BABS and second-level thinking
by Patrick Cloghessy
Overrated, underrated, or properly rated? Our perpetual conundrum is at a fever pitch this time of year. It’s crucial to distinguish between where we personally rank players and where the market places them. Finding value is the name of the game and we must decide whether or not we are buying at the market price.
What factors into the market price is a mix of (mostly) skills, history, potential, hype, and recency bias. When a player doesn’t live up to his draft status, for any reason, it is not uncommon to see his price drop the following season. This bias can carry enormous weight.
If we take for granted that the wisdom of the crowd (ADP/market price) is correct in its construction of the player hierarchy, we are liable to miss something.
For instance, Javier Baez (p,SB.a) has seen his ADP jump from 105 in 2018 all the way to 15. Javier’s fantastic 2018 has the market salivating over his power/speed combo. BABS is less convinced of the power ceiling, throwing a bit of cold water on the consensus. She still values his triple asset combo, just less than the crowd does. The room for profit is slim, and a full bid for his services counts on a repeat with little to no regression.
On the flip side, Jonathan Schoop (p,a) has fallen from ADP 64 in 2018 to 182 this season. The crowd’s opinion of this player changed. Why? He had a down season largely due to an oblique injury. Sure, he leaves the homer-friendly confines of Camden Yards for Minnesota, but might that drop be a touch drastic? BABS still likes his dual asset skills.
So far, we have taken a dive just under the surface of two players, revealing a key component to success in the quest to find value, what Howard Marks calls second level thinking:
First-level thinking says, “It’s a good company; let’s buy the stock.” Second-level thinking says, “It’s a good company, but everyone thinks it’s a great company, and it’s not. So the stock’s overrated and overpriced; let’s sell.”
First-level thinking says, “I think the company’s earnings will fall; sell.” Second-level thinking says, “I think the company’s earnings will fall less than people expect, and the pleasant surprise will lift the stock; buy.”
We are not buying stocks, but we are buying stock in players. We are searching for value. Again, Marks:
First-level thinking is simplistic and superficial, and just about everyone can do it (a bad sign for anything involving an attempt at superiority).
All the first-level thinker needs is an opinion about the future, as in “The outlook for the company is favorable, meaning the stock will go up.”
Second-level thinking is deep, complex and convoluted. The second-level thinker takes a great many things into account:
What is the range of likely future outcomes?
Which outcome do I think will occur?
What’s the probability I’m right?
What does the consensus think?
How does my expectation differ from the consensus?
How does the current price for the asset comport with the consensus view of the future, and with mine?
Is the consensus psychology that’s incorporated in the price too bullish or bearish?
Exploring and analyzing the market on a macro level is useful and necessary, but we can drill down even further.
Are you preparing for your home league, with a few low commitment opponents? Is your “big-money league” the No. 1 priority? Have you invested time and bankroll in high-stakes leagues? Normally, the larger the cost to enter, the stiffer the competition. As we raise the stakes and the level of difficulty, we also increase minimum requirements for victory. Sure, we can search for black swans and hope luck blesses us with a $3 player who becomes a $40 stud but the odds of success are slim. We can do better. Enter second-level thinking.
It doesn’t matter if you are assessing Christian Yelich (p,s,A+) for your home league or your high stakes league, the process is the same. Is the market overreacting to his magical 2018? There are whispers that Yelich may be due for regression. He can’t possibly repeat the ridiculous 35% HR/fly ball rate while only hitting 24% fly balls, so he is not worth a high draft choice.
The whispers could be right. More importantly, you must form an opinion on how right (or wrong) you think they are. At any point, is regression still ok? Will Miller Park continue its HR charity campaign to LHB?
Yelich earned $44 last season. If he falls back, as long as he doesn’t fall all the way back to previous levels, he can still earn over $30, which puts him in the range of first rounders. For her part, BABS sees Yelich as one of the most highly skilled bats in the game, despite her disbelief in his 2018 power spike. Judging by his ADP (7), the market agrees. What say you?
After being selected in the top 5 in 2017 and 2018, Jose Altuve’s (s,A+ | inj-) ADP since February 1st is 13. His first ever DL trip forced him to appear in just 137 games in 2018, torpedoing his fantasy production. It is hard to ignore Jose’s roto values from 2014-17: $50,$40,$46,$45. It’s not without risk, but a late first/early second round selection nets a player with a long history first round finishes.
Adalberto (Raul) Mondesi (p,S+ | inj-,EX) has drafters extrapolating his 275 AB from 2018 into a full season of superstardom. The hype is real, but is the player? BABS sees real skill but also elevated risk. Is the cost commensurate with the likelihood of return? Mondesi (pictured) would not be the first player to tease us with a small sample only to fall short of expectations thereafter. See the 2011 seasons of Desmond Jennings and Brett Lawrie as examples. Strikingly similar output over a small sample elevated draft prices the following season only to disappoint. Anecdotal data points aren’t proof, but experience risk is real, and BABS marks it with a big EX.
Second level thought is incorporated within the cells of the BABS spreadsheet. Each column is a chapter in a player’s story. BABS is able to exclaim, without raising her voice, who stands out in a crowd, and why. At the draft, performance under pressure will be enhanced with BABS as your companion, whispering sweet, actionable intel in your ear.
This is a good article for asking the question, but what is the answer? The last paragraph that attempts to answer the question is mostly platitudes and fluff. I agree that BABS has some actionable intelligence, but I have been constantly asking questions since joining, and I haven’t gotten many answers. BABS seems to like aging, formerly great players without accounting for age other than a marker to note it. It also seems too simplistic—there can be a negative for average because that is a rate stat, but not a negative for power, even though occupying a roster spot with a zero in power (looking at you, Dee Gordon), can tank two categories as much as Joey Gallo can tank average. With limited spots, you can get killed in counting stats with a dead spot in them as much as you can in a rate stat like average, but even Dee Gordon cannot get a negative power rating.
I loved the article on the Athletic today noting the failures of ADP last year–but it didn’t compare ADP’s results with other ranking systems, such as BABS, but my questions about the same were unanswered. Was BABS actually better than using ADP in a snake draft last year or not? That’s information that can inform how to use BABS this year. Inquiring minds want to know, I want to know. (All credits to where I got that from, I just don’t remember, I think it was a commercial.)
I love outside-of-the-box thinking, and I am happy I ponied up to see the full extent of BABS (the “free report” on various places (I saw it on MLBTR) was a great advertisement), but I would love to see the idea tested and contested here, rather than a group think. I have had one person engage in the 3B thread, but otherwise, it has been crickets. That is a bit disappointing. Can BABS put up or just show up? With a bunch of auctions coming up, I would love to know.
Great reply and I cant wait to see what Shandler posts. I have been using BABS for a few years now and she won me a league in year one mainly by finding MRP with save potential. It was epic, one day I had 7 closers and crushed. Truth be told, she feels nothing more than a draft tool based on enough sample to quantify a rating per the skill, but in-season I continuously found myself passing on guys that were blowing up because their BABS ratings didnt suggest a roster spot based on the skills from the report and I found myself in a hole that was impossible to get out of. I still have a lot of faith in her, but I certainly would love to see how she stacks up too and how to progressively use her ahead of the herd in-season.
Nice article here, addressing the toughest area of Babs to manage. For example,Babs rates Archer as er/KK, Nola as ER/K. Last year Archers K/9 was 9.8, Nola’s 9.5. But in the second half of last year Archers was 10.8, Nola 10.2. Who do I think will strike out more guys this year? Babs says buy Archer, I don’t agree. Archer hasn’t helped any team win a league in a few years. I set the over/under for Nola’s ERA @ 2.85, I doubt Babs agrees. I’m with the market on this one. Clevinger, Taillon, Marquez, the market thinks will become aces this year. Babs disagrees, but finding breakout guys is big. I know the endless search for upside is a trap but you have to gamble sometimes on which stocks are going to become Apple. I find myself struggling to follow Babs wholeheartedly in this particular area. Being able, in draft, to bully hitting early, and take shots on breakout pitching is something I like to do for the advantage in offense. I just don’t use Babs for my breakout targets. I do like and use Babs, however, just not so much where in this area. It does help a lot, I just struggle with what the article addressed, thanks!
I agree with everything you say, but think your disappointed tone reflects unreasonable expectations. Don’t throw out the projections in Baseball Forecaster, but supplement them with a BABS filter, which I find particularly useful in late rounds–e. g., Rosario or Hampson to add some speed. You’re right about Dee Gordon, who is, in my opinion, about the tenth-best choice (now I’m exaggerating) at second base. Travis Shaw is a real steal at his ADP if you’re not looking to get speed from your 2d base starter. I also like Goldschmidt more than BABS does, and it’s (the she thing is getting old and distracts from the content) not going to convince me that as a late 2d round pick he’s not a good investment. Still, I think it’s useful; if you can get Ramirez and Story in the first two rounds you’re off to a good start, and BABS agrees with the projections here. Yelich is a better choice mid-1st round than T. Turner if you can get Story (a better choice) in the 2d round. Anyway . . .
Did a mock draft last night using BABS advice for the first time. Definitely improved my results. Still learning!
Most of your questions are answered in The BABS Project eBook. On two points – a negative for an average category is most certainly not the same thing as a negative for a counting stat. A negative in a counting stat is a missed opportunity; often that can be made up later in roster construction. A negative in an average is a double whammy – it’s a missed opportunity PLUS backwards movement in that category as well. It takes at least TWICE as much positive influence to correct that negative. As for runs and RBIs, well, BABS is only about underlying skill.
BABS was back-tested extensively prior to its release and — used correctly (which is vital, but often not accomplished) — it has proven far better than ADPs. However, BABS – and all systems – have been challenged to keep up with the changing statistical landscape in MLB. We adjusted the metric benchmarks that generate the ratings in 2018 to account for some of those changes, but MLB kept moving the fence out. It’s tough catching a moving target.
In the end, the only true measure of success is whether any system works for you. If you win drafting off an ADP list, more power to you. No system is perfect, but BABS thinks she’s pretty darn close, lol. (And yes, it’s a SHE).
Adam, I enjoy talking about BABS vs other approaches, etc. perhaps we open a thread on the Forum, and start with the questions you still think have been unanswered, and I will be happy to explore further with you and others?
I would love to see some of the back-testing numbers. They interest me. And, despite what it seems from my comments, I am not anti-BABS; I generally just see no point in commenting unless I don’t understand something or disagree with something—agreeing with something generally requires no comment.
And no, I am no fan of ADP, I just thought it would be an interesting comparison to try lots of systems through the interesting mechanism you used in the ADP article on the Athletic to test them.
I am also not sure about your average being a double whammy theory—a blank spot in the roster, for example, does not hurt average at all; in a deep league, for average, it is usually better than something you can get off the waiver wire if a player gets hurt and you don’t happen to have a bench player to fill in. A “nothing” in batting average is neutral. Obviously, the opposite is true for the counting stats, which are four out of five in a 5×5, and why we don’t leave blank spots (although it is awful tempting in deep two catcher leagues sometimes, lol).
And maybe you are right about one; it is about how BABS is used. Drafting based solely on the list of BABS ranked players would probably be silly—using the list for bargains, both early and late, is probably the way to go.
Finally, just to be silly regarding pronouns; BABS may well be a she, but she’s a she, and it’s and it, but it’s never a she, simply based on the reflexive properties of objects.
Sounds good to me. Which forum would be the best one to use? The general forum?
I love Adam’s comments as I often think FB ratings are like psychic predictions. You generally only hear of the psychic’s hits, not their 99% misses!! In fantasy those hits are used as fodder to say my site/prognostications are superior. I don’t think from my perusing of sites over the years have I ever found a lot of self analysis IN ANY DEPTH as to how that particular site/person’s predictions feared the prior year.
So, I think one goes with their own ideas, formulated on those from whom you believe give solid advice and hope that it works out for that year. And we need to remember that LUCK is often ranked 2nd or 3rd in surveys addressing the various factors that make for a great fantasy season.
I’ve used BABS for 2 years in my 10 team snake draft, last year relying on near 100% going with a strategy of picking hitters straight thru the 10th round. My thought was I’d get semi decent pitching later, as well as using BHQ’s weekly pitching matchup info for my weekly roster. Yes, I got badly hurt in the pitching cats, primarily in K’s and SVs, not so much in WHIP and ERA. But, along with my great hitting stats I ended up tied for second, missed first by a half point.
This year my draft will again rely primarily on BABS. Again, not so much during the season. I am thinking of going after starting pitching MUCH earlier. Why? As they say the proof is in the pudding. I needed a major arm last year, and from reading the touts I pin my hat on, the consensus this year is mounting that it’s no sin to take a major starter or two in the first 10 rounds.
Great reply. I think BABS is great after 3 seasons. I assume you might be new to BABS – reading into assumptions it might not work. I would say “Give it a try” and sounds like you are. I feel like everyone uses ADP, AAV, $ amounts, etc.
I like trying something different.
With any system, you must “trust” the data and system in which BABS I do but I also take into account (as many should) you own analysis, gut, $ value approach, etc.
My mind is not a computer so many times I find myself reading the BABS ratings and looking at the player and remember a bad year, good trend, bad trend, new park, etc. It helps me out in that regard by far but I also couple it with RotoLabs and I feel I get the most out of each and would never do it with just one or the other.
I tend to use BABS as a mining tool. Turning loose of my comfortable cocoon of numbers was harder than I imagined and honestly I’m not completely sure I want to at this stage. Traditional projections can also account for recency bias, age and injuries, they should be baked into the numbers. But comparing similar players through a marketplace lens is much easier with BABS than a Cheat Sheet full of names and numbers. I look at an Asset group and break down and look for the safer places where I’d want invest my money/picks near the top of the group, where I know it will cost me to play based on the ADP/AAV. And I look for potential bargains near the bottom of the group. And then I start digging into them and what I think their potential is and what will it take to roster them. I use this process to augment what I typically do with a traditional Cheat Sheet. I’m still not drafting full blown BABS yet, but I certainly like what it has brought to my draft prep ritual.
Start it wherever you want, will find you…
And BTW, I meant to mention this earlier, the original article was well-written and thought provoking. Good job, Patrick.
Thanks, Merv.
And thanks to all for reading.
Hi Ron could you post a rotisserie draft board for us. I find it’s a great reference/ visual and tool for our drafts
Draft board? I’m not sure what you are referring to. The Spreadsheet and Database provide the only reports we have.
I meant a copy of one of your drafts with the experts, you had one last year
Coming tomorrow
Thanks!! Love your work