Is the shift effective?

[dominik]

Inspired by this thread

http://www.baseball-fever.com/showthread.php?109576-Brett-Lawrie-is-the-best-player-in-baseball-according-to-WAR

I think it's time to discuss the shift statistically. I think the teams wouldn't do it if it won't work but how much does the shift really reduced batters OPS (or wOBA or whatever hitting stat)? Is the shift proven to work?

with a shift you basically only defend center and right (maybe a little left-center)

http://www.google.de/imgres?um=1&hl=de&sa=N&biw=1231&bih=762&tbm=isch&tbnid=8uTGY6m1lJcIpM:&imgrefurl=http://en.wikipedia.org/wiki/Baseball_positioning&docid=SzCkZ9jLljCgrM&imgurl=http://upload.wikimedia.org/wikipedia/commons/thumb/a/a3/Baseballpositioning-shift.png/180px-Baseballpositioning-shift.png&w=180&h=158&ei=ozzsT6S0AZP44QSEg6iWBQ&zoom=1

Are there data on the shift?

[JDanger]

The Shift has been a popular topic lately in saber circles. The Fielding Bible 3 finds that the shift reduces the batting average on groundballs, short liners and bunts by the top-shifted hitters by 40 to 60 points.

[Ubiquitous]

The problem right now with looking at the shift is that the data needed to do it is in extremely small amount of hands and is being filtered to us through those hands. We don't have 100 different people looking at the data right now and approaching it differently than the next guy. So it is hard to come to any real conclusions on the shift because it is very hard to get all the needed questions answered.

[dominik]

The problem right now with looking at the shift is that the data needed to do it is in extremely small amount of hands and is being filtered to us through those hands. We don't have 100 different people looking at the data right now and approaching it differently than the next guy. So it is hard to come to any real conclusions on the shift because it is very hard to get all the needed questions answered.

yes I see that problem. but wouldn't be an extremely crude approach of comparing wOBA/OPS in shift situations vs no. shift situations already give you a pointer? of course OPS is not a very advanced metric since it doesn't acount for luck, range of fielders... (like ERA) but if lefties let's say hit 50 OPS points higher in no shift vs shift situations this would give a first hint (and would be very easy to do)

[Ubiquitous]

Yes it would but the problem is that we can't quickly and easily identify shift situations vs no shift situations. That is part of the data that only a few people have right now.

[Second Base Coach]

The numbers show the shift is effective if you are looking to reduce the probability of the batter getting a base hit. I am not sure there is enough to debate here. Facts are facts. The debate comes in when you begin to look for more batters to add to the list of players who are to face the shift.

I would think that list would be very short as extreme pull hitters are just not that common.

[Ubiquitous]

Facts are facts but we really don't have all the facts. We have some facts and we have some conclusions based on data we have not seen and that might not really be relevant to 2012.

[Second Base Coach]

I would quote myself, but that would be arrogant. LOL

The question still remains whether or not to employ the shift, and not if the shift lowers a player's anticipated batting average.

In layman's terms, the shift is a good idea if the batter is "predictable enough".

[dominik]

I would quote myself, but that would be arrogant. LOL

The question still remains whether or not to employ the shift, and not if the shift lowers a player's anticipated batting average.

In layman's terms, the shift is a good idea if the batter is "predictable enough".

funny is though that teams even shift the IF against guys who hit very well to opposite field (howard, Agone). I think this is because they feel that those guys often hit liners and flies to oppo field but if they hit on the ground they usually will pull it since the tendency when you are early is to top the ball and hit the ground.

thus against those players often a shift is used when the IF shifts but the OF plays straightaway or even a little oppo.

Yes it would but the problem is that we can't quickly and easily identify shift situations vs no shift situations. That is part of the data that only a few people have right now.

Who has those data?

[Ubiquitous]

Baseball Info Solutions, probably a bunch of MLB teams, STATS probably, and maybe Elias.

[Ubiquitous]

Dewan last month posted a list of the top 5 teams to save runs because of the shift. The #1 team was the Blue Jays and the 5th best team was I think the Indians at 7 runs saved. 4 of the 5 teams were in the AL East.

[leewileyfan]

On a lighter note, less than sabermetrically precise or profound as a secretly held database, I figured I'd kill time looking into how Ted Williams fared in a season I knew Boudreau threw the shift on him full time.

Like everything else baseball [and with insufficient data] it was a mixed bag in 1948. The BOSOX and the Tribe met 23 times, the Tribe taking 12 games and the Sox winning 11. In one streak, May 21 through July 30, the Sox actually went 9-2 with Ted going 20 for 40, a neat .500 clip. Overall, for 23 games, Ted was 33 for 86, .384 with 4 HRs and 6 doubles; but down the stretch, August 1 through October 4, when the pennant went past the wire between the clubs, Ted was 6 for 24 with 1 HR [having been HBP on July 25].

The Tribe took the flag; but Ted overall did well against the shift.

[leewileyfan]

FWIW, Lou Boudreau saw the shift more as psychological warfare than as a strategic method of run prevention. He was quoted as seeing TSW's combined skill and dedicated plate discipline as a possible way to get into his head and upset his timing and effectiveness.

As a positional ploy, the shift has some inherent weaknesses for opponents equally focysed on counter-strategies, like the bunt, bunt-and-run, and base stealing when players are out of position.

Williams himself believed that the shift cost him a total of 200 poits of batting average between 1946 and 1960; but in hindsight, the Korean War, injuries and the aging process may have combined for a portion of that discount.

[SHOELESSJOE3]

If I recall correctly, Cleveland with the shift on Ted, 1946 his home run, to left and not deep, inside the parker, pennant clincher, early Sept., looked like the Bosox really took off finished 12 games ahead.. Against Cleveland, 1-0 Bosox.
The only IPHR Ted had in his entire career.

[leewileyfan]

If I recall correctly, Cleveland with the shift on Ted, 1946 his home run, to left and not deep, inside the parker, pennant clincher, early Sept., looked like the Bosox really took off finished 12 games ahead.. Against Cleveland, 1-0 Bosox.
The only IPHR Ted had in his entire career.

Yep. I didn't refer to that one because it was the first year of the shift; and, sabermetrically speaking, that one ITPHR might be considered insufficient data.