Trout, Freeman pace the field as we enter the homestretch
I should have a little more faith in BABS. She’s been good to me; I should support her even when her ratings seem a little off. To wit:
A few weeks ago, I was a guest of Patrick Davitt’s on Baseball HQ Radio. He was looking over the late July BABS update and noticed some odd ratings. He asked me:
Some names jumped out at me from the top reaches of the July update. ATL 1B Freddie Freeman?
I responded: “Freeman was evaluated in the pre-season with significant power, significant batting tool and on-base ability, but has been showing extreme power this year. Obviously, this doesn’t jibe with the 19 HRs he’s hit so far. This seems like an anomaly – all his indicators point to him being able to hit more HRs. We could take this to mean that there is a power surge yet to come or that he’s just an outlier. There will always be outliers.”
Since August 10, Freeman has been on a tear. Over 12 games, he’s hit 7 home runs and has batted .447 (17 for 38). BABS knows.
Freeman’s current (P+,s,AV+) rating matches Mike Trout at the top of the leaderboard entering September. Josh Donaldson’s .228/.336/.391 month drops him from the top spot after two straight months.
AUGUST 25 UPDATE: [EXCEL] [VIEW/PRINT]
About the charts:
Batters and pitchers, individually, are ranked by skill, and are each separated into two sections. For batters, I’ve grouped the full-time and mid-timers together and separated out those with 30-236 plate appearances. These fringey players clutter up the chart, especially since their BABS ratings are less credible than those with more playing time. Those with less than 30 PA are omitted because their ratings are pretty much useless.
You can still identify the skills tiers within each list, but the separation helps delineate how much faith you should be putting in the ratings. You can still cherry-pick upside bats on the part-timer list, though it’s clearer now that these are speculations.
I’ve separated the pitchers into starters and relievers. The starters are essentially those with around 12+ batters faced per game – so there are some long relievers in there – but this roughly captures all those with rotation potential. By separating out the relievers, we can better see which pitchers are getting the saves and which ones have just as much skill as those anointed with 9th inning work.
BABS has me in 2nd place, close on the heels of the top dog in our league … THANKS! I have 2 questions, hope I’m allowed that. First, is there a reason you don’t list the liabilities of these players (inj, exp) as you did at the beginning of the year? Secondly, how much do you see these skills changing between this last uodate and next preseason? Is that even possible to project?
Two excellent questions!
1. The Liabilities are historical, backwards-looking indicators. While they may change during the course of a season, I don’t want to give the impression that a player’s injury proclivity or experience changes at a particular AB or IP milestone level. So I won’t update them during the season. I suppose I could just republish their pre-season indicators but I fear that I would then get a rash of emails asking “when will Kris Bryant lose his Experience risk?” or “when will you upgrade Clayton Kershaw’s injury risk?” So I’ve opted to omit them in the in-season reports. That said, I am hopeful that the BABS database upgrade this winter will provide more flexibility in this regard, so that at minimum, you’ll be able to pull up the indicators yourself, at will.
2. The database will provide a 3-year scan for every player, which will help us track the progression of their skills. I have added a section in the upcoming “BABS Guide to Fantasy Baseball” that addresses this specifically. Here’s an excerpt of my first pass:
“What’s more, these ratings are not projections. They are gauges of each player’s historical measurable skill. When we start planning for the new season, we may draw some conclusions about how a player may progress or regress, but we’ll never attach a number to those conclusions. You’ll find that, over time, most players do have a more trackable skills progression than their statistics would lead you to believe. Any marked changes in a trend will either be supportable by a change in expectation, or more likely just regress. For instance, let’s say a batter shows the following power trend:
Year BABS
1 p
2 p
3 PW
4 P+
5 PW
This player took a step up in power in years #3 and #4, then regressed in year #5. An analysis of his most recent balance sheet might reveal an injury situation, or a change in leagues, or some variable that might have contributed to the regression.
Going into year #6, we might rate this batter as either PW or P+ depending upon the extent that those variables might affect his future performance. If he was hurt and is expected to be healthy, we might return him to P+, which is a skill level he has shown to possess. If the negative variables will likely continue to be a factor, we might keep him at PW. Or, we might return him to P+ as his natural skill level and reflect the downside on the Liabilities side of his balance sheet.
There are several ways to play it, but you’ll note that we’re still working within a very broad range of outcomes. And we’re not limiting those outcomes to a statistical projection of exactly 34 home runs. Or even 30-35 HRs. Because we just don’t know where that number will end up.”
As I’m mired in last place (well, tied for last with the team that quit playing in June) in my CBS H2H league, my worst finish in 27 years of playing fantasy baseball, I’m trying to sort out what went so very wrong with my BABS team and why.
One thing that is very clear to me is that for H2H purposes, and probably any points league that deducts points for batters striking out, BABS doesn’t punish this “ability” hard enough. It needs to be made more of a liability or at least highlighted in a more useful way. I ended up with a lot of high-K guys who couldn’t do much of anything else, especially hit (Kemp, McCutchen, Gomez, etc.) McCutchen and Kemp have shown mild signs of life lately, weeks after I was mathematically eliminated from the playoffs, but for most of the season they were worse than useless. And the less said about Gomez, the better. I did have the pleasure of beating the first place team last week, so thank heavens for small victories.
The emphasis on experience I think is also problematic; it’s the same defect as we saw in Portfolio 3 and Mayberry, where at the point you think you have enough information so that you can judge that a player has sufficient experience and consistency, they’re already on the downward curve (or in Gomez’s case, plummet). That’s not a recipe for a winning team.
I originally was going to post my whole draft day roster (which you agreed with me looked good at the time) and comment on each pick, but no one wants to hear that. Suffice it to say that I exercised excruciating self-control and held onto all of these guys until Memorial Day (CBS H2H leagues have a very small bench that pretty much has to be reserved for pitcher swapping in and out), which meant that I missed out on a lot of opportunities of young guys coming up and making a mark. As an example: I drafted Eduardo Escobar; in April someone dumped Xander Bogaerts. BABS then rated Escobar more highly than Bogaerts so I didn’t pull the trigger. Oops.
I’m sorry you had a bad season. Nobody really thought that all three of Kemp, Cutch and Gomez would be terrible, but you landed the trifecta. That’s just bad luck, which does happen no matter how good a job you do of draft prep. But also, no tool is going to be able to accommodate every aspect of every type of league. Deducting points for batter strikeouts is not something you see very often. In fact, it’s kinda counter-productive. You award points for HRs (I assume) yet deduct points for the natural collateral damage of swinging for the fences (I know that’s an oversimplification). But yes, you’re correct – BABS does not discount those strikeouts. I don’t think I would make any change to that too, especially not for today’s game where teams often don’t care if their sluggers strike out.
I agree that we need to see a large enough sample before we can accurately measure a player’s skill, and most fantasy leagues don’t provide that opportunity, but I might disagree that players are already on the down curve once they gain that experience. The inherent problem with fantasy is that your competitors are always going to jump on small sample sizes. We have to work within those limitations.
Finally – yes, BABS missed Bogaerts. She thought he was going to regress. Sorry.