The 2 halves of Justin Verlander

After hitting the major leagues for good in 2006 and despite amassing 9.4 fWAR in his first three full years, many Detroit Tigers fans still wanted more, expected more, demanded more from their second overall pick in 2004, Justin Verlander. With an ERA and FIP north of four in this period, many started to wonder if the young flamethrower from Virginia would ever quite live up to the lofty expectations his 2004 pick provided. Coming off an 11-17 record and 4.84 record in 2008 things didn't look all that great. They need not have worried.

From 2009 to 2012, Verlander was a master of his craft, barely human, such was his otherworldly dominance of the man in the batters box.

Let's just remind ourselves how good he was over that four year stretch in case you'd forgotten (of course you haven't but these numbers are so sexy they need to be wheeled out once in a while).

Year	IP	ERA	FIP	WHIP	K/9	K%	fWAR
2009	240	3.45	2.80	1.18	10.09	27.4	7.7
2010	224.1	3.37	2.97	1.16	8.79	23.7	6.3
2011	251	2.40	2.99	0.92	8.96	25.8	6.4
2012	238.1	2.64	2.94	1.06	9.03	25	6.8

A quite ridiculous run of supreme pitching from a bonafide superstar at the peak of his powers. A man so in control of what he wanted to throw, where and when that at times you almost felt sorry for the opposing hitter. Almost. The cumulative stats over said period gave him these rankings within all of Major League Baseball starting pitching.

Wins	fWAR	WHIP	WPA	RE24	ERA	FIP	xFIP	K/9	K%
1st	1st	1st	1st	1st	7th	3rd	7th	6th	3rd

Not too shabby.

Then, seemingly out of nowhere, it all went a little bit wrong and we've never seen quite the same Verlander since...or have we?

The 2013 season, despite many people thinking it wasn't your typical Verlander year, was still actually a very strong one. Verlander racked up 4.9 WAR, though many were disappointed by his 3.58 ERA. It was 2014 where the struggles* for Justin began, culminating in the first half of 2015, and it's this I'm going to pick up on more strongly through this article.

I started by thinking, what was Justin doing differently during the down period compared to that aforementioned golden four year span from earlier? Sure, he's getting older -- aren't we all, Peter Pan aside -- and with that comes a natural decline curve, but not so rapidly and so sudden that your perennial All-Star morphs into some Triple-A lifer. There had to be more to it, didn't there? Besides injury and age, what was causing this?

Let's compare his numbers from both halves of the 2015 season.

*Granted, 90% of the rest of MLB would love a "struggling" year of just under three WAR.

Period	IP	ERA	FIP	WHIP	K/9	K%	HR/FB
First Half	30.1	5.34	5.80	1.38	5.34	13.5	13.6
Second Half	103	2.80	2.81	1.00	8.30	23.6	5.4

While I have split his production in half, his first "half" (if you can call it that) was actually only five starts and 30 1/3 innings, but the marked difference in performance means (a) it's still a valid time frame to use and (b) it's easier to work with for analytical purposes.

Immediately, some stark differences jump off the page. While the ERA, FIP and WHIP all take on the look of an ace starter again, what's more interesting is considering how he did it. The last three metrics are particularly revealing, striking out three more batters per nine innings, almost doubling his strikeout percentage, not to mention cutting his home run to fly ball ratio by more than half.

To do this with any pitcher, it's important to consider not only pitch type but of course location, where he throws it, how often, and what the results look like. Firstly I looked at his pitch usage percentages across halves and while there were some variations the disparities weren't large enough to prove anything either way, however getting deeper into the data revealed the major culprit.

Here's the breakdown of the results against each pitch per half.

Pitch Type	Split	BAA	SLG	ISO	BABIP
Fastball	1st Half	.292	.569	.277	.264
	2nd Half	.213	.340	.127	.268
Changeup	1st Half	.182	.273	.091	.200
	2nd Half	.317	.390	.073	.342
Slider	1st Half	.316	.316	.000	.353
	2nd Half	.178	.301	.123	.200
Curveball	First Half	.182	.318	.136	.231
	2nd Half	.219	.281	.063	.318

Only one of those pitches shares a similar BABIP with wildly differing results and that's the fastball, his most-used pitch, now, then and forever. While his first half slider numbers aren't too hot either, they are the result of a much smaller numbers of pitches and very heavily inflated by .353 BABIP. It's the gas which gave Justin fits in those five starts before the break, and it's the gas which contributed massively to his stellar performance after. Let me show you exactly why.

First half fastball heatmap:

I'm sure its quite obvious where the problem is but sometimes these can be hard to make sense of so I've circled the area in question. The majority of his fastballs were going in the last place any pitcher wants, middle of the plate and up, right in the happy zone of 99 percent of big league bats. They may as well have been on a tee. No wonder these pitches got crushed to the tune of a .569 slugging percentage and contributed hugely to pretty much the worst first half of Justin Verlander's career to date, but crucially, could he rectify this?

Second half fastball heatmap:

I'd say this tells you all you need to know. Of course he could, that's why he's Justin Verlander. Look where the highest concentration of these fastballs now went in the second half; mainly up and away, top right quadrant of the zone and far harder to do anything positive with, hence the league batting just a paltry .213 against it with an isolated power of less than half what he had been allowing pre All-Star break. It's also important to note these wildly differing results were achieved with almost identical BABIPs, eliminating any notion of first half bad luck and second half good luck.

As for the other pitches, it's hard to read much into them simply because of the large fluctuations in BABIP numbers, especially on the changeup and slider, contributing so much to the outcomes, and while there was a variation in the curveball results too, it wasn't anywhere near as drastic as the fastball or indeed as relevant due to far lower usage of the pitch compared to the fastball and the results remaining similar in both halves.

The spectacular improvement in fastball command in the second half of 2015 allowed Justin to put up some of his best metrics on that particular pitch for a number of years. The pitch value score (per FanGraphs) was at its highest since 2012, the contact percentage against it the lowest since 2009, and the league wOBA against it at .294 the lowest of his entire career.

What this has proven is that it doesn't matter that Verlander doesn't average 95-96 miles per hour on his fastball anymore, he can now sit at just above 93 and prosper just as much as he ever did, his batted ball numbers against it in the second half being up there with anything he's put up at any time since he entered the league.

Now we've seen the major reason he was able to rebound, let's take a moment to realise that his second half was so dominant he actually came out third in all of MLB starting pitchers during this timeframe in fWAR, above the likes of Cy Young candidates Zack Greinke, David Price and AL Cy Young winner Dallas Keuchel. The only two above him? NL Cy Young winner Jake Arrieta and some guy called Clayton Kershaw.

So in closing I return to the question I posed at the start of this article, that we'd not seen the likes of a peak 2009-12 Justin Verlander for a long time.

That was, of course, until second half of 2015 Justin Verlander showed up...

Year	IP	ERA	FIP	WHIP	K/9	K%	fWAR
2009	240	3.45	2.80	1.18	10.09	27.4	7.7
2010	224.1	3.37	2.97	1.16	8.79	23.7	6.3
2011	251	2.40	2.99	0.92	8.96	25.8	6.4
2012	238.1	2.64	2.94	1.06	9.03	25	6.8
2015 2nd Half	103	2.80	2.81	1.00	8.30	23.6	3.2

Welcome back, Justin.

Ed.: And welcome to UK Tiger, BYB's newest staff writer! He has the sunflower seeds and beef jerky in the pink Dora backpack, if you need any.