How to beat Herrera, Davis, Holland

Kelvin Herrera, Wade Davis and Greg Holland have combined for a 1.05 ERA in the postseason. Getty Images, AP Photo

As we've seen so far in the postseason, good luck beating the Kansas City Royals late in the game. Their late-game trio of Kelvin Herrera, Wade Davis and Greg Holland were dominant in the regular season -- all posted ERAs under 1.50, making them the first team since the 1907 Cubs with three pitchers who threw at least 50 innings with an ERA under 1.50 -- and they've been dominant in the postseason, allowing just three runs in 25.2 innings, a 1.05 ERA.

There is one big difference between the regular season and the postseason, however: With all the extra days off, Royals manager Ned Yost can use his big three more often. In going 8-0 in the playoffs, the Royals have played 80 innings; Herrera, Davis and Holland have combined to pitch 32 percent of Kansas City's innings. In the regular season, they combined for just 14 percent of Kansas City's innings.

That shows how the postseason is a different beast than the regular season, although that percentage is higher in part because the Royals haven't been behind late in postseason games, except in the wild-card game. Still, the Royals should be able to concentrate more innings into Herrera, Davis and Holland, and you have to think that if the Giants are going win the World Series, they'll have to beat these guys at least once.

So: How do you beat them? Let's take a look at each reliever and see whether we can find a weakness (even if it's small). All stats include the postseason.

Kelvin Herrera

vs. LHB: .246/.318/.299

vs. RHB: .181/.265/.221

Pitch selection: Fastball 75 percent, changeup 19 percent, curveball 6 percent

Key stat: Hasn't allowed a home run all season.

You'd think that a guy who hasn't a home run throws low in the zone, but that's not the case with Herrera, who pumps fastballs that touch 100 mph up in the zone. You can see from the heat map why his fastball is so tough to hit:

Herrera actually throws two kinds of fastballs: a four-seamer and a two-seamer with a little sink to it. About two-thirds of his fastballs are of the four-seam variety. Herrera's strikeout rate actually isn't that high for a reliever -- he's at 25 percent, while Davis and Holland are both above 35 percent -- but 51 percent of his balls in play are ground balls, even though he throws up in the zone.

How to attack Herrera? He almost always throws a fastball on the first pitch -- 89 percent of the time, saving his changeup and occasional curveball after he gets ahead in the count. Batters have hit .387 when putting the first pitch in play. Trouble is, opponents know to be aggressive against Herrera, as he has the highest first-pitch swing percentage of any pitcher on the Kansas City staff. It's easy to say "swing at the first pitch," but harder to execute when it's a 99 mph fastball up and in or up and away.

Herrera had a gopherball problem last year when he allowed nine home runs through July. It looks like that was primarily an issue of location: more fastballs down the middle. Compared to 2013, his fastball location has gone to the upper corner of the zone.

So the best bet is to be aggressive early in the count and hope Herrera leaves one of those heaters down in the zone. Lefties did fare a little better against him, which could help the Giants as six of their eight position players hit from the left side (or switch-hit).

Wade Davis

vs. LHB: .183/.270/.218

vs. RHB: .120/.182/.148

Pitch selection: Fastball 63 percent, curveball 18 percent, cutter 18 percent

Key stat: Has allowed just seven extra-base hits, including no home runs.

Acquired with James Shields in the Wil Myers trade with Tampa Bay, Davis had pitched well out of the bullpen for the Rays in 2012, struggled when moved into the starting rotation in 2013, but had one of the best relief seasons in recent memory after going back to the bullpen. His fastball velocity jumped from the low 90s as a starter to an average of 95.5 mph this year. His cutter has so much movement that our system actually classifies it as a slider. It's a true swing-and-miss pitch as opposed to most cutters, which tend to induce weaker contact but not strikeouts. Overall, batters hit just .121 against it, with no extra-base hits in 66 at-bats.

Since the Giants have so many left-handed hitters, here's how Davis attacks left-handers with his fastball:

Left-handers hit .217/.323/.253 against Davis' fastball -- so it's not like they suddenly turned into Barry Bonds against it. Still, they reached base nearly one-third of the time against the pitch. While opponents tended to be aggressive early in the count against Herrera, that wasn't the case with Davis; only 19 of his first pitches have been put in play all season -- and batters went 1-for-19. It appears the best chance of beating Davis is hoping his command is a little off and he either falls into a fastball count or issues a walk.

In the regular season, the Giants were very aggressive on first pitches, with the third-highest first-pitch swing percentage in the majors (of course, maybe that's slightly skewed by Pablo Sandoval). We've seen that approach work at times during the postseason, but we've seen other times -- including from Sandoval -- when they've shown more patience. Against Davis, it appears that's the approach: Get ahead in the count and try to avoid seeing the cutter and curveball.

Greg Holland

vs. LHB: .170/.253/.237

vs. RHB: .153/.221/.198

Pitch selection: Fastball 54 percent, slider 41 percent, splitter 3 percent, curveball 2 percent

Key stat: Has walked five batters in eight postseason innings.

Holland is your basic power fastball/power slider closer. His four-seam fastball averages 95.6 mph and touches 98-99. Against right-handers, he tends to throw it up in the zone, both in and away, while against left-handers he throws it away, but up and down the zone. It's a pitch without a lot of movement, but batters haven't exactly teed off on it.

His slider is his big two-strike wipeout pitch, as batters hit .133 against it with 72 strikeouts in 125 plate appearances ending in the pitch. What makes the slider so good is that it's just as effective against left-handers as it is right-handers (lefties hit just .106 against it, although with two of the three home runs Holland has allowed).

Holland hasn't had good control so far in the postseason with five walks, so the best approach is to be patient against him. He throws his fastball 75 percent of the time on the first pitch, so it will be interesting to see how he goes after Giants hitters, considering they do like to swing at that first pitch. But if he throws too many first-pitch sliders, he runs the risk of falling behind in the count and then his fastball becomes less effective -- he had 19 walks and 19 strikeouts in plate appearances ending with that pitch, so it's not a pitch he can just wind up and blow past hitters.

How to beat these guys? Obviously, nobody has done it much this season. The best hope appears to be a walk and a bloop, if you can somehow sneak in a hit against that Kansas City outfield. There's a reason these guys were one of the best bullpen trios we've ever seen.