PDA

View Full Version : Is Jay Bruce more streaky than others?

dougdirt
08-18-2011, 07:10 PM
I decided that this probably deserved its own thread aside from the other one. The conversations are slightly different. I didn't do a big in depth study of everyone in baseball. What I did was go to fangraphs and look at the MLB OPS leaders. I took Jay Bruce and the four players above and below them. I then took their end of game seasonal OPS and charted it, game by game, beginning with game #15 of their season (the start of their third week). I then placed all of the 9 players in the same chart. I left player names off of the chart.

Here is what the chart shows:

Can you tell which one is Jay Bruce? Does his fluctuation look much different than the other guys on the chart?

oneupper
08-18-2011, 07:49 PM
I say player A is Bruce.

I appreciate what you're trying to do, but this may not be the best way, since fluctuations in season-to-date OPS will become less pronounced as the season progresses (as is obvious in the chart).

What you might want to do, is get the monthly OPS's of those players and run a standard deviation on the numbers. It's a crude approximation to "streakiness", but it might tell you something.

AtomicDumpling
08-18-2011, 07:51 PM
I am guessing light green is Jay Bruce or it could be dark blue, they follow a similar path. I don't see any smooth lines in the chart. Lots of sharp ups and downs, especially early in the season before the sample sizes get large enough to be meaningful. Inconsistency rules.

AtomicDumpling
08-18-2011, 07:57 PM
I say player A is Bruce.

I appreciate what you're trying to do, but this may not be the best way, since fluctuations in season-to-date OPS will become less pronounced as the season progresses (as is obvious in the chart).

What you might want to do, is get the monthly OPS's of those players and run a standard deviation on the numbers. It's a crude approximation to "streakiness", but it might tell you something.

The problem with using monthly OPS is the arbitrary endpoints. It is unlikely that a player's hot and cold streaks are going to coincide with the beginning and ends of a month. In the rare instances when they do line up neatly with the month it makes them stand out and people notice, which is why some people have come to believe that Jay Bruce is streakier than others.

oneupper
08-18-2011, 08:04 PM
The problem with using monthly OPS is the arbitrary endpoints. It is unlikely that a player's hot and cold streaks are going to coincide with the beginning and ends of a month. In the rare instances when they do line up neatly with the month it makes them stand out and people notice, which is why some people have come to believe that Jay Bruce is streakier than others.

That's why its crude and "might" say something. That said, at least the monthly data is readily available. With more data points (years), it might tell you a little more.

I think the guys of "The Book" did some kind of measure of streakiness. Not quite sure how it was.

Hoosier Red
08-18-2011, 08:05 PM
How about a rolling 20 or 30 day OPS period?

Starting May 3 to today?

RedsManRick
08-18-2011, 08:07 PM
I would take issue with your approach, Doug. More than anything else, that chart just demonstrates the law of large numbers in action. The more data you accumulate, the closer the sample represents the population.

The thing we're looking for is performance variation between discrete time periods. But that variation is getting lost in the natural decrease in the effect of a given game on the cumulative total over time. A slump in April is going look like a massive swing because of it's proportionality to the season at the time. The exact same slump in July will result in much less movement in the line.

A more useful chart would be one that plots the absolute value of the difference between a series of discrete time periods.For example:

Data Point 1: (absolute value of) Week 1 OPS - Week 2 OPS
Data Point 2: (absolute value of) Week 2 OPS - Week 3 OPS
Data Point 3: (absolute value of) Week 3 OPS - Week 4 OPS
And so forth.

You could also analyze that resulting data set by looking at the total amount of variation (sum of the data points) or the standard deviation among them. The biggest problem we've had is people using months, which just gives us WAY too small of a sample to work with (6 per year). But that much more closely gets at what people think they're seeing.

dougdirt
08-18-2011, 08:32 PM
I thought about doing a two week at a time sample Rick.... I am just really bogged down right now with computer problems (I use two computers for work and both are crapping the bed on me right now and I am working on getting both back in shape, but its probably going to be two days worth of work). Maybe this weekend I can get around to that.

nate
08-18-2011, 09:45 PM
Is there anywhere one can get access to each PA in it's own row? I can get games from BR but I can't break it down to each PA.

Then one could look at a window of 600 (for example) PA which would normalize the amplitude of charts like Doug posted.

BTW, I applaud your efforts to study this, Doug!

jojo
08-20-2011, 04:45 PM
Can you tell which one is Jay Bruce?

Clearly he's the streakiest one... ;)

RedsManRick
08-20-2011, 06:03 PM
Is there anywhere one can get access to each PA in it's own row? I can get games from BR but I can't break it down to each PA.

Then one could look at a window of 600 (for example) PA which would normalize the amplitude of charts like Doug posted.

BTW, I applaud your efforts to study this, Doug!

Rotosheet?

08-20-2011, 06:45 PM
Jay Bruce has put up these numbers in the last 365 days.

.286/.365/.554 38 HR's 102 RBI .919 OPS.

Homer Bailey
08-20-2011, 06:52 PM
Jay Bruce has put up these numbers in the last 365 days.

.286/.365/.554 38 HR's 102 RBI .919 OPS.

Yowzas.

reds44
08-20-2011, 06:58 PM
Jay Bruce has put up these numbers in the last 365 days.

.286/.365/.554 38 HR's 102 RBI .919 OPS.
Left handed Jonny Gomes.

jojo
08-20-2011, 07:17 PM
Left handed Jonny Gomes.

If Jonny Gomes was good that is....

CrackerJack
08-21-2011, 04:03 AM
Do we really need graphs and stats since last August (do those 365 days include October through March? Really?) to determine that Bruce is incredibly streaky? That he's a mediocre offensive player "most" of the year at this point, and great for almost two months out of the season?

All this is fine if you're on a team that isn't "built" around said player who has an .852 OPS this year, despite going absolutely nuts for 7-8 weeks every season.

I'd recommend not relying on this player as a cornerstone of your lineup on a small market team that can't afford pitching, but as a solid addition to a well-constructed and reliable 1-4 hitter lineup where he can bat in the 5-6 spot and be plus player in that role.

08-21-2011, 07:10 AM
Do we really need graphs and stats since last August (do those 365 days include October through March? Really?) to determine that Bruce is incredibly streaky? That he's a mediocre offensive player "most" of the year at this point, and great for almost two months out of the season?

All this is fine if you're on a team that isn't "built" around said player who has an .852 OPS this year, despite going absolutely nuts for 7-8 weeks every season.

I'd recommend not relying on this player as a cornerstone of your lineup on a small market team that can't afford pitching, but as a solid addition to a well-constructed and reliable 1-4 hitter lineup where he can bat in the 5-6 spot and be plus player in that role.

Those last 365 days include his last 147 games 596 PA 525 AB.

.

Brutus
08-21-2011, 09:12 AM
How about a rolling 20 or 30 day OPS period?

Starting May 3 to today?

I have felt this is the best method you describe -- rolling 15 or 20-day periods. I think those methods will be more likely to tell us of any consistency or inconsistency that exists among players. It's a lot more work, but more work is probably necessary to find out such a thing.

Homer Bailey
08-21-2011, 12:28 PM
Do we really need graphs and stats since last August (do those 365 days include October through March? Really?) to determine that Bruce is incredibly streaky? That he's a mediocre offensive player "most" of the year at this point, and great for almost two months out of the season?

All this is fine if you're on a team that isn't "built" around said player who has an .852 OPS this year, despite going absolutely nuts for 7-8 weeks every season.

I'd recommend not relying on this player as a cornerstone of your lineup on a small market team that can't afford pitching, but as a solid addition to a well-constructed and reliable 1-4 hitter lineup where he can bat in the 5-6 spot and be plus player in that role.

Some of us like to base our opinions using facts and actual tangible, measurable things, so yes, we need stats so that if someone makes a statement (for example, Jay Bruce is a mediocre player for most of the year), it can actually be at least somewhat supported.

nate
08-21-2011, 01:38 PM
OK, here's something that somewhat attempts to normalize the peaks and valleys. It's not perfect but it's better than using the Earth's incremental rotation around the sun by 1/12th.

I couldn't find data by PA (sorry, Rick...I searched using your link and only saw game data) so starting at Bruce's career game 143 yielded 601 PA so I used a window of 143 games from that point cumulative up to yesterday's game throughout his career. This gave a window of anywhere from 538 to 601 PAs which isn't 100% accurate but I believe to be a better indicator than citing his last 20 PAs vs. pitchers with 3 vowels in their name at night on Kentucky Bluegrass in the 6th inning or later.

The blue line represents Bruce's career wOBA. The Red line is the windowed wOBA of 143 games. The x-axis represents career games (add 142 to it to get the "real" career game number) and the y-axis is the wOBA.

http://img835.imageshack.us/img835/707/jbulativewoba.png (http://imageshack.us/photo/my-images/835/jbulativewoba.png/)

nate
08-21-2011, 01:53 PM
So now, the challenge would be to identify whether or not this performance is streaky. To me, it would seem impossible to do so without comparing it other players.

BTW, for those who know more about standard deviation than me (probably everyone,) the standard deviation for Bruce's wOBA over this graph is 0.020. I take that to mean, the rough variance (or "streakiness") over these samples is .020 over the mean.

oneupper
08-21-2011, 02:08 PM
Nate, if I understand correctly, what you have determined is a 143-game moving average wOBA. (Correct me if I'm wrong).

The chart seems to indicate that Bruce's numbers are trending upwards (a good thing).
(you could do a regression and get an upwards trendline...to get really esoteric and pretend to predict the future)

However, if we are still on the issue of streakiness, the data doesn't address that, since 143 games is essentially a season and peaks and valleys have been evened out.
By the same measure, the standard deviation of a moving average really isn't telling us anything, except that performance has been changing (in this case on the rise).

A shorter period (say 20 games), could give you a better perspective. If compared to an adequate standard (maybe other similar wOBA players), it could give us an indication of how streaky Bruce is compared to that peer group.

Nice work.

nate
08-21-2011, 09:24 PM
Nate, if I understand correctly, what you have determined is a 143-game moving average wOBA. (Correct me if I'm wrong).

The chart seems to indicate that Bruce's numbers are trending upwards (a good thing).
(you could do a regression and get an upwards trendline...to get really esoteric and pretend to predict the future)

However, if we are still on the issue of streakiness, the data doesn't address that, since 143 games is essentially a season and peaks and valleys have been evened out.
By the same measure, the standard deviation of a moving average really isn't telling us anything, except that performance has been changing (in this case on the rise).

A shorter period (say 20 games), could give you a better perspective. If compared to an adequate standard (maybe other similar wOBA players), it could give us an indication of how streaky Bruce is compared to that peer group.

Nice work.

Thanks.

nate
08-21-2011, 09:25 PM
OK, so lets have three players with a similar wOBA (or OPS if you like) to Bruce and we'll compare. Please pick someone who isn't "streaky!"

RedsManRick
08-21-2011, 09:27 PM
I agree with one-upper. The question isn't about a long term trend, but rather stretches of 10-20 games where he gets particularly cold or hot. Like the method, Nate, but I think the question requires a smaller interval.

Scrap Irony
08-21-2011, 09:32 PM
OK, so lets have three players with a similar wOBA (or OPS if you like) to Bruce and we'll compare. Please pick someone who isn't "streaky!"

The people around him in wOBA are Josh Hamilton, Carlos Quentin, Nelson Cruz, and Rickie Weeks.

nate
08-22-2011, 12:27 AM
I agree with one-upper. The question isn't about a long term trend, but rather stretches of 10-20 games where he gets particularly cold or hot. Like the method, Nate, but I think the question requires a smaller interval.

It's easy enough to make that chart but it's pointless without comparing Bruce to other players.

So, who should he be compared to and what should the size of the window be?

edabbs44
08-22-2011, 08:00 AM
It's easy enough to make that chart but it's pointless without comparing Bruce to other players.

So, who should he be compared to and what should the size of the window be?

The names above and I think you need to look at multiple windows. Like 15, 30 and 45 game windows.

Blitz Dorsey
08-22-2011, 09:29 AM
Doug, why did you say "more streaky" and not "streakier" in the thread title?

Inquiring minds want to know. ;)

osuceltic
08-22-2011, 11:09 AM
This thread is funny. Look at the time and energy being spent to prove something that anyone who watches this team regularly already knows. Jay Bruce is streaky. He's not the only guy. Is he more streaky than an average major leaguer? Don't know and don't really care.

Is he more streaky than Drew Stubbs? I don't think there's much of a difference, other than Jay's hot streaks are hotter than Drew's hot streaks. Is Bruce more streaky than Joey Votto? Absolutely. At the risk of stating the obvious, that's the crux of the Bruce arguments. His supporters are OK with the streakiness, point to the end-of-year numbers as the bottom line and say that's good enough. The other side expects him to be more Votto than Stubbs and believe the long cold streaks aren't negated by the hot streaks.

Anyway, someone is going to spend a lot of time studying something that doesn't need to be studied.

nate
08-22-2011, 11:17 AM
OK, I'll run those names and intervals tonight.

nate
08-22-2011, 11:20 AM
This thread is funny. Look at the time and energy being spent to prove something that anyone who watches this team regularly already knows. Jay Bruce is streaky. He's not the only guy. Is he more streaky than an average major leaguer? Don't know and don't really care.

Is he more streaky than Drew Stubbs? I don't think there's much of a difference, other than Jay's hot streaks are hotter than Drew's hot streaks. Is Bruce more streaky than Joey Votto? Absolutely. At the risk of stating the obvious, that's the crux of the Bruce arguments. His supporters are OK with the streakiness, point to the end-of-year numbers as the bottom line and say that's good enough. The other side expects him to be more Votto than Stubbs and believe the long cold streaks aren't negated by the hot streaks.

Anyway, someone is going to spend a lot of time studying something that doesn't need to be studied.

Then don't participate in the thread.

signalhome
08-22-2011, 11:37 AM
This thread is funny. Look at the time and energy being spent to prove something that anyone who watches this team regularly already knows. Jay Bruce is streaky. He's not the only guy. Is he more streaky than an average major leaguer? Don't know and don't really care.

Is he more streaky than Drew Stubbs? I don't think there's much of a difference, other than Jay's hot streaks are hotter than Drew's hot streaks. Is Bruce more streaky than Joey Votto? Absolutely. At the risk of stating the obvious, that's the crux of the Bruce arguments. His supporters are OK with the streakiness, point to the end-of-year numbers as the bottom line and say that's good enough. The other side expects him to be more Votto than Stubbs and believe the long cold streaks aren't negated by the hot streaks.

Anyway, someone is going to spend a lot of time studying something that doesn't need to be studied.

Some of us need cold, hard facts to accept things, rather than just accepting what we feel or think as being the absolute truth. There are plenty of instances throughout my life (not just involving baseball) where I've felt/thought/said something that was eventually proven to be absolutely false; due to this, along with the fact that I hate being wrong about anything, I usually keep quiet until I can say something with a high degree of certainty. It feels like Bruce may be streakier than the average major league player*, but it's something I'll need to see supported with good data before I really believe it (using arbitrary endpoints such as months would not, in my opinion, represent good data). So yes, this is something that I feel needs to be studied. It may not be very important (not sure that being streaky is necessarily a bad thing), but it's a very interesting piece of information that I'd love to know. Keep up the good work, nate.

*Even if he is streaky, I'm not sure it's some big problem we need to be all too concerned with. End-of-year numbers are probably much more important than any measure of streakiness.

Homer Bailey
08-22-2011, 12:02 PM
Then don't participate in the thread.

+1

Some of us need cold, hard facts to accept things, rather than just accepting what we feel or think as being the absolute truth. There are plenty of instances throughout my life (not just involving baseball) where I've felt/thought/said something that was eventually proven to be absolutely false; due to this, along with the fact that I hate being wrong about anything, I usually keep quiet until I can say something with a high degree of certainty. It feels like Bruce may be streakier than the average major league player*, but it's something I'll need to see supported with good data before I really believe it (using arbitrary endpoints such as months would not, in my opinion, represent good data). So yes, this is something that I feel needs to be studied. It may not be very important (not sure that being streaky is necessarily a bad thing), but it's a very interesting piece of information that I'd love to know. Keep up the good work, nate.

*Even if he is streaky, I'm not sure it's some big problem we need to be all too concerned with. End-of-year numbers are probably much more important than any measure of streakiness.

+1

RedsManRick
08-22-2011, 12:43 PM
This thread is funny. Look at the time and energy being spent to prove something that anyone who watches this team regularly already knows. Jay Bruce is streaky. He's not the only guy. Is he more streaky than an average major leaguer? Don't know and don't really care.

Is he more streaky than Drew Stubbs? I don't think there's much of a difference, other than Jay's hot streaks are hotter than Drew's hot streaks. Is Bruce more streaky than Joey Votto? Absolutely. At the risk of stating the obvious, that's the crux of the Bruce arguments. His supporters are OK with the streakiness, point to the end-of-year numbers as the bottom line and say that's good enough. The other side expects him to be more Votto than Stubbs and believe the long cold streaks aren't negated by the hot streaks.

Anyway, someone is going to spend a lot of time studying something that doesn't need to be studied.

If this thread can either put the claim to bed and/or simply establish a reasonable standard for what streaky looks like, it will be time well spent.

bucksfan2
08-22-2011, 12:58 PM
If this thread can either put the claim to bed and/or simply establish a reasonable standard for what streaky looks like, it will be time well spent.

Jay Bruce.

RedsManRick
08-22-2011, 01:05 PM
Jay Bruce.

Thank you for this helpful contribution.

bucksfan2
08-22-2011, 01:19 PM
Thank you for this helpful contribution.

I guess the issue I take is that whenever the issue of streakieness comes about there are always big qualifiers that pretty much find any kind of figures useless. When monthly stats are used people say that they are just arbitrary beginnings and ends and should be taken with a grain of salt. Then we have 365 day numbers which takes into consideration 2 entirely different seasons. Then we try and compare Jay Bruce to good players and we hear "he isn't Joey Votto so why compare him to Votto." And then finally we hear that Jay Bruce is only 24 years old.

Is there anyone on RZ who would argue that Jay Bruce isn't streaky? Is there anyone on RZ would would argue that Jay Bruce is the definition of consistent? Is there anyone on RZ would deny that Jay Bruce is a good player and he is an asset to the Reds but goes through maddening streaks of not being productive at the plate?

RedsManRick
08-22-2011, 02:25 PM

I guess the issue I take is that whenever the issue of streakieness comes about there are always big qualifiers that pretty much find any kind of figures useless. When monthly stats are used people say that they are just arbitrary beginnings and ends and should be taken with a grain of salt. Then we have 365 day numbers which takes into consideration 2 entirely different seasons. Then we try and compare Jay Bruce to good players and we hear "he isn't Joey Votto so why compare him to Votto." And then finally we hear that Jay Bruce is only 24 years old.

Is there anyone on RZ who would argue that Jay Bruce isn't streaky? Yes, there are. Because it's a relative term and we have nothing close to agreement on what Bruce is being compared to nor any actual measurement of what it is to begin with. There are people here who like that word to have some kind of relatively consistent meaning, especially when people are using the word to make an argument that Bruce is costing the team wins.

Is there anyone on RZ would would argue that Jay Bruce is the definition of consistent? I doubt it. But if every player is streaky, then who gives a flip if Bruce is streaky? Clearly the claim is one of degree, that Bruce is streakier than most.

Is there anyone on RZ would deny that Jay Bruce is a good player and he is an asset to the Reds but goes through maddening streaks of not being productive at the plate?
I doubt it. But it's one thing to say how something makes you feel. It's another to make a claim about what that thing means in terms of wins and losses. I would argue that most players go through said streaks.

If the only point is to suggest that Bruce has gone through streaks of poor performance, this whole conversation would have not happened. But that's not the point people have been making.

Either streakiness is meaningless and not worth mentioning or it means something real. If people are going to continue to talk about it like it means something notable, it merits the effort to figure out just what that meaning is.

08-22-2011, 02:32 PM

I guess the issue I take is that whenever the issue of streakieness comes about there are always big qualifiers that pretty much find any kind of figures useless. When monthly stats are used people say that they are just arbitrary beginnings and ends and should be taken with a grain of salt. Then we have 365 day numbers which takes into consideration 2 entirely different seasons. Then we try and compare Jay Bruce to good players and we hear "he isn't Joey Votto so why compare him to Votto." And then finally we hear that Jay Bruce is only 24 years old.

Is there anyone on RZ who would argue that Jay Bruce isn't streaky? Is there anyone on RZ would would argue that Jay Bruce is the definition of consistent? Is there anyone on RZ would deny that Jay Bruce is a good player and he is an asset to the Reds but goes through maddening streaks of not being productive at the plate?

Don't you think a players last 147 games are a better indicator of a his production than any 30 day span within those 147 games?

BCubb2003
08-22-2011, 02:41 PM
Don't you think a players last 147 games are a better indicator of a his production than any 30 day span within those 147 games?

That's the debate in a nutshell. Especially if you're in one of those 30-day spans.

bucksfan2
08-22-2011, 02:48 PM
Don't you think a players last 147 games are a better indicator of a his production than any 30 day span within those 147 games?

It depends. I think that baseball is a seasonal sport. Meaning that quite a bit can change from one year to the next. The big break in the game from Nov to April gives a player time to work on things, get his head right, and maybe most importantly heal. I don't put a lot of stock in someone combining the last month in 2010 with 2011 and trying to make a point. I like to look at it more year to year as opposed to a 365 calendar.

The 30 day span gives you a firm starting and end point. I also think that if your having a bad month you can look to the start of the next month as a fresh start. I think there is a psychological aspect to the changing of the calendar that can help out players.

08-22-2011, 02:59 PM
It depends. I think that baseball is a seasonal sport. Meaning that quite a bit can change from one year to the next. The big break in the game from Nov to April gives a player time to work on things, get his head right, and maybe most importantly heal. I don't put a lot of stock in someone combining the last month in 2010 with 2011 and trying to make a point. I like to look at it more year to year as opposed to a 365 calendar.

The 30 day span gives you a firm starting and end point. I also think that if your having a bad month you can look to the start of the next month as a fresh start. I think there is a psychological aspect to the changing of the calendar that can help out players.

Then you would place more value in year end stats than the ebb and flow of numbers that make up those year end numbers? So would I.

bucksfan2
08-22-2011, 03:11 PM
Then you would place more value in year end stats than the ebb and flow of numbers that make up those year end numbers? So would I.

Sure I place value in the end of year numbers. But I do like to look at the numbers and see if there was a deviation from the norm throughout the year. I like averages but realize that a really good or really bad period of time can skew the overall average.

08-22-2011, 03:17 PM
Sure I place value in the end of year numbers. But I do like to look at the numbers and see if there was a deviation from the norm throughout the year. I like averages but realize that a really good or really bad period of time can skew the overall average.

But the overall year end stats aren't skewed, they are a history of what happened that season.

Homer Bailey
08-22-2011, 03:54 PM
Sure I place value in the end of year numbers. But I do like to look at the numbers and see if there was a deviation from the norm throughout the year. I like averages but realize that a really good or really bad period of time can skew the overall average.

No, it does not skew the average, it makes up the average. And you, along with other posters, only want to consider the negative effects of the cold streaks, and ignore the positive effects of the hot streaks.

bucksfan2
08-22-2011, 03:59 PM
But the overall year end stats aren't skewed, they are a history of what happened that season.

Lets say that your a salesman and you are paid strictly on commission. You make a nice yearly of \$80,000. Uncle Sam doesn't care how that is distributed over the course of a year. Based upon the total aggregate you should make enough money to support your family and not have many worries. However the distribution of your monthly income is all over the place.

In Jan and Feb you make 3000 each, barely enough to pay your bills. You have to charge more and pay less. Your behind the 8 ball early, missing payments and occurring a lot of debt. Then in March you have a banner month and pull in 10,000. Now you attempt to pay down some of your debt but that keeps compounding on itself. The first three months of the year when you average them out they look pretty good on paper. But the overall aggregate doesn't tell the whole story. In the end of the year you 80K will be the summary of how much you made. But it sure as heck doesn't go anywhere to tell the story of the year.

dougdirt
08-22-2011, 04:00 PM
Lets say that your a salesman and you are paid strictly on commission. You make a nice yearly of \$80,000. Uncle Sam doesn't care how that is distributed over the course of a year. Based upon the total aggregate you should make enough money to support your family and not have many worries. However the distribution of your monthly income is all over the place.

In Jan and Feb you make 3000 each, barely enough to pay your bills. You have to charge more and pay less. Your behind the 8 ball early, missing payments and occurring a lot of debt. Then in March you have a banner month and pull in 10,000. Now you attempt to pay down some of your debt but that keeps compounding on itself. The first three months of the year when you average them out they look pretty good on paper. But the overall aggregate doesn't tell the whole story. In the end of the year you 80K will be the summary of how much you made. But it sure as heck doesn't go anywhere to tell the story of the year.

You don't go into debt in baseball stats. Games won/lost in April count exactly the same as the ones in September.

Homer Bailey
08-22-2011, 04:01 PM
Lets say that your a salesman and you are paid strictly on commission. You make a nice yearly of \$80,000. Uncle Sam doesn't care how that is distributed over the course of a year. Based upon the total aggregate you should make enough money to support your family and not have many worries. However the distribution of your monthly income is all over the place.

In Jan and Feb you make 3000 each, barely enough to pay your bills. You have to charge more and pay less. Your behind the 8 ball early, missing payments and occurring a lot of debt. Then in March you have a banner month and pull in 10,000. Now you attempt to pay down some of your debt but that keeps compounding on itself. The first three months of the year when you average them out they look pretty good on paper. But the overall aggregate doesn't tell the whole story. In the end of the year you 80K will be the summary of how much you made. But it sure as heck doesn't go anywhere to tell the story of the year.

How does this relate to baseball? Do losses early in the season count double?

BCubb2003
08-22-2011, 04:15 PM
The season goes beyond May, too. I'm not sure why the two sides are talking past each other, macro vs. micro. It's legitimate to say that Bruce's May counts. And in June and July, when the offense has disappeared, it's legitimate to ask what happened to Bruce. Does he need a break? Does he need to adjust? Will he regain his form? (He did, eventually.) May is only part of the picture.

bucksfan2
08-22-2011, 04:21 PM
How does this relate to baseball? Do losses early in the season count double?

Losses in the early season can mean you become sellers instead of traders at the deadline. Losses early in the season mean you pack it in for next year instead of going for it all. Losses early in the season can put you in a huge hole that is difficult to climb out of.

I kinda like the golf reference for baseball. You can't win the golf tournament on the first day but you sure as heck can lose it. You can't win the division during the first 1/4 of the season but you sure as heck can lose it.

Homer Bailey
08-22-2011, 04:29 PM
Losses in the early season can mean you become sellers instead of traders at the deadline. Losses early in the season mean you pack it in for next year instead of going for it all. Losses early in the season can put you in a huge hole that is difficult to climb out of.

I kinda like the golf reference for baseball. You can't win the golf tournament on the first day but you sure as heck can lose it. You can't win the division during the first 1/4 of the season but you sure as heck can lose it.

How does this relate to Bruce and his .850 pre all star break OPS?

oneupper
08-22-2011, 04:32 PM
Let me chime in, since in the world of investing we go over this all the time :)

Lets compare two investment alternatives, both of which have the exact same expected return over time (a long time). 10% for example.

However, the yearly returns on investment A fluctuate between +5 and +15%, while investment B can show yearly returns between -10 and +30%. On average, they will both yield 10%.

Are they equivalent? Obviously not. B is inherently more risky and therefore less desirable. They might both be good, but A is a better choice. There is value in consistency.

In terms of baseball production, there is also probably some value to consistency. A team of streaky players is more likely to produce streaky results. Do streaky baseball results produce fewer (or more) wins over a season? Hard to say, theoretically they should produce more VARIABLE results,

So arguably if there is some measure that should correlate to wins (such as perhaps run differential), a streaky team could be more likely to deviate one way or another.
(Always keep in mind that good and streaky is always better than consistently BAD. )

This is all terribly hypothetical, of course. And as confabulated as it all may sound, some of us just want to test the hypothesis. Or something like that.

bucksfan2
08-22-2011, 04:32 PM
How does this relate to Bruce and his .850 pre all star break OPS?

2/3 of that time was spend sub .700.

Homer Bailey
08-22-2011, 04:39 PM
2/3 of that time was spend sub .700.

1/3 of that time was spent OPS'ing over 1.100.

Carried the team on his back for an entire month. But that doesn't count I guess?

Scrap Irony
08-22-2011, 04:41 PM
No, it does not skew the average, it makes up the average. And you, along with other posters, only want to consider the negative effects of the cold streaks, and ignore the positive effects of the hot streaks.

This is, again, not true.

For the third time:

All posters on this thread (and the other one) recognize that Jay Bruce is an outstanding ballplayer who, at times, can carry a team offensively. When he's "on", he's a HoF level talent. An MVP talent.

But, Jay Bruce is ALSO a player prone to prolonged and massive slumps. These slumps affect the team negatively.

He is, as a ballplayer, flawed.

Does that mean he's not good? Obviously not.

Does it mean he's streaky? That's what we're finding out here.

In short, Bruce=good. Does Bruce=streaky?

Plus Plus
08-22-2011, 04:41 PM
2/3 of that time was spend sub .700.

And when it was 1.140 and Bruce was playing like Barry Bonds redux, the Reds went 14-15. This is in opposition to June and July of 2011, where the Reds went 14-12 and 11-15 respectively and Bruce was OPSing .649 and .805, respectively. The Reds are 9-10 thus far in August, so maybe Bruce OPSing below .700 (or .800) leads to Reds wins?

Wouldn't that be an interesting situation...

bucksfan2
08-22-2011, 04:47 PM
1/3 of that time was spent OPS'ing over 1.100.

Carried the team on his back for an entire month. But that doesn't count I guess?

Where did I say that? I never have discounted Jay's May. My thinking that while he went on a terror in May it accounted for 33% of the games. He went into a funk in April and June which accounted for 66% of the games. In an ideal world I would have absolutely given up some of his May production for a more even distribution. Would it adversely effect May? sure, but it would positively effect a larger size of games.

signalhome
08-22-2011, 04:50 PM
But, Jay Bruce is ALSO a player prone to prolonged and massive slumps. These slumps affect the team negatively.

He is, as a ballplayer, flawed.

This is where the discrepancy lies. You say those two things as if they are absolutes that cannot be debated, whereas some of us:

a) are not so sure that he is any more prone to massive slumps than any other player, or
b) simply don't believe that being streaky is some awful thing that everyone should be upset about, or
c) believe any major variance in his stats when setting arbitrary endpoints such as months (a flawed exercises as is) is simply the result of chance and isn't repeatable throughout the course of a career, and therefore, isn't anything worth worrying about.

signalhome
08-22-2011, 04:59 PM
Where did I say that? I never have discounted Jay's May. My thinking that while he went on a terror in May it accounted for 33% of the games. He went into a funk in April and June which accounted for 66% of the games. In an ideal world I would have absolutely given up some of his May production for a more even distribution. Would it adversely effect May? sure, but it would positively effect a larger size of games.

It seems as if you are simply looking at games as good vs. bad. However, that's not exactly how it works. We don't just chalk up a tally to the respective side whenever he has a good game or a bad game. There are vast differences in the kinds of "good" games a player can have as well as "bad" games a player can have. 2-3 with a BB and two singles is a good game; 2-3 with a BB and two HRs is an excellent game. 1-4 with a single is a bad game; 0-5 is an awful game. To just group both sets into the same category is unfair. In May, Bruce had a ridiculous amount of excellent games that went a long way to countering the sheer amount of mediocrity he tallied up in March/April and June, and in the overall grand scheme of it, it made for a really good three-month stretch (.842 OPS through June 30th). If we just took seven of those excellent games in May and swapped them with some of his awful games throughout June and April to give him a much more consistent distribution, does it really make one bit of difference in the amount of value Bruce has provided the team? I don't think so.

Homer Bailey
08-22-2011, 05:04 PM
This is, again, not true.

For the third time:

All posters on this thread (and the other one) recognize that Jay Bruce is an outstanding ballplayer who, at times, can carry a team offensively. When he's "on", he's a HoF level talent. An MVP talent.

But, Jay Bruce is ALSO a player prone to prolonged and massive slumps. These slumps affect the team negatively.

He is, as a ballplayer, flawed.

Does that mean he's not good? Obviously not.

Does it mean he's streaky? That's what we're finding out here.

In short, Bruce=good. Does Bruce=streaky?

It is absolutely 100% true that several posters only focus on the negatives for Jay Bruce.

Homer Bailey
08-22-2011, 05:12 PM
Where did I say that? I never have discounted Jay's May. My thinking that while he went on a terror in May it accounted for 33% of the games. He went into a funk in April and June which accounted for 66% of the games. In an ideal world I would have absolutely given up some of his May production for a more even distribution. Would it adversely effect May? sure, but it would positively effect a larger size of games.

Yet, as I have stated, and as Plus Plus just stated, his overall production has not correlated to Reds wins and losses so far this year. And I'm yet to see a single study that suggests that a team is more likely to win more games if a player goes .681, 1.141, .649. than a player that goes .842, .842, .842.

Never, in my history of watching baseball, analyzing, and discussing it, have I ever encountered a player where people are willing to dissect his otherwise good numbers, and tell me why they aren't good enough.

RedsManRick
08-22-2011, 05:18 PM
I do think there's some merit to the point that performance before the trading deadline has a small multiplier effect due to the fact that teams make decisions to acquire/shed talent based on where they are in the standings. That said, we're looking at something like, at most, 1 extra loss pre-deadline vs. what would have happened given an even distribution of performance. I find it hard to believe that the 1 extra loss (again, it's probably less than that) makes that much of a difference in the deadline decision making process. And by the end of the season, the loss is offset by an extra win.

Forest vs. Trees. Arguing about the distribution of trees when the forest is too small to begin with doesn't really make any sense to me.

Scrap Irony
08-22-2011, 06:34 PM
It is absolutely 100% true that several posters only focus on the negatives for Jay Bruce.

And I disagree with your assertion.

I've only read one poster on ORG who believes Bruce is a poor player. Literally everyone else on this board believes him to be a good player. Some believe he has flaws.

Homer Bailey
08-22-2011, 06:35 PM
And I disagree with your assertion.

I've only read one poster on ORG who believes Bruce is a poor player. Literally everyone else on this board believes him to be a good player. Some believe he has flaws.

There's a difference in what you are saying and what I'm saying.

I said there are several that only focus on the negatives for Bruce. I did not say that these people said they were poor players.

Plus Plus
08-22-2011, 06:37 PM
And I disagree with your assertion.

I've only read one poster on ORG who believes Bruce is a poor player. Literally everyone else on this board believes him to be a good player. Some believe he has flaws.

It isn't about posters saying "Jay Bruce is a poor player." It is about posters saying "wow, look how inconsistent Jay Bruce is- months with OPS of .681 and .649 are weighing the team down!" I think that saying that there are several posters who are very inclined to point out the low OPS months and how they may have negatively affected the Reds without also pointing out (see- ignoring) the high OPS months and how they may have positively affected the Reds is absolutely true.

Scrap Irony
08-22-2011, 06:50 PM
I think that saying that there are several posters who are very inclined to point out the low OPS months and how they may have negatively affected the Reds without also pointing out (see- ignoring) the high OPS months and how they may have positively affected the Reds is absolutely true.

So posters are required now to post all the good before discussing the bad?

Were posters supposed to focus on Gomes' good May from last season befofre talking about his poor season? Arroyo's good numbers as a Red before complaining about him? Volquez's good season two years ago before calling him a disappointment?

Redszone posters aren't ignoring his good numbers-- they fully recognize he's been good.

But they also want to discuss his flaws, his slumps, their effect (if any) on the team, the won/loss record (if possible), and his streakiness (if, indeed, if exists).

Why that's looked down upon by a select number of posters is a mystery similar to that of Adam Dunn.

kaldaniels
08-22-2011, 08:56 PM
It isn't about posters saying "Jay Bruce is a poor player." It is about posters saying "wow, look how inconsistent Jay Bruce is- months with OPS of .681 and .649 are weighing the team down!" I think that saying that there are several posters who are very inclined to point out the low OPS months and how they may have negatively affected the Reds without also pointing out (see- ignoring) the high OPS months and how they may have positively affected the Reds is absolutely true.

There was a post made in the past week that said good players don't have bad streaks like Jay Bruce goes on. Now you tell me, what is that saying :dunno:

nate
08-22-2011, 09:42 PM
I would offer that arguing about what "Redszone" thinks is like a mock draft for a fantasy cat herding league.

Raisor
08-22-2011, 10:28 PM
I would offer that arguing about what "Redszone" thinks is like a mock draft for a fantasy cat herding league.

I'm taking Buttons in the first round.

Patrick Bateman
08-22-2011, 10:31 PM
So posters are required now to post all the good before discussing the bad?

Were posters supposed to focus on Gomes' good May from last season befofre talking about his poor season? Arroyo's good numbers as a Red before complaining about him? Volquez's good season two years ago before calling him a disappointment?

Redszone posters aren't ignoring his good numbers-- they fully recognize he's been good.

But they also want to discuss his flaws, his slumps, their effect (if any) on the team, the won/loss record (if possible), and his streakiness (if, indeed, if exists).

Why that's looked down upon by a select number of posters is a mystery similar to that of Adam Dunn.

Well I certainly don't think it's a requirement, but at the same time, when players that clearly have solid value to the Reds are harped on continuously without a reasonable attempt to give some credit to the overall value of the player, I'm not sure why it comes as a surprise that Reds fans might be a little critical of such a one sided analyis.

IMO, most Jay Bruce fans have better demonstrated the flaws in his game than the detractors have in demonstrated the skills that he brings. I may be biased considering I am one of the many who are more in support of his game, but when you look at his overall value in comparison to his peers (both the Reds and the rest of the majors), I find it pretty difficult to lament the fact that his overall distribution of when he does good things over the fact that he actually does do those good things.

Or then again, we could also just be being a little picky with each other the fact that the Reds have grossly underachieved this year, and it's simply easier to try to find warts in players at this point.

nate
08-22-2011, 11:20 PM
I'm taking Buttons in the first round.

Buttons let the league with fewest hairballs two years ago.

nate
08-22-2011, 11:22 PM
Here's the five players mentioned based on cumulative WPA with a 60 PA window. Note, not all players have the same number of PAs. I used WPA because Fangraphs had the play data with WPA for each player. Thanks to texasdave for the tip.

http://img695.imageshack.us/img695/9306/wpa60.png (http://imageshack.us/photo/my-images/695/wpa60.png/)

mth123
08-22-2011, 11:33 PM
Or then again, we could also just be being a little picky with each other the fact that the Reds have grossly underachieved this year, and it's simply easier to try to find warts in players at this point.

Underachieved or over-rated?

Rolen's drop could be seen from miles away and was on display for the entire second half of 2010, yet the Reds opened the season with him in the clean-up spot.

Stubbs had a.715 OPS in AAA which he surpassed quite abit as a major leaguer, yet was counted on as one of those "young players who should improve" when nothing in his resume pointed toward expecting more than we got in 2010.

Janish was a known poor hitter whose pattern is wearing down when he plays a lot, yet his .700+ OPS from 2010 as a sparingly used player was viewed as a new standard for him, rather than expecting him to drop like a rock with regular playing time which is the norm with the weak hitting MI type.

Volquez was less than a year removed from TJ and rarely does anyone on that timetable do much on the mound, yet he was not only counted upon to hold down a rotation spot, but was named the opening day starter and was the designated ace.

Bailey spent 90 days on the DL with shoulder problems in 2010, has never really extablished himself but was counted on to be a solid member of the rotation.

Wood had 100 big league innings and was hardly established, but was also considered a main cog in the rotation.

Gomes had the legendary "one good month" in 2010, but was brought back as the starting LF and a middle of the order bat with seemingly little attempt to find an upgrade.

I could go on. The point is, there may have been reason to hope that these guys could come through, but the organization seemed expect and count on it. They didn't really underachieve. The Reds played the lottery and many of the tickets were losers. That isn't underachieving. Its what should be expected wth Lotto tickets.

This all very much relates to Jay Bruce and his streakiness. Like it or not, as one of the team's few above average players, when he disappears for weeks, an already short handed roster becomes glaringly thin and people note and comment on it. It doesn't really mean anyone thinks he's bad. The problem is the GM who confused hoping with planning and left him out on the edge of the cliff all alone.

Plus Plus
08-23-2011, 12:40 AM
This all very much relates to Jay Bruce and his streakiness. Like it or not, as one of the team's few above average players, when he disappears for weeks, an already short handed roster becomes glaringly thin and people note and comment on it. It doesn't really mean anyone thinks he's bad. The problem is the GM who confused hoping with planning and left him out on the edge of the cliff all alone.

But then why are there like 1,500 posts debating Jay Bruce and where he fits in the "blame game," whether due to production or streakiness or overrated defense and not a fraction of that number providing explanation for the losses beyond the perceived disappearance of Jay Bruce?

The team has a winning record in months that Bruce has "disappeared" in, and a losing record in months that Bruce has "raked" in. Yet the discussion seemingly begins and ends with this one player rather than the other 30+ that have played for the Reds this year.

dougdirt
08-23-2011, 02:54 AM
Here is what I did for this one. I took the same 9 players (Bruce, Gonzalez, Helton, Kotchman, Pedroia, Peralta, Ramirez, Youkilis, Zobrist) and broke their season down by groups of 14 games (in terms of team games, not individual games though it only came into play with one guy who missed time). I then calculated that 14 games worth of OPS. Here are the results:

Every single player on the chart has a similar low point with the exception of one player who had just an atrocious two week stretch. Can you pick out Jay Bruce? If so, does his streakiness really differ from the other guys shown here?

08-23-2011, 05:18 AM
All I and other Jay Bruce supporters in this thread are saying is "Jay Bruce is a very productive young ballplayer". We should all appreciate that fact.

There is no good reason to place so much of the blame for this teams shortcomings on the shoulders of Bruce as some members of RZ and most of the local media so easily do.

Sure Bruce has his flaws. Is there room for improvement? Yes.

There has never been nor will there ever be a perfect ballplayer so there is absolutely no reason to nail Jay Bruce to a cross for not being perfect.

I doubt there is another baseball town in the country with the history Cincinnati has for underrating or undervaluing it's young sluggers. A whole lot of that can be traced back to fans buying wholeheartedly into the opinions of local sports talk host, the guys in the radio booth and the beat writers. It stretches all the way back to the misguided opinions about Frank Robinson.

Seems to me many have already made up their own mind in this discussion so I'll try to bow out gracefully for a day or two. (Sorry about the rant.)

mth123
08-23-2011, 07:26 AM
But then why are there like 1,500 posts debating Jay Bruce and where he fits in the "blame game," whether due to production or streakiness or overrated defense and not a fraction of that number providing explanation for the losses beyond the perceived disappearance of Jay Bruce?

The team has a winning record in months that Bruce has "disappeared" in, and a losing record in months that Bruce has "raked" in. Yet the discussion seemingly begins and ends with this one player rather than the other 30+ that have played for the Reds this year.

I have to agree that the root cause is not Jay Bruce. IMO it was pretty obvious that Jay Bruce would need to be MVP caliber for this team to have a chance based on how it was assembled (and I said so all last winter and this spring and questioned whether even that would be enough). Based on the roadmap the GM laid out, counting on improvement from young players and all, Bruce needed to be about 100 OPS points higher than he is to offset the other predictable drops.

The root cause was expecting too much, but the GM built his team on those hopes, so Jay Bruce merely being very good instead of the being the best in the league means he's not doing enough compared to what the plan seemed to call for. The question marks in this rotation and the predictable drops at other positions pretty much suggested that Bruce and Votto needed to be a 21st century version of Ruth and Gehrig for this team to win. He's not Babe Ruth (yet) but he's still pretty good.

oneupper
08-23-2011, 09:00 AM
Here is what I did for this one. I took the same 9 players (Bruce, Gonzalez, Helton, Kotchman, Pedroia, Peralta, Ramirez, Youkilis, Zobrist) and broke their season down by groups of 14 games (in terms of team games, not individual games though it only came into play with one guy who missed time). I then calculated that 14 games worth of OPS. Here are the results:

Every single player on the chart has a similar low point with the exception of one player who had just an atrocious two week stretch. Can you pick out Jay Bruce? If so, does his streakiness really differ from the other guys shown here?

DD, nice work. The chart, however, is very noisy and won't give us a measure of streakiness (or rather, inconsistency).
If you ran a standard deviation on the data, it would give a better idea of how much each player 'strays" from his season average.

Even better, would be a "rolling" average, such as nate is using in his calculations. This would lessen the effect of arbitrary cutoff points, which can influence when you're using small samples. It also gives you much more data points.
The desired output result would continue to be the standard deviation of those data points, rather than a graph.

RedsManRick
09-07-2011, 01:55 PM
Just came across this. Haven't had the chance to read it, but I imagine it contains a lot of insights related to this conversation: http://www.beyondtheboxscore.com/2011/8/2/2309646/now-available-hitter-volatility-calculator

bucksfan2
09-07-2011, 03:10 PM
Last night I caught a little of the Reds game and Chris Welsh made an interesting comment about Jay Bruce. He said in batting practice he was squaring up every ball thrown. He made mention how there are times when Jay squares up as many as he fouls to the backstop. Just thought it was interesting related to Jay.

mbgrayson
09-07-2011, 04:20 PM
Just came across this. Haven't had the chance to read it, but I imagine it contains a lot of insights related to this conversation: http://www.beyondtheboxscore.com/2011/8/2/2309646/now-available-hitter-volatility-calculator (http://www.beyondtheboxscore.com/2011/8/2/2309646/now-available-hitter-volatility-calculator)

Using the measure that they have created, which factors in wOBA, ISO, BB%, K%, you get a volatility index. Lower volatility scores are around .5 or .6, and high scores (which equal high volatility) are over 1.0. On part IV of the series of articles (http://www.beyondtheboxscore.com/2011/7/8/2251452/is-volatility-an-attribute-player-volatility-part-iv), it lists the most and least volatile hitters over the 2007-09 time period. Their chart shows that the most volatile player was Bengie Molina in 2009, with a 1.09 score.

Using their online calculator, and putting Bruce's 2011 numbers in, the result is 1.05, which is very volatile. I then ran Drew Stubb's 2011 numbers, and his result is 1.146, even higher.

By comparison, Joey Votto is .576, Brandon Phillips is .883, Paul Janish is 1.344(!), Edgar Renteria is 1.087, Scott Rolen is 1.176, Chris Heisey is 1.149, Ryan Hanigan is .858. and Ramon Hernandez is .924. That is all that I had time to run.

If there is anything to all of these numbers, it would show that the Reds have LOTS of very streaky hitters, with Bruce, Renteria, Stubbs, Heisey, Rolen, and Janish all being in the very high range on volatility.

In fact this may help to explain why we have such a big difference between actual wins and pythag wins. When hitters get hot, they score lots of runs, and win games by larger margins. When they get cold, the team doesn't score enough runs. Of course, they ususally don't all get hot or cold at the same time.... In the latest part, the author hypothesizes that a team is better off with steady (low volatility) hitters. He plans to test this in a future supplement.

Very interesting stuff.....

kaldaniels
09-07-2011, 06:36 PM
Using the measure that they have created, which factors in wOBA, ISO, BB%, K%, you get a volatility index. Lower volatility scores are around .5 or .6, and high scores (which equal high volatility) are over 1.0. On part IV of the series of articles (http://www.beyondtheboxscore.com/2011/7/8/2251452/is-volatility-an-attribute-player-volatility-part-iv), it lists the most and least volatile hitters over the 2007-09 time period. Their chart shows that the most volatile player was Bengie Molina in 2009, with a 1.09 score.

Using their online calculator, and putting Bruce's 2011 numbers in, the result is 1.05, which is very volatile. I then ran Drew Stubb's 2011 numbers, and his result is 1.146, even higher.

By comparison, Joey Votto is .576, Brandon Phillips is .883, Paul Janish is 1.344(!), Edgar Renteria is 1.087, Scott Rolen is 1.176, Chris Heisey is 1.149, Ryan Hanigan is .858. and Ramon Hernandez is .924. That is all that I had time to run.

If there is anyuthing to all of these numbers, it would show that the Reds have LOTS of very streaky hitters, with Bruce, Renteria, Stubbs, Heisey, Rolen, and Janish all being in the very high range on volatility.

In fact this may help to explain why we have such a big difference between actual wins and pythag wins. When hitters get hot, they score lots of runs, and win games by larger margins. When they get cold, the team doesn't score enough runs. Of course, they ususally don't all get hot or cold at the same time.... In the latest part, the author hypothesizes that a team is better off with steady (low volatility) hitters. He plans to test this in a future supplement.

Very interesting stuff.....

If true though, wouldn't the Reds have a totally different runs scored distrubution than the other top offenses in the league?

Scrap Irony
09-07-2011, 06:45 PM
So the short answer to this would be:

Yes. Remarkably so?

dougdirt
09-07-2011, 06:46 PM
The short answer may be "yes this year". For the most part, he was pretty consistent last season.

Scrap Irony
09-07-2011, 06:49 PM
Really? I'm checking his numbers from 2010 and, so far, at least, that's not the case.

dougdirt
09-07-2011, 06:53 PM
Really? I'm checking his numbers from 2010 and, so far, at least, that's not the case.

Well, I am certainly only going on month to month OPS in 2010, but he was at .800+ in every month but one. Seems pretty consistent to me. Though I wonder if two monster months to end the season cause a volatile score that looks worse than it is because those two months make 3 other .800+ OPS months look inconsistent?

Scrap Irony
09-07-2011, 07:03 PM
10 game sets, I've got him at 0.99 in 2010. Then again, that was all done with calculator on the computer, so I could be wrong.

RedsManRick
09-07-2011, 10:59 PM
So the short answer to this would be:

Yes. Remarkably so?

THere are two questions:

Has Jay Bruce been streaky in 2011? Is streakiness a "skill" that Jay Bruce possesses (and thus is likely to continue)? These are not the same thing.

wlf WV
09-08-2011, 12:27 PM
I would be interested to see our volatility compared to the other offenses similar to us.

RedsManRick
09-09-2011, 04:01 PM
I LOVE this article - it does exactly what we've been trying to do. For me, this really stood out:
Volatility isn't so much an attribute as an observable implication of hitters with a certain mix of talents/attributes.

So the answer appears to my questions appear to be:

1) Yes, Jay Bruce has been particularly streaky
2) Sort of. Jay Bruce's skill set lends itself to streakiness.

Additionally, this struck me as completely intuitive:
The attribute with the largest affect after wOBA was isolated power (ISO). What was surprising was that it had a positive correlation to Volatility, meaning that the higher one's ISO the higher their Volatility.

If a guy hits for a lot of power but not a high average, he's spreading say, 70 extra base hits across 600 PA and not flattening that out with a lot of singles. That's a relative small number of high value events spread out. There are bound to be significant clumps of performance by random variation alone.

And then,
The second largest affect after ISO was BB%. The more patient a hitter and the better they controlled the zone the lower their Volatility.

Again, completely intuitive. Walks give you another frequent but small positive contribution.

I think of it like roulette. Some guys put big chunks of their stack on a small set of numbers. Other guys put lots of small chunks of their stack out. The first guy is going to see much bigger swings in his chip stack than the second guy. It just makes sense.

Of course, the trick is remembering that in the long run, the chip stacks will grow at the same rate. In the long run, it's the amount of chips you start with, not how you distribute them, that determines who leaves the table with the biggest stack.

edabbs44
09-09-2011, 09:54 PM
So wait, now he is streaky and it isn't just the way the season is being divvied up?

RedsManRick
09-09-2011, 11:05 PM
So wait, now he is streaky and it isn't just the way the season is being divvied up?

All I said is that I wanted better evidence. I'm always willing to change my position based on what a credible argument backed by rigorous evidence shows.

This author actually defined consistency in a rigorous way and analyzed the data objectively.

Patrick Bateman
09-09-2011, 11:25 PM
I don't think a lot of people argued that he wasn't streaky, as much as they argued that it didn't matter.

Has that been resolved yet?