Let's start with a basic premise: coming into 2013, Adam Lind had been pretty awful throughout his career against left-handed pitching. For his career, he is a .227/.269/.351 hitter, with a 5% walk rate and 25% strikeout rate (that includes this year, which is a small part and not worth separating out; and actually bumps it up a bit). In total, a 62 wRC+, which is well, well below average. That compared to a 123 wRC+ against RHP (about 120 coming into 2013), which is above average and has value. In addition, the only season where he did hit LHP (2009, 100 wRC+), he was a monster against RHP, and still had a huge split.

For that reason, I've long been a strong proponent for strictly platooning Lind. It tormented me when Farrell wouldn't do it, but I was very hopeful when Gibbons took over, since he had a history of successful platooning (Reed Johnson and Frank Catalanotto being the best examples). And despite some suggestions to the contrary before the winter, that's exactly what Gibbons did in the early going. And it worked. Though Lind struggled in the early going production wise, he was putting together quality at-bats and it was basically a case of a low BABIP. And in time the hits came...and they came...and they came...

So with Lind essentially carrying the offence, and the Jays suffering injuries and carrying a short bench into interleague road play, Lind started playing against lefties. And surprisingly (to me at least), he hit them and had a decent approach (I thought in most of his few at-bats against LHP earlier in the season he looked pretty bad). Through June 26, Lind sports a 133 wRC+ against LHP. So maybe he's not so terrible against lefties anymore?

In addition, there's a few other data points. In mid-May last year, Lind was optioned to AAA, and ended up there for about 40 days before being recalled. Apparently, he did quite a bit of work with Chad Mottola, and when he returned he was productive not only against RHP (.311/.354/.492) but decent against LHP (.268/.312/.437). Not great, but playable. Also, in Dec 2011, Jason Collette wrote an article at Baseball Prospectus, which showed that Adam Lind's struggled particularly with lefty relievers, and was significantly better against lefty starters. In other words, he was really susceptible to LOOGYs like Randy Choate and Mark Hendrickson.

All this together, I began to wonder if I wasn't just being dogmatic about Lind and LHP. Is it possible he's changed his approach in the last year working with Mottola, such that he's much better against them (or much better overall, such that even with a split, he's playable, as in 2009)? Or, alternative, he's actually decent enough against LHP, or at the least, a subset of LHP starters? I had planned on looking at this a couple weeks ago, but I thought that data collection would be a mess. I forgot Baseball-Reference has added "vs. Pitchers" splits, which makes it pretty easy.

So let's start by looking deeper at Lind's performance v. LHP, focusing on peripherals to isolate different skills and tools, over three difference time periods:

One quick note about the "power" column. Most use ISO (SLG - BA), I use bases/hit (SLG/BA), since ISO is biased by BA and doesn't just measure power. Career, Lind has a below average 5% BB rate, and above average 25% K rate (MLB average for 2006-13 roughly 7-8% and 18-20% respectively). He averages 1.55 bases per hit, which is below the average of about 1.65. Moreover, his BABIP is below average at .281, despite an above average 21% LD rate. Since it's only 795 PA, it's possible the BABIP is poor fortune, it's also possible he's just not making good contact. So basically, on all 4 metrics, he's below average, though one is subject to SSS issues.

In 2013, he's only had 40 PA. That's really too small to say much, other than his success is driven by a BABIP of .448 on the back of a 33% LD rate. Is that sustainable? Certainly not at that level - but it's possible that it's indicative of improvement. So, let's add the 77 PA against LHP in from after his recall, and that gives a bigger, though still quite small sample to evaluate.

In the last year v. LHP, his walk rate is roughly in line with his career rate, his strikeout rate is down by 5 percentage points, the BABIP is solid, and the power in line (I don't have the LD/GB/FB data for 2012 after the recall). So maybe the new Lind is more successful? One thing to remember is that so far this year, he's been generally shielded from the best LHP, so all else equal we should expect that would help his numbers. Let's look at the same data against RHP for signs of improvement that might confirm the LHP trends:

Basically, his walk rate is higher in the last year (mostly 2013, not so much 2012), his K rate is in line with his career numbers, his BABIP has increased in both periods (though the batted ball rates for 2013 are still in line, which suggests some good fortune, though it also some sharper contact). His power slumped last year, but this year is in line.

So what does should we expect from Lind going forward against LHP? Let's be generous and give him a bump on his career BB% to 6%, based on improvement against RHP. Let's also assume the K rate decline is at at least somewhat real, and plug in 22%. Let's give him a .300 BABIP going forward, and his normal power of 1.55 bases per hit. Putting that together, that works out to roughly a .257/.302/.399 slash line, roughly 700 OPS. Which is just not good. To put it in perspective, Rajai Davis is a career .290/.348/.415 hitter against LHP.

But that's not the end of the inquiry. Through June 26, Lind has 795 career PA against 155 LHP. About a week ago, there was a post suggesting that Lind was abysmal against 11 specific LHP, but "[a]gainst all the others about 85-90%, he hits .271/.309/.431 ... actually respectable". And so, the conclusion was, play him against most LHP. Now, I generally hate arguments like this, where an arbitrary number of PA are removed, and then the rest looks okay (or bad, as the case may be). In this case, the 11 pitchers are only 7% of the total, but total 200 of his 795 PA, of 25%. I'm sure if we removed the pitchers who made up every batter's 25% worst PA, then the average would end up so much higher that a 740 OPS would still look pretty ugly.

But that's not even the main issue. Pitcher/batter matchup data is almost inherently the result of small sample sizes. In Lind's case, his 39 PA against David Price is the most against a LHP in his career. As a result, they have very little predictive value. Especially in this case, where there's very little that links pitchers like Mark Buehrle, Jon Lester, and Randy Choate together.

But going back to my initial inquiry, what if there types of pitchers against whom Lind really struggles, and other that he hits okay? Then there might be a legitimate reason to hit him against at least some lefties. So I took all 155 pitchers, and pulled their results from 2006-2013 in terms of FIP (measuring process) and wOBA allowed (measuring results) against LHP. I sorted them according to both metrics, best to worst, and then grouped them into quartiles, and looked at Lind's results against each quartile:

When sorted by FIP, against the first 3 quartiles covering pitchers with a 4.00 FIP or lower v. lefty batters, Lind is almost uniformly awful. A league average wOBA is roughly .330, and he well, well below that. Curiously, he's a little better against the first quartile, the toughest pitchers - which is almost certainly just noise. But against the worst quartile of LHP, he's much, much better, with a wOBA of .347 which is actually decent and suggestive. But here's the kicker - these are the worst of the worst, and he's really just above average. A lefty masher, or even normal everyday player, would likely have much better numbers than Lind against this quality of LHP.

And indeed, when sorting by pitcher wOBA, we don't see the same trend. In fact Lind is remarkably uniform (and uniformly bad) against the 4 quartiles, with a small spike against the 3rd quartile. Pretty bad.

But we can drill down further. There are 56 pitchers Lind has faced at least 5 times, which make 72% of his career PA. Overall, he's a little worse against these pitchers, .216/.252/.327, which makes sense since better players get more playing time. For these 56, I pulled their numbers against RHP as well. I then classified them according to whether they were primarily SP, RP or other (bounce back and forth). For the SP, I then divided them into frontline SP (group 1) and mid-rotation and backend guys (group 2). I also calculated their platoon splits according to an average of FIP and wOBA, grouped them into small/reverse, medium and large platoon splits. For each of these classification, I looked at how Lind's production broke down:

In terms of the role of the opposing pitchers, on average SP are quite rough on him. But that's nothing compared to relievers, who absolutely eat him alive. He hits the "others" best (though not well), which follows since they tend to be journeyman types.

When the SP are grouped according to their quality, unsurprising he does worse against the frontline starters compared to the backend types. He's awful against both.

Let's look at grouping pitchers by their platoon splits. The idea here is to use that as a proxy for the type of pitcher. Pitchers with small or reverse splits often have 3 quality pitchers, and/or a very good change-up. Pitchers with larger splits usually make their bread and butter off dynamite breaking balls. Once again, Lind is well below average against all three categories, struggling in particular with pitchers who tend to be lefty specialists due to large splits.

We can drill down a little more. Let's combine role and platoon splits:

That's the chart on the left, and some of the result are illuminating. He struggles mightily, again, against all the SP. There's just really no type of SP that he can hit. At the bottom, are RP with large splits. Otherwise known as LOOGYs. And Lind is just brutal against them, basically more like a rookie ball hitter than than a major league hitter. He actually hits the other RP well, but he rarely faces those - likely only in garbage time. In important situations late in the game, you just LOOGY him to death.

On the right, I've combined all three splits for SP. That's really the realm of totally small samples, but it's also consistent with everything else.

The conclusion is pretty clear: as good as Lind's been this year, and well as he's done against LHP so far this year, there's no reason to think he's improved enough to be a good option going forward. When it comes to the idea of playing Adam Lind against lefties, the answer is to listen to Nancy Reagan and Just Say No.

Thanks to Fangraphs and Baseball-Reference for the data used above.