Okay, let's talk about a relatively new development.

But it's also an old development.

So what's old is new again, or whatever the song says.

So shifting, which you've probably seen if you've been watching Major League Baseball

for the last couple of years.

So here's a picture of a shift.

I think this may have started with Lou Boudreau, the kid manager,

player manager, the Indians shifting on Ted Williams.

Ted Williams, as you know was a great hitter, maybe the greatest ever.

Left-handed hitter, he usually hit the ball to the right side of the in-field,

and so, if you know the guy, the hitter, is a full hitter, and they're left-handed.

They're going to usually hit the ball between first and second base,

why not put your fielders there?

In other words, you're more likely to stop a ground ball or

a line drive, and that's the principle behind the shift.

And I think it was used behind William McCubby, in maybe the 60's,

the early 70's, and then it sort of went into hibernation.

But now with the focus on analytics, the shift has come back with a vengeance.

And most of what I'm going to talk about comes from that great book,

The Fielding Bible, fourth edition 2015, by John Dewan and Ben Jedlovec.

I'm sorry I can't say that word.

But if you have any interest in analytics, even if it's not sports,

you should buy this book.

Because their approach to finding th right data and

the rigor in their analysis is just fantastic.

And we'll reference them when we talk about pitch rating with

catchers because I think they've done a really nice job with that.

Okay.

So basically, let's look at how a shift might benefit you.

So Adrian Gonzales, who's a full hitter.

If you look at the zones between first and second base, you divide them into six

equal zones where one is closest to first base and six is closest to second base.

And between second and third base divide them in 6 equal size zones and

7 is closest to second base and 12 is closest to third base.

You can see where Gonzalez hit his ground balls and soft liners.

Okay, so there's a big spike up in zone four and in zone eight.

Okay, so the question is where would you put your fielders?

Well, if you don't shift, you probably have the first baseman has to be zone one,

close to first base.

Third baseman would be in zone 12 close to third base.

The shortstop might be in zone eight or nine, I don't really know.

I mean, the second baseman zone four.

And so what I did with Excel is I figured out, basically,

how far each of the fielders is from that zone.

For instance, okay, if you look at zone two, the fielder in zone

one is one zone away, the fielder in zone four is two zones away.

Fielder in zone nine is seven zones away, etc.

And then I found how far the closest fielder was because the idea is

you probably can catch the ball if your closest fielder

is within one zone of the ball.

Okay, so with no shift here, given where Gonzalez hit it,

92 percent of the time we'd have a fielder within one zone of the ball.

And of course it was 0 zones if you're more likely to catch it, but

don't worry about it.

Just trying to give the idea.

Well what I figured out was, and

there's a way to show you how to do this with the Excel solver.

But I'm not going to do that now because don't think we're ready for

the solver yet.

But basically instead of going 1, 4, 9 and 12, which is what we had before

See that was with no shift.

We go one, four, seven, and ten.

There's always a fielder within one zone and

we don't care about that zone 12 because he never hit anything there.

So in other words every ball he hit we've go somebody within one zone.

So it's much more likely with this, this is a first baseman normal position.

You shift over the third baseman two zones, shift over the shortstop two zones,

and leave the second baseman in the right place.

Okay.

And basically you'd 100 percent of the time have somebody within one zone.

So you can see based on the hitting pattern of each hitter,

which is well known.

And you could shift your guys

to the left side of the infield if it's a right hand fold hitter.

Okay that's also fine and basically this seems to work.

Okay, so the good example, the Astros of 2014 were pretty bad but

they're within the American league since I taped this.

They did 1341 shifts, that's about eight a game and

saved 27 runs, which basically is about we know 3 wins by shifting.

You know, three wins could make the difference between making the playoffs and

not making the playoffs.

And in general does this shifting really

stop, reduce the number of runs a team scores?

Well again, we go to the fielding bible and

they have some very good evidence that the players that are most shifted against.

Their average dropped over 25 points.

Depends whether it's a left handed pitcher versus a left handed hitter, etc.

Which is a lot.

Okay.

Now there's a song.

You may know this song.

Something's lost and something's gained in living every day,

if you're as old as I am, that's Judy Collins.

Great song.

Of both sides now, because what happens is there's an upside and

downside to everything.

So you'll save runs on shifting, but what are the things that you give up?

We may not be as good against the bunt.

You may not be as good at catching foul balls,

because you have nobody to catch a foul ball near third base.

You may make less double plays.

So the fielding bible authors brilliantly studied this statement.

For the people who were shifted on the most they figured you

save 289 runs by catching the ground balls and the line drives.

You get 31 less fowl ball catches, 30 less run.

It costs you 31 runs on fowl ball outs you would not catch,

costs you 16 runs on doing a bad job on bunts.

But you actually did two runs better on double plays.

So even with the both sides now, the good and the bad in everything, okay,

the shift is really helping.

Okay, so I mean this is a good rigorous analysis and I think if teams

optimize their shifts, they're going to in general win a couple of extra games.

And maybe that will turn into a playoff, making the playoffs when you wouldn't

have or maybe winning the worlds series when you wouldn't have.

And the questions is should baseball outlaw this?

And that's again an ethical question which I don't feel qualified to comment on.

But basically, if you look at the individual pattern of each hitter like we

looked at Adrian Gonzales, you can figure out where to put those fielders.

So you would maximize the chances that you would catch a field successfully a ground

ball or field successfully a line drive.

Now, the one complication this shifting ushers in when we talked about ultimate

zone rating or the fielding plus/minus, it sort of makes it harder to do.

Because basically when the ball is hit in the zone right over second base

it rarely got fielded if you don't shift.

But if you shifted like this, it would be an easy play to the third basemen and

then it's not the shortstop.

And so it gets to be an issue you probably you need to do the fielding plus minus

separately when there's a shift and when there's not a shift.

And I'm not sure if people have address that issue yet.

But, I hope that was an illuminating introduction to the mathematics and

the rationale for the shift because it makes a lot of sense.

>> It really does.

For certain hitters, you have to look at that pattern and

see where they hit the ball.

In the next video, we'll discuss another recent development in honey bowl or

sports and baseball analytics, safer metrics, catcher pitch frame.

4.11 Defensive Shifts

Math behind Moneyball

Meet the Instructors