Yankees Game 3 Game Plan: Pray
The New York Yankees find themselves in the unenviable position of trailing the Detroit Tigers two games to zero in the ALCS, having given away home-field advantage. The Yankees do have the consolation of not yet having lost with ace CC Sabathia, who could still make two starts. But then, the Yankees have the anti-consolation of not yet having lost to Justin Verlander, who could still make two starts. It’s Verlander who’s taking the hill Tuesday night, as the Tigers look to take a truly commanding lead in the best-of-seven.
The Yankees have struggled to hit in the playoffs so far, and they’ve struggled to hit against a bunch of pitchers who aren’t the best starting pitcher in the world. I don’t need to tell you that Justin Verlander is a little excellent. On top of that, he’ll be pitching Tuesday night at home, against a struggling lineup, on a cool October evening that should only depress offense even further. Tigers fans couldn’t possibly be happier with the way things are set up. Yankees fans, therefore, couldn’t possibly be less happy.
For any given game, you try to come up with a plan. Doesn’t really matter what sport you’re playing — if it’s organized, then you come up with a plan of attack, such that you might better be able to defeat your opponent. The Yankees, surely, have tried to work out a plan to defeat Justin Verlander, but I’m going to go ahead and tell you right now that no such plan really exists, not with any legitimacy. You don’t beat Justin Verlander by sticking to some kind of plan, unless your plan is simply “hit Justin Verlander.” That’s the real key, but that’s rather difficult to pull off.
During the ALDS, we heard a lot about how the A’s were trying to drive up Verlander’s pitch count in an effort to get into the Tigers’ bullpen. This is a typical plan when a team is going up against an awesome starting pitcher; do what you can to get rid of the awesome starting pitcher. In Game 1, Verlander threw 121 pitches over seven strong innings. In Game 5, Verlander threw 122 pitches over nine shutout innings. The A’s had a plan, but I wouldn’t say it worked, and it’s because it’s not actually a good plan. Especially under these circumstances.
This isn’t going to surprise you, but: this year, 186 starting pitchers threw at least 50 innings. The median was just about 95 pitches averaged per start. Justin Verlander had the highest average out of all of them, at 114.2. That was more than four pitches higher than the runner-up, James Shields. You can try to drive up Verlander’s pitch count, I suppose, but he might be more tolerant of a high pitch count than anybody, and he’s no stranger to reaching or exceeding 120. See, the thing everybody’s learned about Justin Verlander is that he knows how to pace himself.
This might be the telling graph, and you’ve probably seen something like this before. Here’s how Verlander paced himself with his fastball this past regular season.
In first innings, Verlander’s fastball averaged 94.1 miles per hour. Come the end, it shot up to 96 and 97, and what this ignores is Verlander’s variation pitch-to-pitch. At any given moment, he’s capable of throwing a fastball pretty much anywhere between 90-100. He always seems to have something left, and never visibly seems to wear down.
Research by Tangotiger and MGL and probably others has shown how starting pitchers get worse with subsequent trips through the lineup, due to fatigue, or batter familiarity, or something else, or plenty of things. This is one of the main arguments for replacing a starter with a reliever before he reaches the magical 100-pitch mark. I like to look at Verlander’s performance since 2009, since that’s when he ascended to ace-hood, and here’s how Verlander has done each time through the order since 2009:
| Time | wOBA | BABIP | K% | BB% | PA |
|---|---|---|---|---|---|
| 1st | 0.268 | 0.273 | 28% | 7% | 1247 |
| 2nd | 0.262 | 0.267 | 26% | 6% | 1204 |
| 3rd | 0.288 | 0.294 | 23% | 6% | 1103 |
| 4th | 0.259 | 0.316 | 25% | 6% | 278 |
You see a spike the third time through the order, but a lot of that is simply due to BABIP increase, and we don’t know how much to read into that. Verlander’s been nails the fourth time through the order, in limited opportunities. Taken at face value, the first two times through the order, Verlander has turned opponents into 2006 Cesar Izturis. The third time through the order, he’s turned opponents into 2008 Cesar Izturis. The fourth time through the order, he’s turned opponents into 2012 Cesar Izturis. A lineup full of various Cesar Izturises wouldn’t score a lot of runs. Also, fun fact: Cesar Izturis played major-league baseball in 2012.
Verlander doesn’t show any pattern of wearing down, and in fact he seems to do the opposite as he’s learned to pace himself with experience. He doesn’t have much in the way of an exploitable platoon split, owing to his four dominant pitches. And while you can try to knock him out of the game by elevating his pitch count by taking pitches and fouling pitches off, that theoretical approach falls apart quickly. For one thing, batters have shown little ability to foul pitches off on purpose, and for another, it does little good to take pitches when those pitches are strikes, which Verlander usually throws. Against a pitcher with good control and command, the key to driving up a pitch count isn’t patience, but generating base-runners. Against Verlander, you need to get base-runners.
Hell, even if you have a bunch of longer plate appearances that result in outs, those are lower-stress pitches, and therefore less taxing pitches. Just taking pitches against Verlander isn’t going to accomplish much of anything. You can take the balls, but then you should always take the balls, and that isn’t as easy a thing to do as it is a sentence to type.
If anything, if anything, the Yankees might want to be aggressive. According to Brooks Baseball, Verlander throws 69-percent first-pitch fastballs to lefties, and 75-percent first-pitch fastballs to righties. Once Verlander gets ahead, he’s lethal, because he balances all of his pitches. The Yankees could key in on first-pitch swings. But Verlander might well adjust to that, and additionally, not all of his first-pitch fastballs are the same, since he varies his velocity. Knowing that Verlander throws 75-perfect first-pitch fastballs to righties seems exploitable; it seems less exploitable when you realize those fastballs fall within a broad range.
The key is basically prayer. The Yankees are left praying that Verlander has a rare bad game, their hitters generate some lucky balls in play, or Phil Hughes is absolutely brilliant. If Verlander posts zeroes, but Hughes also posts zeroes, then Verlander is effectively negated. But Hughes gives up a lot of fly balls, and a lot of fly balls turn into home runs, so that could go out the window in a jiffy. I will note that, on six occasions this year, Verlander allowed at least five runs. Additionally, over four appearances in last year’s playoffs, he allowed 12 runs in 20.1 innings. Those opponents were left having to pray, and their prayers were answered. October Justin Verlander was mortal.
But you can’t really count on Verlander’s mortality. All pitchers are mortal, but no starting pitcher is less mortal than Justin Verlander. The Yankees’ odds of winning Game 3 Tuesday night are higher than you might think they would be. But, because of Verlander, they’re a lot lower than the Yankees would like them to be, and we’re one dominant outing away from the Tigers being on the verge of an ALCS sweep.
Let’s be honest, the Yankees have zero chance of winning tonight.
Let’s be honest, there is zero chance this isn’t hyperbole.
didn’t the yankees beat verlander 2 of 3 times this year?
This is baseball. The Astros would have a >0% chance of beating the Tigers.
Bravo
Swisher and A-Rod out, Nunez and Gardner in. Apparently the plan is to steal a run and have Hughes pitch a shutout.
I guess I think it’d be more likely that A-Rod or Swish hits a 3-run homerun and Hughes gives up 2….
Nunez is at SS, Chavy’s the one replacing A-Rod.
It’s pretty hard to steal first…
Rickey says that’s a clown statement Bro
According to fangraphs, Verlander’s fastball has been the least effective pitch for him this year. According to B-R, batters have an .862 OPS against him when swinging at the first pitch. It seems to me that if he’ll give them a pitch to hit on the first pitch, they’d better hit it b/c he’s just deadly when he gets ahead in the count. I know…easier said than done.
Or they could just pray or use Nunez and Gardner to try and steal first.
Got a great laugh out of the title to this article. Classic. I hate the Yankees and would love nothing more than to see JV mow them down tonight. That said, Jeff is right. The Yankees have a better chance than most realize. Chad Ogea beat Kevin Brown at his apex twice in the 97′. Think about how many times the 90′s Braves played a playoff series where they had a huge advantage at starter for every game of the series and didn’t win. Or in 2001 when Miguel Batista basically matched Unit and Schilling, his HOF rotation mates, in the WS. Or the time Kenny Rogers decided to turn into Steve Carlton once the 2006 playoffs started. It’s one of the many reasons we love this game.
I think the best chance they have to score is simply to hit a homer or 2. The way this team is striking out, they seem to little chance of actually stringing together a few hits/walks. Verlander was sandwiched between Aaron Harang and Trevor Cahill in HR/9 this season, so not like he’s immortal on that front.
Obviously Girardi doesn’t see it that way, giving Gardner the start over Swish.
This.
1.77 HR/9 in 20.2 IP against the Yankees, along with a .360 BABIP.
At home he had 1 masterful start in August, one not so special (but not terrible) in June in which he gave up 2 HRs.
There’s also the fact that Verlander becomes better as runners get on base. What’s harder than hitting .200 against verlander when the bases are empty? hitting .170 when you get them into scoring position.
gardner has a better chance of getting on base by standing there and taking pitches than any of swisher, granderson, cano, or a-rod seem to at the moment
Just a slight objection to the argument about each time through the order. It seems that any pitcher will have really good stats the fourth time through the order because if they have made it that far, they are having a very good game, right? I mean a bad start will rarely result in facing 30+ batters. In other words, bad starts will hurt the first few trips through the order but not the fourth.
Of course, the fact that Verlander is so great against the lineup regardless of times he’s seen them remains true, but I don’t know that it’s “turning it on” the 4th time through or if it’s a matter of only facing them a 4th time when he’s already “on.”
My thoughts exactly.
Perhaps. But how much do we believe in the idea of being “on” in a single game in the first place? Perhaps many bad starts are simply a case of a pitcher being unlucky early on.
Still, it would be interesting to see his average fastball velocity in the early innings only in starts where he went 8+, and then compare to the late inning velocities, I agree.
Except that no other starter can put up a table like that.
It’s the other way around. If you pitch a perfect game, you only get three times through the order, the more hits/walks you give up, the more times through the order you go.
i think back to verlander’s start in the ALCS last year against the tigers, where nelson cruz homered off that 100 MPH fastball on the hundred-and-umpteenth pitch, and i gotta figure that’s all the yankees can really hope to do
Isn’t a lot of this selection bias?
First time through the order =100% of Verlander starts, good or bad.
2nd time through the order =almost 100% of Verlander starts, excluding any starts in which he’s terrible, but including bad starts.
3rd time through the order= only good starts with some brilliant starts
4th time through the order =only brilliant starts
Verlander would be a bit of an exception seeing that he hasn’t had a start that hasn’t gone 100 pitches this season and last year was much of the same in that regard.
I agree that the 4th time through the order category is likely to skew in his direction because he’s doing well but I wouldn’t say 3rd time through the order necessarily automatically means he’s performing above his average.
It’s arguably less selection bias than many stats based on the other possibility (which would, I agree, be interesting to look at) of only taking the fastball velocities from earlier innings in starts where he went deep. (In this case I would like to look at the velocity in those cases.)
After all, in those cases, you *know* that the earlier times through the order were good, or else he wouldn’t be in the game.
It’s only selection bias if the “good start” is a real thing, instead of luck being a strong factor. It’s possible that his performance in different innings is largely independent, and what looks like a “bad start” is really just bad luck but he had nearly the same stuff.
While I do, to some degree agree that luck has a decent amount to do with “bad” starts but I think that anyone who has played a sport knows there are days when you’re just “off” if that makes sense. There are days where, for whatever reason, your curve just doesn’t curve as much or other things of the sort. In those cases, he’s probably out before the fourth time even if he is Justin Verlander.
I know he always throws 100 pitches but if it’s a struggle to get through the order 3 times that might take nearly 100 and he would face hardly any of the lineup a fourth time.
In terms of *knowing* he’s on early if he gets to the fourth time through the order, this seems to contradict your other point. Either “being on” exists and he is and thus early numbers reflect that or it doesn’t and the numbers are driven by luck. However, if that is the case then you have to take into account the “off” days as well and those would be much more heavily weighted in the early stats then the late ones.
In terms of fastball velocity, which we’re looking at here, you’re right. But, as I said, in many other stats based on looking at early versus late inning appearances, the two effects are different. If you’re just looking at results rather than stuff, then it’s very possible that the “good start” is a myth and yet you’ll guarantee that if you look only at a appearances when a pitcher went into the 8th or 9th that the earlier innings had good results for the pitcher.
See above. If other good starting pitchers were doing the same you could make that argument. You could somehow correlate quality with number of times through the lineup but then you would have to show that those really good pitchers were all doing this. I don’t think anybody is.
They say no plan survives first contact with the enemy, but the Yankees may have difficulty making any contact at all.
Looking at that graph of fastball velocity per inning leads me to draw the following stat-oriented conclusion…
Justin Verlander is just sick.
That’s probably the most scientifically accurate statement ever uttered
That graph in the article is obscene, NSFW and should be rated, at the very least, R. I can’t define pornography, but I know it when I see it and it is right here, midway down this post.
It’s just not fair and should be illegal.
What he does to opposing hitters was only made legal in some states by the Supreme Court’s Lawrence vs. Texas decision
Well, that didn’t work. What else ya got?
This post was brought to you by the movie “Looper.”
At least they won’t have to worry about facing Jose Valverde- he is pregnant!
http://www.youtube.com/watch?v=lNrds8OBYHM
In hindsight, you nailed it.