Lead and Antimony Medal Awards for lousy charts

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.

Lead and Antimony Medal Awards for lousy charts

Postby nisiprius » Thu May 09, 2013 9:12 pm

Generally, the quality of financial charts is pretty bad judged by... anyone who uses any other kind of chart.

But these deserve a prize. The Lead Medal for Who Cares About Numbers As Long As The Zigzag Line Goes Up Goes Up goes to this chart:

Image

Question: is the vertical axis

a) a linear scale with a suppressed zero, or
b) a logarithmic scale?

:annoyed :? :)

Source: http://www.dfaus.com/pdf/fact_sheets/us ... 1_port.pdf

And, the runner-up; the Antimony Medal for Strangest Choice Of Tick Values goes to [sources omitted from original posting, now added]
Putnam Investments Multicap Growth. Source: https://www.putnam.com/individual/ and https://content.putnam.com/prodinfo/cha ... A_LIFE.jpg.

Image
Last edited by nisiprius on Fri May 10, 2013 9:53 am, edited 2 times in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
nisiprius
Advisory Board
 
Posts: 24274
Joined: 26 Jul 2007
Location: "Citizen of the terrestrial sphere"--O. Henry, "A Cosmopolite in a Cafe"

Re: Lead and Antimony Medal Awards for lousy charts

Postby livesoft » Thu May 09, 2013 9:19 pm

Boy, they are making one think, aren't they?

The first chart is likely to be linear. Reason: The leftmost values are listed in the line labels: $153K or so.

The second chart has labels on the left for the high value and the ending value along with horizontal lines marking this levels. Sure is strange though.
It's all about short-term opportunistic rebalancing due to a short-term change in one's asset allocation, uh, I mean opportunistic rebalancing, uh I mean rebalancing, uh I mean market timing.
livesoft
 
Posts: 30781
Joined: 1 Mar 2007

Re: Lead and Antimony Medal Awards for lousy charts

Postby LadyGeek » Thu May 09, 2013 9:25 pm

What's the source of the 2nd chart?

As for the 1st chart, the minimum Y-value (X-axis intercept) doesn't need to be 0. It's just offset and scaled to present appropriately. I don't see anything wrong here.
To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
LadyGeek
Site Admin
 
Posts: 16012
Joined: 20 Dec 2008
Location: Philadelphia

Re: Lead and Antimony Medal Awards for lousy charts

Postby k66 » Fri May 10, 2013 12:48 am

There's also something a little hinky with the abscissa in Chart #1. The first tick is marked 9/2005 while the next is 2007 (presumed to be 1/2007) making the first interval about 1.25 years. Yet the next two intervals, of comparable size to the first, are 2 years each and the final interval, which is much smaller than any of the others, is again 1.25 years. Neither logarithmic nor a normal scale is apparently in use on this axis... perhaps it just "charter's choice".

Chart #2, while not incorrect, is certainly hard to visually incorporate. One presumes that it is the result of allowing unsophisticated software to slice the rudimentary data with no concern for creating neat or regular intervals.

As one who spends a great deal of time dealing with (non-financial) charts in daily work, I would concur that those produced by the investment industry are decidedly difficult to read. It is also perhaps the deliberate intention of these chart producers to create unintelligible information so as to not allow investors the benefit of the real data.
User avatar
k66
 
Posts: 108
Joined: 27 Oct 2012

Re: Lead and Antimony Medal Awards for lousy charts

Postby NightOwl » Fri May 10, 2013 2:17 am

Ah, this brings back memories. Not particularly happy memories, but memories nonetheless.

In what seems now like another life, while I was in graduate school I had a temp job working 6 pm to 4 am helping investment banking analysts prepare their pitchbooks -- that tells you how long ago this was, because those jobs moved overseas a long time ago.

One night at around 3 am, an exhausted-looking junior analyst asked me to help him with an Excel chart in his deck. The data was all plotted on a hidden second y-axis to get the shape of the data right, while the visible y-axis was labeled with inflated values. Even as a poor graduate student who really needed that job, I took one look at that chart and said "I'm sorry, I can't help you with that."

I am highly confident, by the way, that no client ever saw that chart -- he was trying to trick his bosses. I worked with many analysts who were obviously in waaay over their heads and who washed out very quickly. Here today, gone tomorrow.

NightOwl
"Volatility provokes the constant dread that some investors know more than we do, making us fearful of ignoring such powerful price movements." | Peter Bernstein, "The 60/40 Solution."
User avatar
NightOwl
 
Posts: 625
Joined: 6 Feb 2009
Location: New York, NY

Re: Lead and Antimony Medal Awards for lousy charts

Postby nisiprius » Fri May 10, 2013 6:38 am

LadyGeek wrote:As for the 1st chart, the minimum Y-value (X-axis intercept) doesn't need to be 0. It's just offset and scaled to present appropriately. I don't see anything wrong here.
I think not being able to tell whether it's linear or log when the range of values is that large is wrong. And I think charts where you can't determine the maximum and minimum values are wrong, i.e. they ought to be scaled so that they end on a labeled tick mark, or at least a tick mark whose value can be determined.
What's the source of the 2nd chart?
Oops, sorry. Putnam Investments Multicap Growth. Can't find a direct link to the chart page, you have to select from a popup, but:

https://www.putnam.com/individual/

Aha, you can "email" the page to yourself. Oho, I did but it never arrived. But I can link to the actual image,
https://content.putnam.com/prodinfo/cha ... A_LIFE.jpg

And how about the Bismuth Medal for Illustrating Why You Might Sometimes Want a Log Scale, to Putnam's growth chart for the George Putnam Balanced Fund:

Image

Pretty cool, actually; I must add that to my list of "funds that have been around a while." (Relax, Vanguard fans, Wellington beats it from the same starting point... and there is some very unpleasant recent stuff in the Putnam growth chart, like dropping about 50% during 2008-2009--in a balanced fund? and not recovering).
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
nisiprius
Advisory Board
 
Posts: 24274
Joined: 26 Jul 2007
Location: "Citizen of the terrestrial sphere"--O. Henry, "A Cosmopolite in a Cafe"

Re: Lead and Antimony Medal Awards for lousy charts

Postby LadyGeek » Fri May 10, 2013 9:46 am

nisiprius wrote:
What's the source of the 2nd chart?
Oops, sorry. Putnam Investments Multicap Growth. Can't find a direct link to the chart page, you have to select from a popup, but: https://www.putnam.com/individual/

Aha, you can "email" the page to yourself. Oho, I did but it never arrived. But I can link to the actual image, https://content.putnam.com/prodinfo/cha ... A_LIFE.jpg

Sorry, that's not the right one. Here's what you linked, it's from the The Putnam Fund for Growth and Income (The filename contains the fund ID number, 37.)

Image
Putnam's website creates their pages using a convoluted set of internal applications and scripts. You can't get a direct URL to a specific fund, which does not seem like a good approach (note constant churning of jsessionid in the URL as the page is created). Here's the top-level page: Putnam Mutual Funds.

For Putnam Multi-Cap Growth Fund, I get:

Image
They really should embedded a title inside those charts images, along with proper Y-axis labeling. There's no way to tell if this is the right chart except by the filename (there could be an error there as well). If their website designers don't catch a mismatch (quality assurance, I hope), the investors will.

They might put a lot of work into their product line, but the first thing their customers see is a poor quality chart. “You never get a second chance to make a first impression.”
To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
LadyGeek
Site Admin
 
Posts: 16012
Joined: 20 Dec 2008
Location: Philadelphia

Re: Lead and Antimony Medal Awards for lousy charts

Postby nisiprius » Fri May 10, 2013 9:59 am

LadyGeek wrote:Sorry, that's not the right one.
:oops: OK, I fixed it. I think.
They really should embed a title inside those charts images, along with proper Y-axis labeling.
To the investment industry, charts appear to be regarded as persuasion, not information.

I don't know if it's a legitimate complaint, but it seems odd to me to have the labelled ordinate be at the right end of the graph instead of the left. I don't know if that's just techie bias on my part.

Maybe we really should design a formal scoring system, score websites and hand out awards.

:?: points for a sensible number of tick marks, labels, and grid lines, all set at sensible round-number values
:?: points for the growth chart being right there among the things you see first. Minus points if they're not in any obvious place, like a "performance" tab. More minus if you have to download a PDF ("factsheet," annual report, etc.) to find one.
:?: points for a logarithmic Y axis. (Selectable option is OK)
:?: points for letting you choose more than one fund to plot on the chart.
:?: points for letting you include other fund companies' funds in that comparison.
:?: points for going back to inception and not just 10 years.
:?: points for dynamic readout of numeric values by moving the mouse.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
nisiprius
Advisory Board
 
Posts: 24274
Joined: 26 Jul 2007
Location: "Citizen of the terrestrial sphere"--O. Henry, "A Cosmopolite in a Cafe"

Re: Lead and Antimony Medal Awards for lousy charts

Postby nisiprius » Fri May 10, 2013 10:25 am

And the molybdenum medal goes to Franklin Growth Fund, for not having any value labels on the Y axis at all. $10,000! $5,008,342! Ziggy line starts at nothing and goes up, up, up! What more would you want to know?

Image
"He had brought a large map representing the sea
Without the least vestige of land;
And the crew was quite pleased when they found it to be
A map they could all understand.

'What's the use of Mercator's North Poles and Equators,
Tropics, zones, and meridian lines?
So the captain would cry, and the crew would reply
'They are merely conventional signs.'"--Lewis Carroll, The Hunting of the Snark.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
nisiprius
Advisory Board
 
Posts: 24274
Joined: 26 Jul 2007
Location: "Citizen of the terrestrial sphere"--O. Henry, "A Cosmopolite in a Cafe"

Re: Lead and Antimony Medal Awards for lousy charts

Postby jsl11 » Fri May 10, 2013 11:38 am

The Metallurgy Award goes to Nisiprius for knowing what antimony, bismuth, and molybdenum are.

Jeff
jsl11
 
Posts: 2915
Joined: 27 Feb 2007
Location: Cleveland, OH

Re: Lead and Antimony Medal Awards for lousy charts

Postby momar » Fri May 10, 2013 11:42 am

I don't appreciate the badmouthing of Antimony and especially Bismuth. These can form beautiful crystals, especially Bismuth, and anyone would be happy to win a medal made of them.

Bismuth
Image

Antimony (actually Stibnite, Sb2S3)
Image
"Index funds have a place in your portfolio, but you'll never beat the index with them." - Words of wisdom from a Fidelity rep
User avatar
momar
 
Posts: 1359
Joined: 13 Nov 2011

Re: Lead and Antimony Medal Awards for lousy charts

Postby Chan_va » Fri May 10, 2013 11:45 am

Hey, why slap down these metals?

Lead and antimony do far more for society than the precious metals.
Chan_va
 
Posts: 281
Joined: 5 Dec 2012

Re: Lead and Antimony Medal Awards for lousy charts

Postby Oicuryy » Fri May 10, 2013 1:05 pm

Does anyone else see anything wrong with this graph from In the Vanguard?

Image

Ron
Money is fungible | Abbreviations and Acronyms
User avatar
Oicuryy
 
Posts: 801
Joined: 22 Feb 2007

Re: Lead and Antimony Medal Awards for lousy charts

Postby LadyGeek » Fri May 10, 2013 2:01 pm

1. The reference to "1989 = 100" is not on the chart. I think the label is a short-cut because they didn't get the scaling correct and were in a hurry to publish.
2. The left and right side axes are the same.
3. There are no labels on the Y-axes.
4. There should be a legend indicating which axis (left or right) the plot refers to.
5. Which S&P 500 index is used; Total Return or Price?
To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
LadyGeek
Site Admin
 
Posts: 16012
Joined: 20 Dec 2008
Location: Philadelphia

Re: Lead and Antimony Medal Awards for lousy charts

Postby Oicuryy » Fri May 10, 2013 11:30 pm

Compare this graph to the one I posted above.

Image

Ron
Money is fungible | Abbreviations and Acronyms
User avatar
Oicuryy
 
Posts: 801
Joined: 22 Feb 2007

Re: Lead and Antimony Medal Awards for lousy charts

Postby Valuethinker » Sat May 11, 2013 3:27 am

Chan_va wrote:Hey, why slap down these metals?

Lead and antimony do far more for society than the precious metals.


Not wanting to start a debate about this, other than from an informational point of view ('benefits to society of using lead') but
I am struggling to find a *useful* contribution for lead.

Lead acid batteries to be sure.

But the harm in lead, in particular lead in gasoline and in paint, has been a huge societal deficit in the 20th century (not to mention what lead pipes did to the Romans).

[OT environmental effects and policy comments removed by admin LadyGeek]

Lead is like asbsetos. I don't think human civilization has, overall, been advanced by its discovery and widespread use.
Valuethinker
 
Posts: 23715
Joined: 11 May 2007

Re: Lead and Antimony Medal Awards for lousy charts

Postby nisiprius » Sat May 11, 2013 6:19 am

Oicuryy wrote:Compare this graph to the one I posted above.

Image

Ron
Ron, I think I will need a hint on this, as I'm not sure what you see as the problem other than the use of a linear scale when a log scale might be more appropriate. The Vanguard chart is sloppy in many ways. But to me, I guess the principle of the thing is whether it is possible to make rough numeric quantitative judgements, comparing values on the chart by reading where the curve falls on the scale, to the degree of precision represented by the chart resolution and the ability to sight horizontally by eyeball.

Of course the use of a price index instead of a total return index is somewhat misleading--on the S&P if total return were used, the 2007 peak would be about 10% higher than the 2000 peak, for example.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
nisiprius
Advisory Board
 
Posts: 24274
Joined: 26 Jul 2007
Location: "Citizen of the terrestrial sphere"--O. Henry, "A Cosmopolite in a Cafe"

Re: Lead and Antimony Medal Awards for lousy charts

Postby LadyGeek » Sat May 11, 2013 9:09 am

Oicuryy wrote:Compare this graph to the one I posted above.

Thanks for going back to the data source. Where is the plot from?

The plot lines are not lined up vertically with the chart. IOW, 1989 starts around 90 - not 100 as Vanguard stated. The fact that Vanguard stated 1989 is referenced to 100 is incorrect and misleading.

That goes to my earlier comment in which they rushed the chart out rather than fix it, i.e. add a label and hide 1989 instead of fixing the analysis. So, the investors caught it - or it slipped past everyone except Oicuryy.
To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
LadyGeek
Site Admin
 
Posts: 16012
Joined: 20 Dec 2008
Location: Philadelphia

Re: Lead and Antimony Medal Awards for lousy charts

Postby Habarnam » Sat May 11, 2013 12:08 pm

Valuethinker wrote:Not wanting to start a debate about this, other than from an informational point of view ('benefits to society of using lead') but
I am struggling to find a *useful* contribution for lead.
[snip]
Lead is like asbsetos. I don't think human civilization has, overall, been advanced by its discovery and widespread use.


I don't know enough about metallurgy but I'm pretty sure the malleability of lead was fairly important to the early printing presses. And the books were of significant importance to the development of Western civilization and its dominance for the past 400 years.

Definitely not in the same class as asbestos.
Habarnam
 
Posts: 8
Joined: 17 Sep 2012

Re: Lead and Antimony Medal Awards for lousy charts

Postby baw703916 » Sat May 11, 2013 12:19 pm

momar wrote:I don't appreciate the badmouthing of Antimony and especially Bismuth.


Here's a recently-discovered scientific fact: bismuth is radioactive.*

So enjoy your bismuth medal now--it won't be there forever. (I do have a cool-looking piece of bismuth like the picture).



*the half-life is about a million times greater than the age of the universe--so no need to be too alarmed.
Most of my posts assume no behavioral errors.
User avatar
baw703916
 
Posts: 5335
Joined: 1 Apr 2007
Location: Northern Virginia

Re: Lead and Antimony Medal Awards for lousy charts

Postby nisiprius » Sat May 11, 2013 12:20 pm

Habarnam wrote:
Valuethinker wrote:Not wanting to start a debate about this, other than from an informational point of view ('benefits to society of using lead') but
I am struggling to find a *useful* contribution for lead.[snip]Lead is like asbestos. I don't think human civilization has, overall, been advanced by its discovery and widespread use.
I don't know enough about metallurgy but I'm pretty sure the malleability of lead was fairly important to the early printing presses. And the books were of significant importance to the development of Western civilization and its dominance for the past 400 years....
I was going to ask whether transistors would have been developed without the inspiration of the galena crystal. (Galena crystals with point contacts were natural rectifiers used in "crystal radios." But in fact, perhaps, a galena-based transistor was developed in 1933--but overlooked.

Of course, lead is used in bullets--I am carefully going to express no opinion whatsoever on whether bullets have contributed to civilization--but I imagine that if lead had not been available, there would have been tin bullets or zinc bullets or frozen quicksilver bullets or something.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
nisiprius
Advisory Board
 
Posts: 24274
Joined: 26 Jul 2007
Location: "Citizen of the terrestrial sphere"--O. Henry, "A Cosmopolite in a Cafe"

Re: Lead and Antimony Medal Awards for lousy charts

Postby momar » Sat May 11, 2013 12:32 pm

Lead is useful as a component in compound semiconductors (such as Galena - PbS - and PbSe) that are used in IR detectors.
"Index funds have a place in your portfolio, but you'll never beat the index with them." - Words of wisdom from a Fidelity rep
User avatar
momar
 
Posts: 1359
Joined: 13 Nov 2011

Re: Lead and Antimony Medal Awards for lousy charts

Postby baw703916 » Sat May 11, 2013 12:35 pm

nisiprius wrote: I was going to ask whether transistors would have been developed without the inspiration of the galena crystal. (Galena crystals with point contacts were natural rectifiers used in "crystal radios." But in fact, perhaps, a galena-based transistor was developed in 1933--but overlooked.

Of course, lead is used in bullets--I am carefully going to express no opinion whatsoever on whether bullets have contributed to civilization--but I imagine that if lead had not been available, there would have been tin bullets or zinc bullets or frozen quicksilver bullets or something.


PbS and PbSe are quite good infrared photovoltaic materials, often used for thermal imaging (as momar just said simultaneously with this post).

For bullets, lead has a combination of a high density (so the projectile can carry more momentum) as well as being rather malleable, which means it can withstand the shock wave of the combustion chamber without shattering, and will deform itself to follow the rifling of the barrel, so it can have spin imparted and follow a more predictable trajectory. If cost were no object, gold would work even better in these two ways. :)
Most of my posts assume no behavioral errors.
User avatar
baw703916
 
Posts: 5335
Joined: 1 Apr 2007
Location: Northern Virginia

Re: Lead and Antimony Medal Awards for lousy charts

Postby momar » Sat May 11, 2013 12:37 pm

baw703916 wrote:For bullets, lead has a combination of a high density (so the projectile can carry more momentum) as well as being rather malleable, which means it can withstand the shock wave of the combustion chamber without shattering, and will deform itself to follow the rifling of the barrel, so it can have spin imparted and follow a more predictable trajectory. If cost were no object, gold would work even better in these two ways. :)

So wait, all those people ridiculing gold bugs with "when society collapses, you'll want bullets and food" need to revise their statement?
"Index funds have a place in your portfolio, but you'll never beat the index with them." - Words of wisdom from a Fidelity rep
User avatar
momar
 
Posts: 1359
Joined: 13 Nov 2011

Re: Lead and Antimony Medal Awards for lousy charts

Postby LadyGeek » Sat May 11, 2013 2:04 pm

Getting back on-topic, does anyone have comments on Oicuryy's chart?

(Please stay on-topic.)
To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
LadyGeek
Site Admin
 
Posts: 16012
Joined: 20 Dec 2008
Location: Philadelphia

Re: Lead and Antimony Medal Awards for lousy charts

Postby umfundi » Sat May 11, 2013 2:19 pm

Leaded glass windows? A good use for lead.

Again, I am drawn to Edward Tufte's books, including "The Visual Display of Quantitative Information". Classics.

Which brings to mind Peter Norvig's classic on the Gettysburg Address in Powerpoint.

http://norvig.com/Gettysburg/

Keith
Déjà Vu is not a prediction
umfundi
 
Posts: 3361
Joined: 7 Jun 2011

Re: Lead and Antimony Medal Awards for lousy charts

Postby Oicuryy » Sat May 11, 2013 4:16 pm

nisiprius wrote:Ron, I think I will need a hint on this, as I'm not sure what you see as the problem other than the use of a linear scale when a log scale might be more appropriate. The Vanguard chart is sloppy in many ways. But to me, I guess the principle of the thing is whether it is possible to make rough numeric quantitative judgements, comparing values on the chart by reading where the curve falls on the scale, to the degree of precision represented by the chart resolution and the ability to sight horizontally by eyeball.

Yes, the problem I see is the use of a linear scale. IMO, the picture should tell the story. The viewer should not have to convert the picture into numbers then do calculations on the numbers to see what happened.

Vanguard's graph does not show actual index values. They scaled the indexes to 100 at year-end 1989 to show the relative performance since then. The Nikkei fell to one-fourth its value. The S&P rose to four times its value. A quartering and a quadrupling are basically the same thing in opposite directions. They should travel the same vertical distance in the picture but in opposite directions.

Suppose Vanguard had chosen to scale the indexes to 100 at year-end 2010. Then their graph would have looked something like this. The data and the y-axis scale are the same but the visual impression of the history of the indexes is much different.

Image

But with a log scale on the y-axis the visual impression stays the same.

Image

LadyGeek wrote:Thanks for going back to the data source. Where is the plot from?

It's from the summer 2011 issue of In the Vanguard.
https://personal.vanguard.com/pdf/itvsummer2011.pdf

Ron
Money is fungible | Abbreviations and Acronyms
User avatar
Oicuryy
 
Posts: 801
Joined: 22 Feb 2007

Re: Lead and Antimony Medal Awards for lousy charts

Postby #Cruncher » Sun May 12, 2013 8:38 am

Oicuryy wrote:IMO, the picture should tell the story. ... A quartering and a quadrupling are basically the same thing in opposite directions.
I guess it depends on what you think the story is. To me a quartering and a quadrupling are not basically the same thing in reverse. Assume I put $10,000 into a fund that quadruples and another $10,000 into one that falls 75%. One would increase to $40,000 and the other fall to $2,500. A gain of $30,000 and a loss of $7,500 don't seem like "the same thing in opposite directions".

In my opinion, the Vanguard graph which you disparage does tell that story (where the S&P 500 and Nikkei represent the two funds). Graphs that change the 100 base from the beginning to the end and / or use a log scale do not.
Oicuryy wrote:Image
User avatar
#Cruncher
 
Posts: 1302
Joined: 14 May 2010
Location: New York City

Re: Lead and Antimony Medal Awards for lousy charts

Postby nisiprius » Sun May 12, 2013 1:46 pm

Oicuryy wrote:...IMO, the picture should tell the story....
Apologies for quibbling here but I would have said that I want charts to be "visual displays of quantitative information," not storytelling. I do not want them to "tell stories!" One of my quarrels with the financial industry is that charts are seen as vehicles for storytelling. As a source of power in making points.

I personally have a very very strong preference for semilog charts for many reasons, but in some cases I think they can be used to tell stories where a linear chart might present information more accurately. There's one big problem with semilog charts that I don't know how to solve, which is the problem of making comparisons if a) the two charts cover very different lengths of time, or if b) they cover very different ranges on the Y axis.

I've thought it should probably be a standard feature of financial semilog charts to have a few well-positioned straight lines represent round-number percentage growth rates.

I've also thought that probably the most helpful axis graduation would involve factors of 2, since "doubling your money" is such a natural way of thinking about financial growth.

One would like to have any mutual fund chart presented in such away that one could instantly and intuitively tell at a glance whether, during 2008-2009, the value of the fund fell by "about 50%," "about 25%," "about 12.5%", "about 6%"...

Parenthetically, it seems to me that nobody has combined the power of electronic media with the insight of someone like Tufte. Tufte, for all his admirable virtues, deals primarily with stationary charts, ink on paper. What a pity that you can't have a pair of growth charts on the screen and dynamically slide one of them up or down with respect to the other, for example.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
nisiprius
Advisory Board
 
Posts: 24274
Joined: 26 Jul 2007
Location: "Citizen of the terrestrial sphere"--O. Henry, "A Cosmopolite in a Cafe"

Re: Lead and Antimony Medal Awards for lousy charts

Postby Oicuryy » Sun May 12, 2013 4:26 pm

#Cruncher wrote:I guess it depends on what you think the story is. To me a quartering and a quadrupling are not basically the same thing in reverse. Assume I put $10,000 into a fund that quadruples and another $10,000 into one that falls 75%. One would increase to $40,000 and the other fall to $2,500. A gain of $30,000 and a loss of $7,500 don't seem like "the same thing in opposite directions".

I guess it depends on whether you measure gains and losses in dollars or in percent. I think the story is percent change. It takes a 300% gain to offset a 75% loss. That is why I say a quadrupling is the reverse of a quartering.

Ron
Money is fungible | Abbreviations and Acronyms
User avatar
Oicuryy
 
Posts: 801
Joined: 22 Feb 2007

Re: Lead and Antimony Medal Awards for lousy charts

Postby nisiprius » Sun May 12, 2013 4:47 pm

Ron, I'm 90% with you and certainly to anyone who is used to interpreting semilog charts they are usually the best presentation of financial data. But there is one aspect that bothers me.

Linear charts have one property that semilog charts do not: on a linear chart, provided the zero isn't suppressed, if you double the number of dollars, the distance on the graph doubles, and that's independent of the chart scaling.

On semilog charts, doubling the number of dollars may look big or small depending on the chart scaling, and certainly there is a problem with those two-century stocks-for-the-long-run type charts on which 1987 is hardly visible and the Great Depression is just a little wiggle.

In fact, I'd say there's a problem with semilog charts in that no semilog chart gives you a good visual impression of what a 50% drop in the market really means, let alone what an 87% drop in the market, circa 1929, means.

If you were king of the world, or at least emperor of financial chart presentations, what would you suggest as an unequivocal presentation of vertical scale in semilog charts to make sure that one stayed oriented to "how much is doubling?" Maybe light green bars on every chart, lightly shading in 1 to 2, 4 to 8, 16 to 32, etc.?

There's no perfect answer. Sometimes a Mercator projection is best! (Parenthetically: when I was in grade school, there was a vogue using equal-area projection on classroom maps of the world, with everything in accurate relative size but everything outside the Americas having its shapes stretched and sheered to the point of being unrecognizable. From my brief forays into modern high schools it appears as if things have shifted back, to Mercator or Mercator-like projections, in which everything is recognizable but Greenland is as big as South America...)
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
nisiprius
Advisory Board
 
Posts: 24274
Joined: 26 Jul 2007
Location: "Citizen of the terrestrial sphere"--O. Henry, "A Cosmopolite in a Cafe"

Re: Lead and Antimony Medal Awards for lousy charts

Postby Oicuryy » Sun May 12, 2013 7:55 pm

nisiprius wrote:Ron, I'm 90% with you

I'll take 90%!

nisiprius wrote:Linear charts have one property that semilog charts do not: on a linear chart, provided the zero isn't suppressed, if you double the number of dollars, the distance on the graph doubles, and that's independent of the chart scaling.

It took me a minute to understand what you are saying. One of my main gripes with linear graphs is that a doubling from 400 to 800 covers four times the distance of a doubling from 100 to 200.

But it is true that the distance from 0 to 800 is double the distance from 0 to 400 and the distance from 0 to 200 is double the distance from 0 to 100. Vanguard's graph does show that both the S&P and the Nikkei roughly doubled from 2003 to 2007. But you have to look carefully to see it. The S&P's doubling covered nine times the distance on the graph that the Nikkei's doubling did.

nisiprius wrote:On semilog charts, doubling the number of dollars may look big or small depending on the chart scaling, and certainly there is a problem with those two-century stocks-for-the-long-run type charts on which 1987 is hardly visible and the Great Depression is just a little wiggle.

In fact, I'd say there's a problem with semilog charts in that no semilog chart gives you a good visual impression of what a 50% drop in the market really means, let alone what an 87% drop in the market, circa 1929, means.

Now you're pulling my leg. Can you even see the Great Depression on this linear graph from measuringworth.com?

Image

nisiprius wrote:If you were king of the world, or at least emperor of financial chart presentations, what would you suggest as an unequivocal presentation of vertical scale in semilog charts to make sure that one stayed oriented to "how much is doubling?" Maybe light green bars on every chart, lightly shading in 1 to 2, 4 to 8, 16 to 32, etc.?

I sometimes set the gridline spacing to log(2), sometimes to log(1.1) or log(1.5). Only on very long timeframes is the standard log(10) spacing suitable. The idea is to have the distance from one gridline to the next represent a percent increase that is both appropriate for the data and meaningful for the viewer (10%, 50%, 100%, 900%).

If I was czar of financial graphs I would first decree that all time series be drawn with a log scale on the y-axis with equally spaced gridlines. I would decree that all graphing software, especially Excel, allow the user to set the gridline spacing. I might require a vertical line in the margin of the graph. The length of the line would be the distance of a doubling; like the mileage scale on a map. (You mean Greenland is not as big as South America?)

Ron
Money is fungible | Abbreviations and Acronyms
User avatar
Oicuryy
 
Posts: 801
Joined: 22 Feb 2007

Re: Lead and Antimony Medal Awards for lousy charts

Postby Dulocracy » Mon May 13, 2013 10:42 am

I broke down your statement into the important parts for you:

baw703916 wrote:
Here's a recently-discovered scientific fact: bismuth is radioactive.

(I do have a cool-looking piece of bismuth like the picture).
I'm not a financial professional. Post is info only & not legal advice. No attorney-client relationship exists with reader. Scrutinize my ideas as if you spoke with a guy at a bar. I may be wrong.
Dulocracy
 
Posts: 623
Joined: 27 Feb 2013
Location: Atlanta, GA

Re: Lead and Antimony Medal Awards for lousy charts

Postby Epsilon Delta » Mon May 13, 2013 11:40 am

Dulocracy wrote:I broke down your statement into the important parts for you:

baw703916 wrote:
Here's a recently-discovered scientific fact: bismuth is radioactive.

(I do have a cool-looking piece of bismuth like the picture).


Detecting the bismuth decay required a cutting edge detector. Pure bismuth is significantly less radioactive that many common objects, including granite counter tops and human beings. You emit at least 20,000 times more radiation than would your body weight of pure bismuth.
User avatar
Epsilon Delta
 
Posts: 3077
Joined: 28 Apr 2011

Re: Lead and Antimony Medal Awards for lousy charts

Postby frugaltype » Mon May 13, 2013 12:34 pm

Valuethinker wrote:I am struggling to find a *useful* contribution for lead.


Shielding from radiation. What do you think is inside that thing they drape over you during dental x-rays.
User avatar
frugaltype
 
Posts: 1952
Joined: 24 Apr 2013

Re: Lead and Antimony Medal Awards for lousy charts

Postby Epsilon Delta » Mon May 13, 2013 12:58 pm

frugaltype wrote:
Valuethinker wrote:I am struggling to find a *useful* contribution for lead.


Shielding from radiation. What do you think is inside that thing they drape over you during dental x-rays.


Darn near the entire development of electronics depended on lead. Sure we can replace lead now we have a developed industry, with electronically controlled solder machines, but the original development would have been much harder with leadless solder.
User avatar
Epsilon Delta
 
Posts: 3077
Joined: 28 Apr 2011

Re: Lead and Antimony Medal Awards for lousy charts

Postby LadyGeek » Mon May 13, 2013 4:36 pm

Any more comments on charts (vs. material properties and applications)?
To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
LadyGeek
Site Admin
 
Posts: 16012
Joined: 20 Dec 2008
Location: Philadelphia

Re: Lead and Antimony Medal Awards for lousy charts

Postby umfundi » Mon May 13, 2013 5:25 pm

LadyGeek wrote:Any more comments on charts (vs. material properties and applications)?

Well,

Yes. Let's start with the observation that all financial charts are lousy. Does anyone have nominations for any good ones?

Engineering and simulation (particularly fluid dynamics) have long been at the forefront of visualizing 3-dimensional time dependent data, i.e., data in 4 dimensions. Some of this has spilled over into visual analytics for big data, but the graphics used on all financial sites that I know is just embarrassing.

Here's a good example of innovation in looking at and analyzing data:

http://blogs.hbr.org/cs/2013/04/visuali ... ocess.html

Keith
Déjà Vu is not a prediction
umfundi
 
Posts: 3361
Joined: 7 Jun 2011

Re: Lead and Antimony Medal Awards for lousy charts

Postby nisiprius » Mon May 13, 2013 6:49 pm

I'd love to see a three-dimensional dynamically rotatable presentation of all the stocks listed on the New York Stock Exchange, plotted according to their Fama-French factor loadings.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
nisiprius
Advisory Board
 
Posts: 24274
Joined: 26 Jul 2007
Location: "Citizen of the terrestrial sphere"--O. Henry, "A Cosmopolite in a Cafe"

Re: Lead and Antimony Medal Awards for lousy charts

Postby umfundi » Mon May 13, 2013 7:01 pm

nisiprius wrote:I'd love to see a three-dimensional dynamically rotatable presentation of all the stocks listed on the New York Stock Exchange, plotted according to their Fama-French factor loadings.
Well, let's see. Something like Ensight http://www.ceisoftware.com/ensight10/what-is-ensight/ can plot 3D data, animate or dynamically slice it for 4D and provide false color for 5D.

Do you have the data you'd like to visualize?

Keith
Déjà Vu is not a prediction
umfundi
 
Posts: 3361
Joined: 7 Jun 2011

Re: Lead and Antimony Medal Awards for lousy charts

Postby nisiprius » Mon May 13, 2013 8:11 pm

umfundi wrote:
nisiprius wrote:I'd love to see a three-dimensional dynamically rotatable presentation of all the stocks listed on the New York Stock Exchange, plotted according to their Fama-French factor loadings.
Well, let's see. Something like Ensight http://www.ceisoftware.com/ensight10/what-is-ensight/ can plot 3D data, animate or dynamically slice it for 4D and provide false color for 5D.

Do you have the data you'd like to visualize?

Keith
No. That's another problem with financial data--most of it has a fairly high paywall around it. Expensive enough that I haven't bothered to cost it out exactly, but thousands of dollars anyway for stuff like CRSP or Morningstar Principia Pro.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
nisiprius
Advisory Board
 
Posts: 24274
Joined: 26 Jul 2007
Location: "Citizen of the terrestrial sphere"--O. Henry, "A Cosmopolite in a Cafe"


Return to Investing - Theory, News & General

Who is online

Users browsing this forum: 22twain, Bing [Bot], bvp, Deceptor, Devil's Advocate, FAST Enterprise [Crawler], packer16, roymeo, TO, Yahoo [Bot] and 38 guests