<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Who&#8217;s Going to Win Nathan&#8217;s Hot Dog Eating Contest?</title>
	<atom:link href="http://flowingdata.com/2009/07/02/whos-going-to-win-nathans-hot-dog-eating-contest/feed/" rel="self" type="application/rss+xml" />
	<link>http://flowingdata.com/2009/07/02/whos-going-to-win-nathans-hot-dog-eating-contest/</link>
	<description>Strength in Numbers</description>
	<lastBuildDate>Sun, 12 Feb 2012 04:57:37 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: Jon Peltier</title>
		<link>http://flowingdata.com/2009/07/02/whos-going-to-win-nathans-hot-dog-eating-contest/#comment-31523</link>
		<dc:creator>Jon Peltier</dc:creator>
		<pubDate>Fri, 03 Jul 2009 04:30:46 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=2005#comment-31523</guid>
		<description>What would make this a real good infographic is if you used hot-dog-shaped bars.

(Pardon me while I dislodge my tongue from my cheek!)</description>
		<content:encoded><![CDATA[<p>What would make this a real good infographic is if you used hot-dog-shaped bars.</p>
<p>(Pardon me while I dislodge my tongue from my cheek!)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Naomi B. Robbins</title>
		<link>http://flowingdata.com/2009/07/02/whos-going-to-win-nathans-hot-dog-eating-contest/#comment-31515</link>
		<dc:creator>Naomi B. Robbins</dc:creator>
		<pubDate>Thu, 02 Jul 2009 18:07:35 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=2005#comment-31515</guid>
		<description>Thanks for your kind words about Creating More Effective Graphs. :)</description>
		<content:encoded><![CDATA[<p>Thanks for your kind words about Creating More Effective Graphs. :)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Noah Iliinsky</title>
		<link>http://flowingdata.com/2009/07/02/whos-going-to-win-nathans-hot-dog-eating-contest/#comment-31514</link>
		<dc:creator>Noah Iliinsky</dc:creator>
		<pubDate>Thu, 02 Jul 2009 17:31:28 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=2005#comment-31514</guid>
		<description>I completely agree with Joan. 

The distinction that is most relevant is whether the data is continuous or discrete. Temperature over a 24 hour period is continuous, and therefore a line graph is an accurate representation of the the reality. 

Event winners, whether hot dogs, or racing times, may trend over time, but are discrete events and therefore bars are a better choice. 

A potentially nebulous middle ground might be sales figures for a time period. The periods are discrete, but sales may have some trending and influencing that imply continuity.</description>
		<content:encoded><![CDATA[<p>I completely agree with Joan. </p>
<p>The distinction that is most relevant is whether the data is continuous or discrete. Temperature over a 24 hour period is continuous, and therefore a line graph is an accurate representation of the the reality. </p>
<p>Event winners, whether hot dogs, or racing times, may trend over time, but are discrete events and therefore bars are a better choice. </p>
<p>A potentially nebulous middle ground might be sales figures for a time period. The periods are discrete, but sales may have some trending and influencing that imply continuity.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: jcukier</title>
		<link>http://flowingdata.com/2009/07/02/whos-going-to-win-nathans-hot-dog-eating-contest/#comment-31511</link>
		<dc:creator>jcukier</dc:creator>
		<pubDate>Thu, 02 Jul 2009 16:10:07 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=2005#comment-31511</guid>
		<description>I like your book :)</description>
		<content:encoded><![CDATA[<p>I like your book :)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: jcukier</title>
		<link>http://flowingdata.com/2009/07/02/whos-going-to-win-nathans-hot-dog-eating-contest/#comment-31510</link>
		<dc:creator>jcukier</dc:creator>
		<pubDate>Thu, 02 Jul 2009 16:09:31 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=2005#comment-31510</guid>
		<description>line charts are useful if the slope of the line is meaningful. this is often the case for time series, which are often continuous functions. so if you have an observation for time T, and another one for time T+1, the slope can give you an indication of the value for the function anytime in between. 

here you have discrete, unrelated observations, so a slope between them is not meaningful and probably misleading.

Bar charts it is.</description>
		<content:encoded><![CDATA[<p>line charts are useful if the slope of the line is meaningful. this is often the case for time series, which are often continuous functions. so if you have an observation for time T, and another one for time T+1, the slope can give you an indication of the value for the function anytime in between. </p>
<p>here you have discrete, unrelated observations, so a slope between them is not meaningful and probably misleading.</p>
<p>Bar charts it is.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Joan</title>
		<link>http://flowingdata.com/2009/07/02/whos-going-to-win-nathans-hot-dog-eating-contest/#comment-31509</link>
		<dc:creator>Joan</dc:creator>
		<pubDate>Thu, 02 Jul 2009 14:48:49 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=2005#comment-31509</guid>
		<description>I think bar charts are just as adept at showing trends, but people default to line charts for time series data because it&#039;s thought of as a &lt;a href=&quot;http://stattrek.com/Lesson2/DiscreteContinuous.aspx&quot; rel=&quot;nofollow&quot;&gt;continuous variable&lt;/a&gt;.  That&#039;s not to say it&#039;s an absolute rule though.  I agree with Naomi that the bar format you have helps to visualize the new world records.</description>
		<content:encoded><![CDATA[<p>I think bar charts are just as adept at showing trends, but people default to line charts for time series data because it&#8217;s thought of as a <a href="http://stattrek.com/Lesson2/DiscreteContinuous.aspx" rel="nofollow">continuous variable</a>.  That&#8217;s not to say it&#8217;s an absolute rule though.  I agree with Naomi that the bar format you have helps to visualize the new world records.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Chris P</title>
		<link>http://flowingdata.com/2009/07/02/whos-going-to-win-nathans-hot-dog-eating-contest/#comment-31508</link>
		<dc:creator>Chris P</dc:creator>
		<pubDate>Thu, 02 Jul 2009 14:43:40 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=2005#comment-31508</guid>
		<description>Time series with lines makes sense when there is serial relationship between the events.How connected are the events from one year to the next?  Are the contestants even the same?  

Kobayashi would be considered a level-shift in the series.  Even scarier is the jump from 20 to north of 60 in just 30 years.</description>
		<content:encoded><![CDATA[<p>Time series with lines makes sense when there is serial relationship between the events.How connected are the events from one year to the next?  Are the contestants even the same?  </p>
<p>Kobayashi would be considered a level-shift in the series.  Even scarier is the jump from 20 to north of 60 in just 30 years.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Naomi B. Robbins</title>
		<link>http://flowingdata.com/2009/07/02/whos-going-to-win-nathans-hot-dog-eating-contest/#comment-31507</link>
		<dc:creator>Naomi B. Robbins</dc:creator>
		<pubDate>Thu, 02 Jul 2009 14:31:56 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=2005#comment-31507</guid>
		<description>I&#039;ve always heard that one should use a line chart if the trend is more important and a bar chart if the individual values are of interest. Here you are more interested in the individual values so I agree with your choice. Also, the new world records would be less obvious with a line chart.</description>
		<content:encoded><![CDATA[<p>I&#8217;ve always heard that one should use a line chart if the trend is more important and a bar chart if the individual values are of interest. Here you are more interested in the individual values so I agree with your choice. Also, the new world records would be less obvious with a line chart.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Nathan Yau</title>
		<link>http://flowingdata.com/2009/07/02/whos-going-to-win-nathans-hot-dog-eating-contest/#comment-31506</link>
		<dc:creator>Nathan Yau</dc:creator>
		<pubDate>Thu, 02 Jul 2009 14:12:37 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=2005#comment-31506</guid>
		<description>what&#039;s wrong with that? you can read it, time is labeled, nothing is misrepresented.</description>
		<content:encoded><![CDATA[<p>what&#8217;s wrong with that? you can read it, time is labeled, nothing is misrepresented.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: John S.</title>
		<link>http://flowingdata.com/2009/07/02/whos-going-to-win-nathans-hot-dog-eating-contest/#comment-31503</link>
		<dc:creator>John S.</dc:creator>
		<pubDate>Thu, 02 Jul 2009 10:03:35 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=2005#comment-31503</guid>
		<description>Using a bar chart for time series data?! I&#039;m shocked. What is this, USA Today?</description>
		<content:encoded><![CDATA[<p>Using a bar chart for time series data?! I&#8217;m shocked. What is this, USA Today?</p>
]]></content:encoded>
	</item>
</channel>
</rss>

