Food For Thought
Here’s two pretty similar players with slightly different approaches at bat, but overall, basically the same skills – you pick the one you would rather have. The differences aren’t huge, but they are there.
Player A:
3.0% BB%, 8.5% K%, 52.6% Swing%, 32.1% O-Swing%, 85.5% Contact%, 20.1% LD%, 3.40 Pitches Per PA
Player B:
4.8% BB%, 8.7% K%, 49.4% Swing%, 28.65% O-Swing%, 88.5% Contact%, 21.1% LD%, 3.61 Pitchers Per PA
Player A walks a little less, swing a little more often, swings at a few more pitches out of the strike zone (that’s O-Swing%), makes a little less contact, and swings a bit earlier in the count. Player B is a bit more patient, does a better job of not chasing balls, and puts the bat on the ball more often, while hitting a few more line drives.
Player A is 2007 Kenji Johjima. Player B is 2008 Kenji Johjima.
Yes, Kenji just had a terrible year, and yes, the contract is a pretty big albatross. But as the Mariners prepare for 2009, they had better recognize that if they’re looking for evidence that Johjima’s skills as a hitter disappeared for good in 2008, it’s not there. The only real difference between 2007 Johjima and 2008 Johjima was in his batting average on balls in play – .291 in ’07 versus .233 last year.
Based on his batted ball data, we would have projected his 2007 OPS to be close to .785 – it was .755. Based on his batted ball data, we would have projected his 2008 OPS to be close to .765 – it was .609.
Because he’s slow, he’ll probably never match his batted ball projected OPS, but you’d have to be nuts to think there wasn’t a pretty significant amount of non-skill noise in Johjima’s 2008 results. Filter out that noise, and the 2009 expectation of Johjima’s performance suddenly looks quite a bit better. Let’s not give up on him yet, okay?
We can’t give up on him yet. 3 years/$16.5M sees to that.
Who wants to bet he has a better 2009 and we get to read a bunch of stories about how offseason work with a particular hitting instructor and/or some small mechanical adjustment is the explanation?
The more I look at the numbers the more I scratch my head. What the heck happened last year?!?!
Dave’s numbers seem to demonstrate that there was some very bad luck effecting his overall production numbers in 2008. Looking from the past three years each year EXCACTLY 30% of his hits went for extra bases every year. So that seems OK. Yet his HR/F numbers went from 11.2%, 9.3%, to 6.2%.
It would be one thing if we’re talking about a well-rounded hitter in his age-29 season having a down year compared to his age-28 season or a young prospect failing to build on his breakout rookie season but it isn’t: it’s Kenji Johjima, a (soon-to-be) 33-year-old catcher whose on-base ability relies primarily on making contact. He was always going to be susceptible to the vagaries of balls in play but his age and the positional decline history of catchers suggests that he’ll have even more trouble legging out those extra hits and trending towards more “luckâ€-neutral tendencies. (Is LD% + 120 = expected BABIP still the acceptable quick-and-dirty formula for hitting luck?)
Sure, he could experience a bounce-back to something like 85%-90% his 2007 level but, even then so, that’s still pretty poor. More so when you consider the length and amount of the contract.
I’d imagine PECOTA will be as pessimistic he even reaches that level of production as most of us here.
85 to 90% of 2007 Johjima is one of the best hitting catchers in the league. His down year really skews people’s memories of how good he was, and how scarce catchers who can hit a bit really are.
Interesting, Dave. What’s headscratching about that is the sustained nature of the anomaly. How does that kind of bad luck hold over an entire season? A few weeks or even a month or so seems more statistically sustainable … but this? Going into the BABIP tank for that long a stretch … the odds against that seem like they’ve got to be in the millions. Is there much historical precedent for one-year dips like this that are bracketed by career-norm rates?
So, given all this, you play the GM: How do you sort out the Clement/Johnson/Johjima conundrum?
Actually, now that I said that, it sure looks like his BABIP rebounded close to those pre-2008 norms in September. That’s just my guess, though … is there month-to-month data available?
Very, very interesting.
I never would have suspected this was the way you were going from just the stat lines.
Wait, Kenji was .287 / .322 / .433 in ’07. 90% of that is .258 / .290 / .390, for a .680 OPS. That would put him 16th among catchers with at least 400 PAs this year, ahead of only Brandon Inge, Jason Variteck, John Buck, Jason Kendall, and his own actual 2008 self. At 85% (.642) he’d only be ahead of himself. He’d still be much better than the ’08 version, but I don’t think that would make him one of the best-hitting catchers in the league.
Feel free to pick other stats, but I’m not seeing an easy way to consider 90% of ’07 Kenji to be quite that good.
Jim,
Johjima’s BABIP for 2008 by month:
March/April: .187
May: .296
June: .245
July: .146
August: .178
September: .333
Just to piggyback on joser’s comment but, say we figure a conservatively optimistic 80% return to Johjima’s 2007 figures. His projected 2009 VORP (17.7 – 80% of 22.2 in 2007) would rank him 12th in the league based on 2008 league catcher totals. Not bad, top half of the league, but not “one of the best”. And that’s also taking some huge assumptions into consideration – similar playing time, no regression in skills, etc.
Two things…
1) “in the league” means AL, not all of MLB.
2) You’re just looking at raw numbers. Context matters…
Yes. And I’m not sure that park factors are “scalpely” enough to treat righties fairly in a place like Safeco.
It’s complicated.
For instance, Beltre might be considered a warrior rather than a bust by the Steve Phillips’ of the world, say, if he played in Philadelphia.
Jim Thomsen-
One in a million is a vast overestimation.
The “quick and dirty” estimate for BABIP is 12% + LD%. As Dave mentioned, Kenji is slow; so that may overestimate his BABIP. In 2006 and 2007, Kenji’s BABIP was 29.2% and 29.1%, respectively, while his line drive rates were 19% and 20.1%, respectively. So quickly averaging these years (a weighted average would be better, but this will be close) Our best guess for Kenji’s “quick and dirty” BABIP is LD% + 9.6%.
Last year Kenji had a 23.3% BABIP while posting a 21.1% LD rate in 339 balls in play. Assuming 2008 Kenji is the same ol’ Kenji BABIP wise, we would guess that his BABIP would be 21.1% + 9.6%, or 30.7%.
Using a proportions test, we can see what the probability is that a .307 BABIP hitter would only get 79 or fewer hits in 339 balls in play. This probability is .2%. So 1 in 500 is a good guess to the probability that Kenji’s BABIP has not regressed.
That may seem like a lot, but about 200 people had 339 or more balls in play last year. Given that baseball is played every year, by many, many players, 1 in 500 is not crazy enough to justify an after the fact judgment.
I am not saying anyone should expect the 2007 Kenji, but Dave’s Kenji post should bring a rare little bit of optimism to all of us fans.
Two things…
1) “in the league†means AL, not all of MLB.
2) You’re just looking at raw numbers. Context matters…
I totally agree with Conor. When you say a 90% Kenji Johjima, that does not automatically mean 90% of his OPS.
Also, who are we to say 85% of him, or 90%. Let’s not pull number’s out of our ***es just to prove our point. If you have a valid reason for saying 85 or 90% of Kenji, other than that it sounds right, let’s hear it.
Dave’s post has it right on, outside of a slight dip in HRFB (which is not outside of the realm of bad luck when HR are so few per AB anyways), I do not see one negative sustainable statistic from 2008 Kenji. Everything except HR/FB points to an improved hitter.
I think the best guess for 2009 Kenji is to think 2007 Kenji + 2 years older + 2 years better plate discipline. Also, he only played 112 games last year, he might be rested. I say 2009 Kenji is going to be every bit as valuable per game (in less games) as 2007 Kenji, only with a higher OBP and lower SLG.
Food for thought: the Tampa Bay expansion baseball team will reach the world series before the Seattle Mariners.
Last post on this, sorry for those sick of reading my junk. Basically leave out the whole last sentence of the 4th “paragraph”. I did not mean that .2% is the probability that Kenji’s BABIP skills did not regress, that is an incredibly wrong statement. I was in a hurry (as the grammar also shows). Basically .2% is the probability that any .307 BABIP hitter would get 79 or less hits on 339 balls in play.
Too late for me. . .
Woot! We’re going to be in the Terrible Threesome along with The Nats/Los Amours and the Walker, Texas Rangers.
I wonder who might be the last. . .