Coming to terms with who BABS is
She’s my mistress, I know.
Many of us have found great value in BABS, but we have also been trying to make her into someone else. We’re not satisfied with all that she provides us; we need her to give us more. It took me a few weeks in the Mediterranean to realize that maybe we should value her for who she is – a draft prep tool – and not who we want her to be – an all-encompassing season-long tool.
The Broad Assessment Balance Sheet’s ratings and annual trends are driven by the perspective that underlying player skill has to be evaluated over long periods of time. A 1 for 4 outing tells us nothing. 100 for 400 tells us more. 1,000 for 4,000 tells us even more. Sample size is everything.
I’ve devoted several articles over the past few months to discussing whether BABS can be used during the season. Most pertinent is this one here.
My last comment is where I left things:
“The jury is out whether BABS is more than a draft prep tool.
It is entirely possible that this is where we will end up. As much as we want to try to force a round peg into a square hole, we might just be better off making sure BABS is completely optimized for draft application. I am not at this point yet, but I have not ruled it out.”
Well, I’m getting close to that point. I’ve been following the ratings each month when I update them (the database was updated earlier today, August 7). What I see is the same as I’ve been observing for all three years that I have been using BABS:
- April is all over the board.
- Things start stabilizing little in May.
- June shows more similarity to historical levels.
- July stabilizes some more.
- August starts locking in to the final seasons ratings.
- By September, we pretty much know how things will end up.
Needless to say, this in-season volatility does little good for our roster management decisions. There is really nothing that we can tease out of the ratings for more insight, at least not anything that would apply to all players as benchmarks or rules.
So I began to think, what would it look like if I focused on optimizing BABS for draft purposes? I took a look at your survey results from earlier this season for ideas. Here are a few:
- Review and refine the Assets and Liabilities rating system.
- Improve the database report presentations.
- Provide BABS ratings for minor league players.
- Offer a BABS as part of a draft software package.
These are all high-level action items; each one contains a laundry list of sub-tasks. My early plans are to focus on these over the next six months, from feasibility, to design to implementation. The goal is for all of them to be ready for the 2018 draft season.
In addition, I am thinking about partnering again with a commissioner service and designing a game format that makes best use of BABS and where her true value lies.
It’s all on the first iteration of my 2018 drawing board.
That all said, I do not plan to completely abandon BABS during the season. There will always be in-season updates. But the optimal in-season use of BABS can be best described by one of our readers:
“You can look at the BABS ratings from previous years and compare to current year. It eliminates the clutter of small sample sizes. Could a player show growth in power or average? Sure. However, if its not supported by BABS in the past, it likely isn’t a skill the player owns and you have to dig into Fangraphs data to see what’s going on.”
Or BaseballHQ.com, or Rotowire, etc. Essentially, each player’s historical ratings and the associated trends provide the performance benchmarks. You can take a sideways glance at a player’s in-season BABS ratings but only to identify large variances. A player who has never shown speed but has a (S+) rating on May 1 can probably be safely ignored (barring some change in circumstances). An historically consistent (PW) hitter showing no power requires future research. The bottom line is that you can’t take the in-season ratings to the bank at face value.
From there, we can consult these other information sources that excel at getting into the weeds. I don’t know that their projections using more granular data are any more accurate with small sample sizes. But at least they offer the tools.
Onward…
FWIW. As a skeptic, I did not believe that BABS would be useful in-season. However, my experience this season, is that it has been useful. I am glad that you will continue to publish the database, which is helpful. I understand that the theory is that BABS is most useful as a draft prep tool. The in-season database highlights information that is useful to me.
Thanks!
Yes, I’ll continue to post database updates at least monthly during the season but I will probably be spending less time trying to lure major epiphanies out of the data. Glad you are finding value in it.
Although the reliability and the validity of statistical analysis depends upon sample sizes, That doesn’t mean BABS lacks utility during the season, but only that the utility might have a different application than during preparations for the draft. For instance, I would not expect her to distinguish the difference between an in-season streak (Eric Thames) and a sustainable skill set (Joey Votto). But I would think she can be useful in determining whether to drop a player who is underperforming, or to hang on and wait for a correction (Andrew McCutchen, Matt Carpenter?).
She may also have a roll in assessing just how serious an injury is. For instance, Cespedes doesn’t seem to have his usual power since his legs keep going out from under him.
But there will be those anomalous situations. What has happened to Miguel Cabrera? I suspect that back injury he sustained during the World Baseball Classic is continuing to plague him. Or (I hate to think this) he may be struggling with alcohol again. But his season this year is unaccountable. I hope I’m wrong, but I don’t see signs of a recovery. He’s such a quiet man that his struggles have not seemed to attract much attention from the press.
Yes, there is still utility, but I fear that we might be more apt to cherry pick those players who are well-described by BABS. As noted, I will continue to keep the database updated but likely spend less time in quest of a holy grail. However, external tools will continue to be discussed and embraced.
I used BABS throughout the year but never your updates. The simple reason is that if I believe in BABS fundamental principals of sample size and appropriate risks and skills. Of course there are anomalies, always were and will be but for me, original BABS ratings still endured. Every move I made and am still making reference the March 31st list. That’s why I think you’d learn a ton by doing some kind of study of how the year went ADP vs. BABS. How many Andrelton Simmons were there? (folks with 50/100/150 difference between where BABS had them and their ADP. Here’s how that worked for me.
Even now when I look at FAABs I’m looking for ADP-type managers overestimating short samples, both bad and good. Going back to original ratings caused me to pick up Castellanos, JMcCann, Rupp, Joseph, Sonny Gray and Alex Cobb when they were kicked to the curb. They of course are not big stars but they have been serviceable. I got them by referencing and remembering BABS longer term rating when they were dropped.
More importantly, I drafted almost entirely based on BABS value vs. ADP. I looked at gaps you had listed and then did a probability of the earliest the biggest “BABS-gappers” in your 250 could be drafted. That set my consideration list for each round. The result was of course my ADP draft report card was horrific and yet I’m still competing for league champ.
* 13 of my top 15 picks have spent time on the DL and yet all have still eventually have produced.
* My mid round (15-22) BABS-gappers included low ADPrs RRay, CRodon, OrArcia
* In the reserve rounds my gappers included Pomeranz, Conforto and CDickerson
Summary of the Summary: I love that you’re continuing to be open and learning more but for me BABS, especially when combined with ADP to determine popularity gaps, works all season long. If you use it how it was intended.
As I’ve said before, Ron, fantasy sports is never going to be much more sure than the stock market but that’s why the best market pickers are the people who a. think long term and b. are counter-intuitive (they buy when others are selling, sell when others are buying).
Thanks for your good work, I enjoy being your customer and fan.