Logan Watkins was good and rested for today's game after getting yesterday off. He hit his 4th home run to lead off the game, and went 2-4 with 2 RBI. Brett Jackson 2: Aftersucks went 1-3 with a walk and a strikeout. Dave Sappelt was 2-3 with a walk and a run. Josh Vitters had a nice day also, 2-3 with a walk. Ian Stewart, pinch-hitter emeritus, was 1-1 in his one plate appearance. In his last 10 appearances, he's .316/.409/.526. For fun, Chris Rusin also had a double. Our pitchers are probably better hitters than our hitters are. DH go home!
Great day for the Cubs' pitching staff. Chris Rusin allowed no runs in 6 innings. He stranded 6 hits and 2 walks, with 4 strikeouts. Yoanner Negrin is just biding his time until he gets the callup to the majors. He had 2 innings of scoreless ball. Eduardo Sanchez has a scoreless ninth.
Huntsville Stars 0 @ Tennessee Smokies 2
Ronald Torreyes was 0-4 today. The slack was picked up by his fellow MINF, Arismendy Alcantara was 1-2 with a home run and 2 walks. Rubi Silva and Christian Villanueva each were 0-4 with 2 strikeouts. Everyone else was bleh.
Kyle Hendricks had another great start. He allowed 4 baserunners in 7 innings, with no run scored. Even better, he struck out 7. Brian Schlitter has an ERA of 0.90. How tight is that? He had a dicey, scoreless 8th. Marcus Hatley had his second save in the ninth.
Daytona Cubs 4 @ Jupiter Hammerheads 8
Another day off for Javier Baez. John Andreoli was the star of the game, going 3-5 with a triple and 2 runs. Wes Darvill had 2 walks and a double. Jorge Soler had a double and a walk. He's so dreamy…
Ben Carhart also had a double.
Zach Cates had a bad start. In 5 innings, he allowed 4 runs. Jeffrey Lorick walked 3 people in one inning. That didn't work out that well. Sheldon McDonald had 2 innings of 3 run work. Good job guys!
Kane County Cougars 8 @ Peoria Chiefs 9 (14 innings)
The Cougars had 53 at bats in this game. Albert Almora went 4-7 with 2 doubles and an RBI. Dan Vogelbach went 1-6 with 2 strikeouts and a walk. Jeimer Candelario was 2-7 with a double and 4 RBIs. Good job, Jeimer. Worst day in the organization was maybe Rock Shoulders. He went 0-6 with a walk and 5 SO. In fact, 3-9 each had at least 2 strikeouts.
Dillon Maples had the start and was ROCKED. He allowed 8 runs in 3.2 innings (5 earned due to his own throwing error). Lendy Castillo relieved Maples and went 5.1 scoreless innings with 9 strikeouts. Why can't Maples be more like Castillo? Jeffry Antigua threw 2 perfect innings. Justin Amlung eventually surrendered the game winning run in the 14th after 2.1 innings of work.
Humberto Garcia committed 2 errors today, one from just dropping an infield fly.
Comments
Wow @ http://www.fangraphs.com/blogs/miguel-cabreras-ridiculousplate-coverage/
Rice CubeQuote Reply
Myles wrote:
I don’t understand why everyone is so insistent about this. If you have evidence that player X (or a group of players with certain characteristics), given seven previous games of bad performance (or five or 10 or whatever), predictably hits 200 OPS points lower in game 8, then you pinch hit for him in high leverage situations. If you can’t define those types of criteria, then you really don’t know he is streaky.
GWQuote Reply
Are there any pitchers in the upcoming draft that have elite control/command tools? I remember one of the things that was most exciting about Hendricks was his elite command.
EdwinQuote Reply
And by the way, the problem with just looking at week to week variance in production (or whatever your preferred method) over a period of years to identify streaky players, as was suggested in the last thread, is that even if there is absolutely no difference in true streakiness in players, you will expect a certain fraction of players to have streaky performance over any given time period just based on random chance. The past is riddled with artifacts. In order to “prove” streakiness exists, you have to isolate streaky players and test your hypothesis on a future dataset (or split half, or whatever).
GWQuote Reply
Jim Callis’s latest mock has Appel to the Cubs.
GWQuote Reply
Also says that the rumors have the Astros negotiating with one of the hitters, adopting similar strategy to last season.
GWQuote Reply
GW wrote:
You answer your own question with this:
GW wrote:
MylesQuote Reply
@ GW:
@ GW:
That suggests that Cubs will end up with their pick of Appel/Gray (probably Appel) with the other going to Colorado.
Rice CubeQuote Reply
@ Myles:
ugh. i’m a terrible writer.
if you know a player is streaky, then you can act on that data. this was the original hypothetical, and the basis of my response.
and it can be known. this is my position. but just looking at variance over a certain period isn’t going to cut it. at minimum, one must show that a population diverges from a predicted distribution.
does anyone know at this point? maybe. i’ve never seen a convincing writeup, but it seems plausible enough to me.
GWQuote Reply
and by the way, “The Book” did find streakiness for pitchers, so how does the “we can’t know” argument hold any water?
GWQuote Reply
Myles wrote:
That’s either some funny clipboard fail or a great example for Mucker of how far some people will go for the snark.
Suburban kidQuote Reply
i should also mention that the hot hand has been investigated ad nauseum in hoops, and if it exists, the effect seems to be very small.
GWQuote Reply
@ GW:
I don’t think streakiness can be enough of a known quantiy to act on reliably. Even if you know that a player is streaky, the randomness of how long a streak can last, and when a streak can start/end would make it hard to make decisions.
I also don’t see how you would separate out a “streaky” player from a player experiancing normal stat distribution. Even if you knew that Player A is demonstratably streaky, I think by the time you get a large enough data set to be confident that Player A is actually on a Hot/Cold streak instead of just random stat distribution, it might be too late to make a meaningful decision anyways. Ultimately, you’re still using a small sample of recent performance to try and predict future performance, and I just don’t hink that would ever work.
EdwinQuote Reply
@ Rice Cube:
I still think the ‘stros end up taking a pitcher, but in the event that they don’t, it would be interesting to see which way the Cubs go, given that the two pitchers are the prospects that we layman know most about around these parts.
GWQuote Reply
@ Edwin:
This.
BerseliusQuote Reply
@ Edwin:
If you found that after seven games of poor performance, high K%, high power guys tend to hit .200 OPS points lower then their projection on day 8, then that is absolutely enough evidence to act on. If your argument is that the real level of streakiness, to the extent that it exists, is small enough that it won’t make any difference, then that’s a perfectly valid position to hold, and you could very well be correct. But again, the premise was that we know Alfonso Soriano is streaky. My position is that we don’t, but it’s possible that he is.
GWQuote Reply
http://sports.yahoo.com/blogs/mlb-big-league-stew/cincinnati-baseball-team-employing-next-level-photobomb-techniques-151835630.html
GWQuote Reply
[gw]
if you know a player is streaky, then you can act on that data. this was the original hypothetical, and the basis of my response.
and it can be known. this is my position. but just looking at variance over a certain period isn’t going to cut it. at minimum, one must show that a population diverges from a predicted distribution.
does anyone know at this point? maybe. i’ve never seen a convincing writeup, but it seems plausible enough to me.
[/gw]
I do agree that it’s theoretically possible to use variance to your advantage if you knew it was coming. You keep using an example of 7 consecutive games of poor performance. Can’t argue with you there.
The problem is that it can’t really be known. It’s been shown that we can’t really know it. You keep using this example of seven games of poor performance, high K%, high power guys hitting lower in the 8th game. Does this really happen (I’d bet it doesn’t)? What constitutes a “poor performance?” Are these players facing tougher pitchers? What parks are they hitting in? Is 7 games a large enough sample to actually define streakiness (I’d bet it’s not)? Are we SURE this isn’t random noise?
Past events have no bearing on future events. If streakiness does exist (and I haven’t read about pitchers having it but I haven’t read the Book yet, to my eternal shame), I’d bet it is so miniscule to the point of being unactionable.
MylesQuote Reply
GWQuote Reply
Myles wrote:
no, it hasn’t. this, I would argue, is something actually can’t be shown. What can be shown is that is the effect is smaller than is generally believed and difficult to tease out.
GWQuote Reply
GWQuote Reply
@ Myles:
my high k%, high power example, was just a prod in the direction you would want to go if you wanted to show that adolfo is streaky. nothing more.
GWQuote Reply
It’s not just about knowing whether a player is streaky, it’s about knowing how long of a streak a player typically goes on, and whether or not the past X games are a strong enough indication that the player is currently streaking, or just going through normal distribution.
EdwinQuote Reply
@ Edwin:
of course
GWQuote Reply
Suburban kidQuote Reply
http://sports.yahoo.com/blogs/mlb-big-league-stew/those-red-pens-ready-royals-television-graphic-contains-142559444.html
Spot the errors!
Rice CubeQuote Reply
@ Suburban kid:
What a jerk.
EdwinQuote Reply
Cabrera HRs
BerseliusQuote Reply
Myles wrote:
btw, I should mention that this is only true for independent trials, which is precisely the issue in question here.
GWQuote Reply
http://www.pewresearch.org/quiz/science-knowledge/
GWQuote Reply
@ GW:
Gambler’s fallacy, right? Just because Red has been hit 19 times in a row doesn’t mean the 20th time there is a higher chance of it being Black (although this is how a gamble sometimes :))
WaLiQuote Reply
GW wrote:
Watch each card you play and play it slow.
WaLiQuote Reply
@ WaLi:
a roulette wheel is an independent trial. (except when it isn’t, iirc there is a famous and perhaps apocryphal story about blaise pascal sitting and watching a certain wheel for days on end, determining it wasn’t random and then cleaning up over the course of the next few months)
GWQuote Reply
@ GW:
I missed the one about which gas compromises the most of our atmosphere, and I feel like an idiot for missing that. I got the rest.
MishQuote Reply
@ GW:
Wow, there are a lot of idiots out there
>40% of college grads say an electron is bigger than an atom. Also, this poll shows that the Grateful Dead song “Man Smart, Woman Smarter” is wrong.
WaLiQuote Reply
@ GW:
I thought you were saying that each batting attempt is an independent trial?
@ Mish:
You should skip dinner for your mistake.
WaLiQuote Reply
@ Mish:
the results on that one really surprised me. maybe it’s because of all the hype around global warming? dunno.
GWQuote Reply
WaLi wrote:
it’s definitely a very useful approximation. but i doubt that it’s completely true.
GWQuote Reply
@ GW:
I think more that you don’t learn shit about Nitrogen in general science courses. Anything involving the atmosphere is always about the oxygen-co2 cycle.
BerseliusQuote Reply
@ Berselius:
do most people learn about fracking in general science? (asking seriously, I don’t know) More than twice as many people got that one right. The nitrogen success rate is worse than random (or at least it was when I took it).
GWQuote Reply
@ GW:
I have no idea, I just know that from the news.
The nitrogen question had around 20% correct when I just took it.
BerseliusQuote Reply
The biggest surprise to me was that only 58% got the global warming question right.
BerseliusQuote Reply
especially considering the other options were
hydrogen
helium
radon
if one of them was something like ‘ozone’ I could see people going for that, but wtf.
BerseliusQuote Reply
BerseliusQuote Reply
@ Berselius:
that’s a good point. if everyone had CO2 fresh in their minds you might expect higher success rate on that one.
@ Berselius:
i think they tried to make it unambiguous given that ozone and water vapor are actually greenhouse gases.
GWQuote Reply
@ GW:
maybe the key to teaching science is giving all concepts sexually-suggestive names. more excess entropy, if you will.
/bonus points if you get that super-nerdy joke
GWQuote Reply
Berselius wrote:
The biggest surprise to me is definitely the Atom one which had about the same % right from college grads. Overall the % right was only 47%!
Are people still following Dalton’s model of the atom from the early 1800’s? The freaking electron was discovered in 1897. I guess that isn’t recent enough for some folks.
WaLiQuote Reply
Results:
http://www.people-press.org/2013/04/22/publics-knowledge-of-science-and-technology/
WaLiQuote Reply
So this one time, I was speeding on these back country roads, and this cop pulls me over. He gets out of is car, comes up to me, and asks “Do you know how fast you were going?” and I answer “No, but I can tell you exactly where I am.”
EdwinQuote Reply
Returning to what I think 424 was asking, the answer is yes, consistent is better than streaking even if the production is the same. This has been shown at the team level. Teams who are more consistent, which are usually low HR, low strikeout teams, tend to get more bang for their buck. It’s important to note that these offenses aren’t similar. The one that is more consistent is not as good.
What is a streaky player? I don’t have any idea. Is Soriano streaky? I don’t know. We’d have to compare him to similar players to find out. Is he more streaky than Darwin Barney? That would be relatively easy to test, but Barney’s not as good a hitter. Soriano is going to look a lot worse at times because he has been a good hitter. Those hitters around replacement level don’t have far to fall and you don’t notice it when they do. My guess is that Soriano is no more streaky than similar players because if he was, one of the teams he was playing for would have taken advantage of it by now.
All of the evidence suggests there’s no ability to predict streaky with the exception of pitchers as GW noted.
I would say every baseball player is streaky. It’s a matter of luck. In small samples even a bad hitter can post a .500 wOBA. A great hitter can post a wOBA of .085 in small samples.
If you have two .350 wOBA hitters and one makes more contact than the other, you might be wise to go with that player. The ups and downs may not be as large, but if we’re talking .350 and .380, forget about it. Take the .380 player and smile.
dmick89Quote Reply
I must confess, I got the atom question wrong. But I got all the rest right.
EdwinQuote Reply
@ Edwin:
Well now I feel like a jackass (dying laughing)
WaLiQuote Reply
dmick89 wrote:
I never got around to writing that article about variance because work was busy today, but I’ll say that reducing variance is only advisable if you are better than your opponent. A bad team wants to increase variance at every opportunity because it allows them “luck out.” This is why bad basketball teams shoot more 3s against good teams then against bad ones, among other things.
Think of it this way. If you average 5 runs a game, you’d rather get there via 10 games of 5 runs scored. If you average 3 runs a game, you’d rather get there via 5 games of 5 runs scored and 5 games of 1 run scored than 10 games of 3, because you have a negative expectation every time you score 3 runs and a positive one with 5 runs. The games you score 1 (and almost certainly lose) you likely would have lost anyway.
MylesQuote Reply
@ GW:
I got to the results and this was in red:
What? I know I got it right, but apparently it didn’t think I answered. Bullshit exam.
dmick89Quote Reply
WaLi wrote:
So do I, but as I grow older and the feeling becomes more and more common, I’ve learned to ignore it.
EdwinQuote Reply
I got 100%, but I probably would have got the Nitrogen one wrong if I hadn’t seen the spoiler in the thread, and I wasn’t totally sure about one or two others.
Considering I’m a science moron, I was pretty surprised.
Suburban kidQuote Reply
A laboratory test for the detection of a certain disease gives a positive result 5 percent of the time for people who do not have the disease. The test gives a negative result 0.3 percent of the time for people who have the disease. Large-scale studies have shown that the disease occurs in about 2 percent of the population.
(a) What is the probability that a person selected at random would test positive for this disease?
(b) What is the probability that a person selected at random who tests positive for the disease does not have the disease?
EdwinQuote Reply
a) 6.894 %
b) 71.1 %
MylesQuote Reply
To whit:
98% of the population doesn’t carry this disease. 98% * 95% (true negative) = 93.1%
98% of the population doesn’t carry this disease. 98% * 5% (false positive) = 4.9%
2% of the population do carry this disease. 2% * 99.7% (true positive) = 1.994%
2% of the population do carry this disease. 2% * .3% (false negative) = 0.006%
Positive tests: 4.9% + 1.994% = 6.894%
false positive (4.9%) / all positives (6.894%) = 71.1%
Of course, this is why when they “test” you for a disease, they actually test you like 5 times.
MylesQuote Reply
@ GW:
I missed the last one because I had no fracking idea what they were talking about.
Rice CubeQuote Reply
@ Myles:
Great job.
EdwinQuote Reply
@ Edwin:
2 and 288
Suburban kidQuote Reply
Edwin wrote:
I have a statistics degree and all it got me was points on a Cubs blog
MylesQuote Reply
@ Myles:
Fuck yeah it did.
EdwinQuote Reply
New shit: http://obstructedview.net/commentary-and-analysis/the-crosstown-classic-two-teams-one-cup-zero-fucks-given.html
Aisle424Quote Reply