SF 3 TEX 1
Bullpen woes Friday night cost the Giants a win and made things uncomfortably close yesterday afternoon. But today the 'pen delivered a scoreless 2-2/3 in support of a strong start from Derek Holland (6-1/3, 3 H, 1 R) and that led to a series win for the home squad. Mark Melancon got the save, just like yesterday. Steven Duggar's two-out triple drove in two in the 4th, and Evan Longoria scored on a passed ball after a triple in the 1st and that was just enough to prevail. How often have the Giants scored in an inning in which they hit a triple? I'd venture not too often. The triple, a feature of PhoneCo Park, is one of those baseball appendages that has fan appeal but not much value in the age of the home run. According to B-R there have been almost seven times as many homers (4484) as triples (662) this season, and ten times (6685) as many doubles.
Chris Stratton is apparently the starter tomorrow night against the Diamondbacks. I'm guessing we'll see MadBum Tuesday and D-Rod Wednesday, but they may take advantage of the off-day on Thursday to to push the rookie (just off the DL) back to Friday. In that case they'll probably use Ty Blach. Go Giants!
--M.C.
10 comments:
what about Kelly instead of Blach? We need a few hitters to break out and have this team score some runs, while the pitching continues to be good.
I believe Kelly was optioned to AAA.
Why aren't triples valuable? If the Giants could bunt properly, they would be.
Triples aren't as correlated to run-scoring as you would think. They are such an odd event that perhaps it's just small-sample noise, but teams that score a lot of runs get on base and hit doubles and homers. How many triples a team hits is largely irrelevant in the big scheme. Not that they aren't exciting. (And if your offense has to rely on bunting to score runs, you are screwed.)
FanGraphs speculates on what teams might pursue Cutch. Cleveland, maybe?
Buster goes under the knife today. Thoughts and prayers.
Jay Jaffe of FanGraphs muses on Buster's career, recovery, etc.
If triples are not correlated to run scoring, I think what you are saying is that they are no sure thing to score, not, for example, twice as likely to score as a double (which one might think if you decide that a triple only has to move one base to score rather than two). Is that right? That it is not as correlated to runs makes sense, but that doesn't mean triples aren't valuable. What I need to know is the % of triples where the runner scores as opposed to doubles. I bet it's higher. There are of course many more doubles and home runs - they are much easier to hit, which makes the rarity of a triple more exciting.
Well, I was being a little facetious, but triples are sort of irrelevant mostly because of their rarity. If guys make it to third base with no outs they have a pretty good chance to score, that drops precipitously when there is one or two outs. <a href="http://www.tangotiger.net/re24.html>Check out Tom Tango's Run Expectancy Matrix.</a> The data is empirical. In other words, this is what happened over the course of the measured time frame. Obviously there are individual teams/games/contexts that are not included here, just the overall picture. The Giants have been a poor hitting team all year so I'd expect they probably had scoring numbers worse than what you'd expect from the data.
So how many runs, on average, does a leadoff triple (for example) produce? From 2010-2015, as you can see from the top chart, 1.350, and it drops to 0.950 with one out and to 0.353 with two. That's pretty good. Only way to beat that is to put two runners on. Another way to look at it is to use the second chart. There's an 84% chance a leadoff triple will score, 66% with one out, and 26% with two outs. But you get the same results or better if you put two guys on base. (Note how little improvement in your odds of scoring you get by bunting a runner that's on second base over to third base.)
Most of the time triples are just legged out doubles, or fluky park quirks, or homers that fell short!
Whoops. Here's the link.
Post a Comment