The true test of BABS
Success in any one season of playing fantasy baseball is a combination of two things: skill in drafting a roster plus skill in managing that roster during the season. At its core, this process might be described as the combination of: a) comprehensive off-season research and analysis, plus b) reactionary, small sample reflex decisions.
For BABS, this is problematic.
Before our drafts, we do painstaking deep analyses of skill, opportunity and risk, and the context in which it all lives. This effort creates the “projections” that shape the choices we make in constructing our roster.
But once the season begins, we are driven by tiny data samples, recency bias and panic in reshaping our roster to keep up with the other teams. We have no choice, right?
Coming into last season, Detroit pitcher Buck Farmer had a career MLB ERA of 6.82 and was projected for a 4.80 ERA in a smattering of inconsequential innings. BABS rated him as (|-ER,e) – no assets and enough liabilities to warn off even the most devoted Tigers fan. He went undrafted in pretty much every league.
But in his first start on May 27, Farmer struck out 11 batters in 6.1 shutout innings. If you were lucky enough to be in a league with owners astute enough to play it cautious, you might have gotten the opportunity to see another start before jumping in. Regardless, Farmer threw 6.2 shutout innings the second time around and then all bets were off. The FAAB wallets opened.
Over his next four starts, Farmer threw 13.1 innings and gave up 21 runs (14.21 ERA). It’s fun to cherry-pick, but the riskiness of small-sample analysis is exceedingly high.
The bottom line is this: Every player we add to our team during the season means we’re potentially eliminating a player we rostered through more careful analysis. And given how much turnover a typical fantasy roster undergoes during the course of a year, our in-season roster management can negate much of our off-season player evaluation efforts.
I wrote about this in the 2012 Baseball Forecaster:
“Even if our off-season analyses paint glorious pictures about the players we like, our limited roster space and quest for the best statistics force us to toss some of them away to take chances on hot hands. And even though we do look at past track record, the “hot hand” analysis tends to be the overwhelming decision-maker. In other words, we put in obsessive draft prep to analytically determine which players are going to give us the best six months of production. Then we’ll undo all that work by gambling on a fringe player having a hot April. (You know the thought process: “What if this is real?!”)”
The issue is this: given that BABS is a tool best used for draft preparation, the in-season undoing of our draft rosters also undoes any insight into how BABS fares at player evaluation.
We need a way to more accurately evaluate the core competency of roster construction, and at the same time, hone our skills and gauge our efforts at assembling an optimal roster. I think the best way to accomplish both is by creating some type of fantasy competition that emphasizes our winter efforts. And we need a BABS-specific format.
Some key elements:
1. At one end of the spectrum, a fantasy format that might best serve this purpose is a Draft & Hold league. Build your 23-man roster and let it ride. D&H leagues are truer representations of our prognosticating prowess because our rosters are built solely from our off-season analytical efforts. These leagues let us watch and analyze, but remove the temptation to continually tinker and tweak with new talent that comes into the league during the season.
However, we know that injury turnover would devastate this type of competition. Even adding a small reserve list would not be enough to offset the attrition. But I think the concept does need to start here.
2. The key phrase in the Forecaster excerpt above is “limited roster space.” That is a real obstacle. So the next part of BABS Baseball (let’s call it BABL) is large rosters. Maybe we keep the 23-man actives but we need to have a deep enough reserve list to weather the inevitable attrition. Forty players total would be a start; 50 might be even better. NFBC’s 50-player leagues have the right idea.
3. For many of us, the best part of the fantasy experience is the draft. I think we should make this game accessible enough that we can easily participate in many leagues.
Again from the 2012 Forecaster:
“My first reaction was pleasure at the idea of more drafts. I think this is why people prefer the idea of drafting in many leagues; it creates more opportunities to be right.” It’s true. That’s why we play; to find out if we’re right.
“Each year, I value several dozen players more highly than others do. I’d love to figure out how to roster all of them and to test my expectations against others. But I’m limited by the number of leagues I’m in and I can’t commit to more leagues than my time will allow. Playing multiple D&H leagues would let me accomplish those goals.”
More leagues give us more chances to roster more of the players we like, not just the 23 we end up with. More leagues give us more chances to experiment with various roster construction strategies. More chances to be right.
On Friday, I’ll introduce the format we’ll use for BABL in 2018. This will be a test year. BABL will be our laboratory. Then we’ll start setting up some leagues.
I’ve started to draft prep a little and I think roster construction continues to be the issue for most BABS players. The right mix of healthy productive players with injury plagued ones with upside is what we are striving to get. For example, if you need 6 starters, how many should be without injury issues ?? How many with less skill but logs 180+ innings?? What’s the right mix?? I continue to think BABS extrapolates this better than any system.
How difficult would it be to re-structure BABS lists (in the database) for OBP leagues? Doing it all by hand is obviously possible (lower AV guys, raise PW & * guys) but with so many OBP leagues getting more popular I figured its at least worth asking…
We’ve provided the OBP indicators. There are too many format variations to provide custom lists for everyone.
This feels like a dumb question, and probably covered last year; but if one was comparing a pitcher from each of the Phillies, Reds and Mariners, and the Mariners pitcher was a bit more injury prone than the others, would that be at all mitigated by his playing for what we perceive to be a team that provides more run support? While some games may be missed by the Seattle guy to injury, Philly and Cincy pitchers might miss out on a win from time to time because their team can’t score and reward a solid pitching performance. I’m probably in the weeds and should ignore the team designation and stick to the ratings, but the thought does keep arising.
You’re overthinking this. You are trying to compare disparate variables. A benefit ion one doesn’t necessarily offset a deficiency in the other. Regardless, you can’t predict them, or their relationship, anyway. Stick to the skills; BABS already said to pretty much ignore team context.
Thank you. Btw, the article describes how BABS sticking to her ratings might lead to fewer in-season moves, but this isn’t always the case. Opponents are often adjusting their perceptions of players so BABS remaining constant creates a different view. For instance, last season I took Eric Thames in a draft because Brandon Belt was already taken. One week into the season (yes, that quickly) someone dropped Belt. I made the switch for Belt since he was more highly rated and I ignored the first week of stats. I regretted the move for a little while, but by the end of the season neither player was on my roster. My example isn’t great, but BABS can create opportunities, especially if the guy who picked up Buck Farmer dropped someone like Kenta Maeda, for instance.
Perhaps this is a question for Ron. But it seems to me that the BABS roster construction (for pitchers) is based upon no minimum IP. My NL league has a measley 900 IP minimum. I feel as well served with 3-4 SPS, with the remainder of the pitching staff filled in with higher skilled RPs (which are cheaper and more abundant). Am I wrong?
Does your league use the category of Strikeouts? If it does, you already have a proxy for a minimum innings rule. If you need to accumulate Ks, you need to accumulate IP.
3 QUESTIONS:
1. I KNOW THE GUYS I WANT, BUT WHERE DO I DRAFT THEM? BY ADP OR MY ORDER?
2. WHEN DO YU START PICKING SP FOR YER TEAM, AFTER THE TOP 5-6 ARE GONE? CLOSERS?
3. HOW DO YU ‘MESH’ HITTERS AND PITCHERS, USING BABS?
THANK YU!
PS: CAPS = VISUAL PROBLEM
1. The BABS ranking is the foundation of “who is better than who” but the ADPs are your reality check to make sure you don’t draft someone too early. Matt Carpenter may be listed high, but you can wait awhile on him, knowing that you’ll potentially be getting advanced skills for where you can draft him.
2. It’s really your call. Follow the market. Where are others drafting SPs and closers? Jump in to make sure you are not closed out – on any position – but that decision is based on your needs and the marketplace.
3. The database does not provide an integrated list, but I will be publishing a spreadsheet with batters/pitchers ‘meshed’ – soon. I’m waiting for a few more free agents to sign. Also, you can get an integrated list with RotoLab right now. But be aware, our suggested ranking of where to draft pitchers/closers is not gospel. (See #2.)
NOT TO SAY I’M A SMART-ASS, BUT THOSE WERE MY INTERPRETATIONS ALSO! THANK YU!
LASTLY, THE DOLLAR AMOUNTS ARE JUST INFO, WRIGHT?
If you are in a draft league, then yes.
WHOOPS! I AM… ARE THEY A TARGET BID THEN?
No, just a general benchmark of what the marketplace might pay. They are benched to a 15-team mixed league, so they won’t work for any other format. But your target bid is going to be based on what your league’s tendencies are, so you can’t really used published dollar values for that purpose anyway.