP Team Stats:

OBP Team Stats

Now, I’m sure we all have horrible memories of our high school math classes, so let me do the hard work for you. Okay…here we go:

BA Team Avg. SLG:   .534
OBP Team Avg. SLG:  .547

I’m getting tired, so I’ll just wrap this up for you. When entering his stats, Fungoes used SLG percentage…but when he got his results, he automatically attributed the entirety of those results to higher OBPs when in fact, the OBP Team’s SLG percentage was higher on average than the BA Team’s. Now, this doesn’t account for the full difference between the teams – the flaw in the tool still gives the edge to the team with the higher OBP (flaw ONLY because it shouldn’t be used for a BA related comparison…it does what it is designed to do, however) – but it should at least be taken into account when presenting the results.

Alright…I’m getting the “wrap-it-up” signal from my wife, so let me close with two points. First, Fungoes does a great job, and I value his insight, but the post about this particular topic on his site is tainted by the assumption that OBP “trumps” BA and therefore is a poor evaluation of whether it does in fact trump BA. It’s just bad statistical analysis to use a tool that leans so heavily in favor of one outcome. If you use a tool that uses OBP but not BA, OF COURSE the team with the higher OBP numbers is going to score more runs. I could have saved him a whole lot of time on that one.

Secondly…I want to make this clear…I am a big fan of OBP. At no time did I argue that BA “trumps” OBP…or anything even close. Instead, my argument has always been simply this: OBP, like any statistic, gives us a more detailed, fuller picture of a player when used in conjunction with other statistics (traditional and modern). It does not, however, replace traditional stats. You cannot build a team of only OBP guys and then match up against a team that takes a more balanced approach. Okay…the last part of this point is this: My contention was not even about BA alone…it was about all stats other than OBP. RBIs, RISP, BA, SLG, etc. I like them all when constructing a team. I’m not going to look at just BA or just OBP. I want a team of guys who can get on base and guys who can bring them in.

Okay, okay…enough of that…I have to get home for dinner. You all have a wonderful night…and GO CARDS!!!

Advertisements

Discussion

No comments yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

CLICK TO READ: The Magneuris Sierra Effect…

5 Ways the Cardinals Jumped into First Place

Player of the Month Luke Voit Leads Plethora of Memphis Mafia Hitmen

CLICK TO READ: Can Jedd Gyorko Rescue the Cardinals from Themselves?

CLICK TO READ: With Peralta on His Way Out, Cardinals Need a Third Baseman

CLICK TO READ: Switch It Up! Cardinals Need to Swap Diaz and Carpenter in the Lineup

CLICK TO READ: If They Won’t Fire Matheny, Here’s 5 Things the Cardinals Must Do ASAP

CLICK TO READ: As Piscotty Goes, So Go the Cardinals…

CLICK TO READ: 10 Cardinals to Get Excited About…

Find Posts By Category

The NL Central

Click to visit each team's site.

Buy the 2011 St. Louis Cardinals World Series 8-Disc DVD Set!

Click here to view the 2011 St. Louis Cardinals World Series Collector's Edition DVD set!

2011 World Series Game 6 BluRay & DVD Combo Pack

Enter your email address to subscribe to Cards N Stuff and receive notifications of new posts by email.

Join 1,563 other followers

Archives

%d bloggers like this: