For every real number of somebody whose life or the life of a loved one was saved because a gun was in the house, there is another real number of somebody who died in an accidental shooting or shot a loved one by mistake.
Yes, it wouldn't surprise me if overall it turned out to be pretty much a wash. And I think that a person could make a rational choice that even if there is no net advantage, they would rather take the risk of accident if it enabled them to be more prepared to defend themselves or their family. Moreover, if the numbers are close to a wash overall, it is certain that there are some people--people who are meticulous, not inherently overexcitable, who do not have anger management problems, who take careful measures to keep the guns out of the hands of children, and who have invested the training time to learn how to handle a gun responsibly--for whom the risks work out in favor of keeping a gun in the house. This would certainly include the vast majority of police officers, and probably many combat-trained military veterans and quite a few private citizens.
I used to find this a more persuasive argument than I do today. Whereas when this amendment was written, civilian weapons were comparable to those carried by the military, this has long not been the case. Even under existing laws, civilians cannot possess fully automatic weapons or rocket-propelled grenades, not to mention high-caliber vehicle-mounted weapons, tanks, drones, etc., etc. So I don't think the legality of assault weapons with large clips would have much impact on the ability of the resistance to mount an armed insurrection against the government in the event of a military coup in the USA. Moreover, experience in the Middle East has shown that insurrections can pose a substantial challenge to heavily armed military using improvised weapons such as IEDs. And of course, there are weapons held by local police, at least some of whom would presumably fight on the side of the resistance in such a scenario. Moreover, while this worry may have had some validity in the early days of the union, I think that we now have enough history to fairly judge that a military takeover of the USA is not a rational fear. The kind of people who are crazy enough to believe that this is a real concern are not the sort that I care to have heavily armed.
How about from the whitehouse itself, research on guns was never banned, just advocacy. BOOM
This is hair-splitting. What if the results indicated that gun control reduces crime or deaths? How could such results even be reported without running afoul of the Congressional prohibition? So in effect, the restriction says, "You can do research on gun control so long as your findings do not favor gun control." Since the agency can't know what the outcome of the research will be in advance, the net effect is to make such research impossible. What Federal administrator is going to be willing to risk having his career ended by an accusation of misappropriation of Federal funds because the research came out the "wrong" way? The practical effect is clear: the research was shut down. From the JAMA article:
Precisely what was or was not permitted under the clause was unclear. But no federal employee was willing to risk his or her career or the agency's funding to find out. Extramural support for firearm injury prevention research quickly dried up. Even today, 17 years after this legislative action, the CDC's website lacks specific links to information about preventing firearm-related violence.
The only thing 'blocked by the NRA' is a line item for budget from congress. This does not prevent or preclude any kind of research or overhead funding of such a research project by any agency. Please get your information straight.
j
Also false, as documented in the Journal of the American Medical Association article (which I gather that you still haven't bothered to read) which has links to the relevant appropriations acts for the agencies in question, which includes general restrictions on the use of the appropriated funds, not merely blocking a single line-item project.
If you'd actually bothered to read the link, you would see that political pressure has not merely blocked research by the CDC, but also by other federal agencies, including the NIAAA and other institutes of the NIH, as well as some state agencies. So this is not a matter of a single director--the gun lobby is afraid of any real research on the impact of guns and gun regulation on human welfare.
So you are saying, with the only line from the link you posted that is actually relevant (the rest being a retarded emotional appeal, and in some cases totally off topic??), that congress is the only institution that can fund research like this?
Yes, because public health research is generally not profitable, it is almost exclusively funded by federal agencies like the NIH and the CDC. If these are prohibited from doing such studies, it is virtually certain that it won't be done.
If this kind of data manipulation can be done to account for cause and cure with no control groups or understanding of which individuals have taken the medication, please, run the numbers as the data is already available. We have this wonderful invention called the internet where you can self publish your results.
Yes, epidemiology routinely deals with this sort of situation. But that does not mean that it is the sort of thing where somebody can "run the numbers" in an hour or two. To do this properly requires a significant investment of time and effort by people with a high level of expertise and experience in doing this kind of study. Which is why blocking funding is a very effective way of preventing such research from being done.
You can already pull crime statistics, nobody is stopping you from doing this - same as disease statistics. Violent crime rates have gone down, ALOT, over the past 30-40 years. With a height of 52 victims of violent crimes per 1000 people down to less than 17 victims of violent crime per 1000 people in 2009. We had less than 50 guns per 100 people in 1969, as of 2009, we are at a 1 for 1 ratio of guns to people. There went your epidemiology calculations.
Unfortunately for your thesis, crime rates have also come down "ALOT" in states with strict gun control. It is thus quite clear that there are social and demographic factors other than guns that influence crime rates. This sort of situation in which there are multiple potential causes is exactly the sort of thing that epidemiological methods are designed to disentangle (and yes, this can indeed handle situations where "cause" and/or "cure" is randomly distributed within a population)--the very sort of research that the gun lobby is trying to prevent.
Surveying people about "defensive uses" is an idiotic way to go about it, and obviously invalid. How do you determine in an objective manner whether what somebody claimed in a survey as a "defensive use" actually prevented injury? Here's a clue--you cannot do valid scientific research by surveying people and asking them to guess what might have happened if they had not had a gun. Real research is factual, not contrafactual.
The only valid way to go about this is to look at objective, verifiable outcomes, such as measured incidence of criminal victimization and injury--and this is exactly the kind of research that the gun lobby has been trying to derail--presumably because they believe that the knowledge obtained from such factual research would not be in their best interest.
You obviously did not even read the reference I linked to, and are just parroting some bullshit that you heard somewhere. What "principal researcher" are you talking about? What was his name? If you'd actually read the link, you'd have known that it did not refer to a single study, but to a coordinated political effort to prevent any research from being done by any agency on the impact of guns and gun control on human health.
Moreover, the argument that you are parroting is foolish guilt-by-association. The fact that this kind of statistical methodology is used to study the impact of pathogens does not mean that anything studied using this kind of statistics is a "pathogen in the human body." The same methodology is used to study the impact on human health of any environmental factor.
Do you imagine that epidemiologists try to determine whether a public health measure is effective by counting people who were "almost" infected by a disease? Instead, they count the number of people who were actually infected. If a public health measure results in an appreciable number of people being "almost" infected who would otherwise have been actually infected, then that will show up as a decrease in the disease incidence. Similarly, it is possible to investigate whether there is a decrease in people who were actually mugged/raped/assaulted.
And yes, just as there are multiple reasons why the incidence of disease may be different from place to place (demographic differences, differences in reporting, etc.), there are multiple reasons (many of them similar) why crime rates can vary. And there are well developed statistical methods for teasing these different factors apart.
Crime and legal gun ownership do not correlate on any meaningful level.
If that is really true, it argues that legal gun ownership does appreciably reduce or deter crime. However, the available information is limited, because careful epidemiological investigation of effects of guns on human welfare has been systematically suppressed by the gun lobby--presumably because they suspect that the results would not favor their position.
There's more that one way to approach a question. Different states have very different gun laws. So one could compare rates of victimization for various types of crimes (with appropriate statistical adjustment for demographic factors). One could look at rates of accidental gun injuries and "friendly fire" shootings. There have been efforts to research these issues using the same sort of sophisticated epidemiology that has been developed to assess disease risk and drug safety. Unfortunately such research has been largely blocked by political pressure from the gun lobby. Apparently, they feel that their interests are best served if we keep arguing about stupid anecdotes instead of real science.
In one sense, assault rifles with big clips are not a big problem, because mass shootings are rare, despite the heavy media coverage. Relative to the population of the country, "almost no one" is killed in mass shootings. On the other hand, an assault rifle with a big clip is a big advantage to somebody who want to commit this sort of atrocity, and likely increases the body count, and it's hard to see how such weapons provide great benefits for sport or self defense.
Of course, it will never be possible to prohibit large clips entirely. It was clearly just a matter of time before somebody started printing them. But that is not an argument against restricting their sale or possession.
With a question like this, anecdotes are pretty much worthless, just a way of distracting people from thinking rationally about the real issues of risk and benefit. For every anecdote of somebody whose life or the life of a loved one was saved because a gun was in the house, there is another anecdote of somebody who died in an accidental shooting or shot a loved one by mistake. There are examples of people who survived an auto accident only because they were thrown from the car, but that doesn't mean that you shouldn't fasten your seat belt--we know that because we have actual statistics that show that [I]on the average[/I], seat belts save lives.
So if you want to make a real case, forget the anecdotes and cite some real numbers.
I've been annoyed at MIT for years because they ran a financial aid bait and switch on me and my parents, offering a very attractive aid package to get me there, then cutting it substantially in the second year (a friend with less family resources dropped out for the same reason). Every time the fund-raisers call me up to ask for an alumni donation, I tell them about it, but in recent years I've relented enough to make modest donations so long as they are earmarked exclusively for undergrad research. But in the wake of this (and MIT behaved abominably, regardless of whether it contributed to his suicide), I'll find it very hard to give MIT any money at all.
Now if it was iPad screens, I'd consider the report more plausible. One thing that nobody can know until Apple's quarterly report is the impact of the iPad mini on sales of the full-size iPad. There are three iPad mini buyer scenarios, not all mutually exclusive.
Scenario 1: The iPad mini attracts buyers who might otherwise have bought an iPad. The iPad mini has a healthy margin, but it's a cheaper product, so Apple's total revenue will suffer.
Scenario 2: The iPad mini appeals to buyers who would not have bought a full-size iPad, but who might otherwise have bought a low-priced mini-tablet from Amazon or Samsung. Many of these will go on to buy apps, and may want to trade up in the future to a full-sized unit. Apple's total revenue benefits.
Scenario 3: The iPad mini appeals who to buyers who already have a full-sized iPad and who are not in the market for another full-sized unit, but who are willing to consider the mini as a second pad for kids or spouse to take greater advantage of their existing Apple app and media library. Apple's revenues benefit (but perhaps not as much as Scenario 2).
It's mainly a win for iDevice owners. Probably because of competition from Apple, the new Google Maps has the fast vector-based maps that Google never added to the old (Google) Maps app. The lack of a standard Apple app providing transit routing has led to a bunch of new apps offering this feature, so whereas before we had this only from Google, we now have it from a variety of apps. And we also have turn-by-turn from Apple, which Google not only never added to Maps, but even prohibited developers from implementing using Google's data. Indirectly, it is also a win for Apple, since it adds value to Apple's devices.
So the solution to globalization is to encourage American industries to pollute like those in the 3rd world, so Americans can live like Chinese peasants (and your 3% get even richer)? At some point, it's up to the Chinese people to get their own environmental house in order, but even there we can help by setting an example, instead of letting them set the example for us.
This definitely one of the front runners in the "really stupid objection to global warming" competition. It only takes about 10 seconds on the internet to discover that Mauna Loa is only one of many CO2 monitoring stations. It's valued because it is one of the longest running, but there are many, many others that provide cross checks to verify that it isn't only recording volcanic emissions (even if, by some bizarre coincidence, those were increasing). It only takes a little more time to learn something about the many controls done by the Mauna Loa observatory to ensure that their measurements are not corrupted by volcanic CO2.
Or you can simply go to Skeptical Science, who has already done the work for you, and laid out links to the references that you can verify yourself, just in case you've a mind to make another really stupid objection (like "They're a warmist site, so you can't believe anything they say")
A cap and trade program on sulfur emissions led to substantial reductions in acid rain, so it's one approach with real-world evidence that it can work. Because it provides a financial incentive for reducing emissions, the "invisible hand" directs resources to just those areas that are most efficient, and stimulates innovation to develop new ways of reducing emissions http://en.wikipedia.org/wiki/Acid_Rain_Program
Depends on how you measure speed. Both had extremely advanced graphics hardware that handled the most computationally intensive stuff in hardware separately from the cpu, so they could accomplish tasks that competing systems lacked the power to achieve, and did not require particularly fast cpu's (although the cpu's were quite capable. The 65816 in the SNES was essentially an early RISC-like processor, and highly cycle efficient--it was what made it possible for the Apple II gs to implement a full Mac-like GUI of much higher quality than Intel based PCs with nominally faster hardware could handle.) The SNES could handle large 2D sprites in hardware and could scale and rotate bitmaps, which made it possible to reproduce the graphical appearance of arcade games that ran on hardware that cost thousands of dollars, and could produce graphical effects that competing systems could not manage. The N64 was the only machine of its generation that could do true perspective graphics, with antialiasing and correct 3D perspective scaling of textures. The competing systems of the time (the PS1 and Saturn) lacked the power to do this at all--both of them "cheated" by scaling texture bitmaps in 2-dimensions, producing a variety of graphical artifacts including textures that shifted bizarrely as the perspective changed.
It's not like calculating the probability of a dice roll. There are a lot of empirical assumptions in Silver's model. He adjusts for economic conditions, he normalizes polls for historical bias, etc. He has to decide what distributions to use for his Monte-Carlo simulations. So there's plenty of room for him to refine his methodology based on performance. Sam Wang using a simpler median-based method achieved similar accuracy and he argues that Silver's confidence limits are too wide.
Slashdot Mirror
Yes, it wouldn't surprise me if overall it turned out to be pretty much a wash. And I think that a person could make a rational choice that even if there is no net advantage, they would rather take the risk of accident if it enabled them to be more prepared to defend themselves or their family. Moreover, if the numbers are close to a wash overall, it is certain that there are some people--people who are meticulous, not inherently overexcitable, who do not have anger management problems, who take careful measures to keep the guns out of the hands of children, and who have invested the training time to learn how to handle a gun responsibly--for whom the risks work out in favor of keeping a gun in the house. This would certainly include the vast majority of police officers, and probably many combat-trained military veterans and quite a few private citizens.
I used to find this a more persuasive argument than I do today. Whereas when this amendment was written, civilian weapons were comparable to those carried by the military, this has long not been the case. Even under existing laws, civilians cannot possess fully automatic weapons or rocket-propelled grenades, not to mention high-caliber vehicle-mounted weapons, tanks, drones, etc., etc. So I don't think the legality of assault weapons with large clips would have much impact on the ability of the resistance to mount an armed insurrection against the government in the event of a military coup in the USA. Moreover, experience in the Middle East has shown that insurrections can pose a substantial challenge to heavily armed military using improvised weapons such as IEDs. And of course, there are weapons held by local police, at least some of whom would presumably fight on the side of the resistance in such a scenario. Moreover, while this worry may have had some validity in the early days of the union, I think that we now have enough history to fairly judge that a military takeover of the USA is not a rational fear. The kind of people who are crazy enough to believe that this is a real concern are not the sort that I care to have heavily armed.
How about from the whitehouse itself, research on guns was never banned, just advocacy. BOOM
This is hair-splitting. What if the results indicated that gun control reduces crime or deaths? How could such results even be reported without running afoul of the Congressional prohibition? So in effect, the restriction says, "You can do research on gun control so long as your findings do not favor gun control." Since the agency can't know what the outcome of the research will be in advance, the net effect is to make such research impossible. What Federal administrator is going to be willing to risk having his career ended by an accusation of misappropriation of Federal funds because the research came out the "wrong" way? The practical effect is clear: the research was shut down. From the JAMA article:
j
Also false, as documented in the Journal of the American Medical Association article (which I gather that you still haven't bothered to read) which has links to the relevant appropriations acts for the agencies in question, which includes general restrictions on the use of the appropriated funds, not merely blocking a single line-item project.
If you'd actually bothered to read the link, you would see that political pressure has not merely blocked research by the CDC, but also by other federal agencies, including the NIAAA and other institutes of the NIH, as well as some state agencies. So this is not a matter of a single director--the gun lobby is afraid of any real research on the impact of guns and gun regulation on human welfare.
Yes, because public health research is generally not profitable, it is almost exclusively funded by federal agencies like the NIH and the CDC. If these are prohibited from doing such studies, it is virtually certain that it won't be done.
Yes, epidemiology routinely deals with this sort of situation. But that does not mean that it is the sort of thing where somebody can "run the numbers" in an hour or two. To do this properly requires a significant investment of time and effort by people with a high level of expertise and experience in doing this kind of study. Which is why blocking funding is a very effective way of preventing such research from being done.
Unfortunately for your thesis, crime rates have also come down "ALOT" in states with strict gun control. It is thus quite clear that there are social and demographic factors other than guns that influence crime rates. This sort of situation in which there are multiple potential causes is exactly the sort of thing that epidemiological methods are designed to disentangle (and yes, this can indeed handle situations where "cause" and/or "cure" is randomly distributed within a population)--the very sort of research that the gun lobby is trying to prevent.
Surveying people about "defensive uses" is an idiotic way to go about it, and obviously invalid. How do you determine in an objective manner whether what somebody claimed in a survey as a "defensive use" actually prevented injury? Here's a clue--you cannot do valid scientific research by surveying people and asking them to guess what might have happened if they had not had a gun. Real research is factual, not contrafactual.
The only valid way to go about this is to look at objective, verifiable outcomes, such as measured incidence of criminal victimization and injury--and this is exactly the kind of research that the gun lobby has been trying to derail--presumably because they believe that the knowledge obtained from such factual research would not be in their best interest.
You obviously did not even read the reference I linked to, and are just parroting some bullshit that you heard somewhere. What "principal researcher" are you talking about? What was his name? If you'd actually read the link, you'd have known that it did not refer to a single study, but to a coordinated political effort to prevent any research from being done by any agency on the impact of guns and gun control on human health.
Moreover, the argument that you are parroting is foolish guilt-by-association. The fact that this kind of statistical methodology is used to study the impact of pathogens does not mean that anything studied using this kind of statistics is a "pathogen in the human body." The same methodology is used to study the impact on human health of any environmental factor.
Do you imagine that epidemiologists try to determine whether a public health measure is effective by counting people who were "almost" infected by a disease? Instead, they count the number of people who were actually infected. If a public health measure results in an appreciable number of people being "almost" infected who would otherwise have been actually infected, then that will show up as a decrease in the disease incidence. Similarly, it is possible to investigate whether there is a decrease in people who were actually mugged/raped/assaulted.
And yes, just as there are multiple reasons why the incidence of disease may be different from place to place (demographic differences, differences in reporting, etc.), there are multiple reasons (many of them similar) why crime rates can vary. And there are well developed statistical methods for teasing these different factors apart.
Crime and legal gun ownership do not correlate on any meaningful level.
If that is really true, it argues that legal gun ownership does appreciably reduce or deter crime. However, the available information is limited, because careful epidemiological investigation of effects of guns on human welfare has been systematically suppressed by the gun lobby--presumably because they suspect that the results would not favor their position.
There's more that one way to approach a question. Different states have very different gun laws. So one could compare rates of victimization for various types of crimes (with appropriate statistical adjustment for demographic factors). One could look at rates of accidental gun injuries and "friendly fire" shootings. There have been efforts to research these issues using the same sort of sophisticated epidemiology that has been developed to assess disease risk and drug safety. Unfortunately such research has been largely blocked by political pressure from the gun lobby. Apparently, they feel that their interests are best served if we keep arguing about stupid anecdotes instead of real science.
In one sense, assault rifles with big clips are not a big problem, because mass shootings are rare, despite the heavy media coverage. Relative to the population of the country, "almost no one" is killed in mass shootings. On the other hand, an assault rifle with a big clip is a big advantage to somebody who want to commit this sort of atrocity, and likely increases the body count, and it's hard to see how such weapons provide great benefits for sport or self defense.
Of course, it will never be possible to prohibit large clips entirely. It was clearly just a matter of time before somebody started printing them. But that is not an argument against restricting their sale or possession.
With a question like this, anecdotes are pretty much worthless, just a way of distracting people from thinking rationally about the real issues of risk and benefit. For every anecdote of somebody whose life or the life of a loved one was saved because a gun was in the house, there is another anecdote of somebody who died in an accidental shooting or shot a loved one by mistake. There are examples of people who survived an auto accident only because they were thrown from the car, but that doesn't mean that you shouldn't fasten your seat belt--we know that because we have actual statistics that show that [I]on the average[/I], seat belts save lives.
So if you want to make a real case, forget the anecdotes and cite some real numbers.
I've been annoyed at MIT for years because they ran a financial aid bait and switch on me and my parents, offering a very attractive aid package to get me there, then cutting it substantially in the second year (a friend with less family resources dropped out for the same reason). Every time the fund-raisers call me up to ask for an alumni donation, I tell them about it, but in recent years I've relented enough to make modest donations so long as they are earmarked exclusively for undergrad research. But in the wake of this (and MIT behaved abominably, regardless of whether it contributed to his suicide), I'll find it very hard to give MIT any money at all.
Now if it was iPad screens, I'd consider the report more plausible. One thing that nobody can know until Apple's quarterly report is the impact of the iPad mini on sales of the full-size iPad. There are three iPad mini buyer scenarios, not all mutually exclusive.
Scenario 1: The iPad mini attracts buyers who might otherwise have bought an iPad. The iPad mini has a healthy margin, but it's a cheaper product, so Apple's total revenue will suffer.
Scenario 2: The iPad mini appeals to buyers who would not have bought a full-size iPad, but who might otherwise have bought a low-priced mini-tablet from Amazon or Samsung. Many of these will go on to buy apps, and may want to trade up in the future to a full-sized unit. Apple's total revenue benefits.
Scenario 3: The iPad mini appeals who to buyers who already have a full-sized iPad and who are not in the market for another full-sized unit, but who are willing to consider the mini as a second pad for kids or spouse to take greater advantage of their existing Apple app and media library. Apple's revenues benefit (but perhaps not as much as Scenario 2).
iPhones in Walmart are selling with a contract, which means that Apple is still making a healthy profit on them.
It's mainly a win for iDevice owners. Probably because of competition from Apple, the new Google Maps has the fast vector-based maps that Google never added to the old (Google) Maps app. The lack of a standard Apple app providing transit routing has led to a bunch of new apps offering this feature, so whereas before we had this only from Google, we now have it from a variety of apps. And we also have turn-by-turn from Apple, which Google not only never added to Maps, but even prohibited developers from implementing using Google's data. Indirectly, it is also a win for Apple, since it adds value to Apple's devices.
You've probably been listening to Fox news or talk radio too much. Try stepping out of the epistemic closure echo chamber and looking at some actual data on satisfaction with health care
So the solution to globalization is to encourage American industries to pollute like those in the 3rd world, so Americans can live like Chinese peasants (and your 3% get even richer)? At some point, it's up to the Chinese people to get their own environmental house in order, but even there we can help by setting an example, instead of letting them set the example for us.
This definitely one of the front runners in the "really stupid objection to global warming" competition. It only takes about 10 seconds on the internet to discover that Mauna Loa is only one of many CO2 monitoring stations. It's valued because it is one of the longest running, but there are many, many others that provide cross checks to verify that it isn't only recording volcanic emissions (even if, by some bizarre coincidence, those were increasing). It only takes a little more time to learn something about the many controls done by the Mauna Loa observatory to ensure that their measurements are not corrupted by volcanic CO2.
Or you can simply go to Skeptical Science, who has already done the work for you, and laid out links to the references that you can verify yourself, just in case you've a mind to make another really stupid objection (like "They're a warmist site, so you can't believe anything they say")
http://www.skepticalscience.com/mauna-loa-volcano-co2-measurements.htm
And lightning has been known to cause forest fires, so we don't have to do anything about our habit of tossing lit cigarette buts out the car window.
A cap and trade program on sulfur emissions led to substantial reductions in acid rain, so it's one approach with real-world evidence that it can work. Because it provides a financial incentive for reducing emissions, the "invisible hand" directs resources to just those areas that are most efficient, and stimulates innovation to develop new ways of reducing emissions
http://en.wikipedia.org/wiki/Acid_Rain_Program
Depends on how you measure speed. Both had extremely advanced graphics hardware that handled the most computationally intensive stuff in hardware separately from the cpu, so they could accomplish tasks that competing systems lacked the power to achieve, and did not require particularly fast cpu's (although the cpu's were quite capable. The 65816 in the SNES was essentially an early RISC-like processor, and highly cycle efficient--it was what made it possible for the Apple II gs to implement a full Mac-like GUI of much higher quality than Intel based PCs with nominally faster hardware could handle.) The SNES could handle large 2D sprites in hardware and could scale and rotate bitmaps, which made it possible to reproduce the graphical appearance of arcade games that ran on hardware that cost thousands of dollars, and could produce graphical effects that competing systems could not manage. The N64 was the only machine of its generation that could do true perspective graphics, with antialiasing and correct 3D perspective scaling of textures. The competing systems of the time (the PS1 and Saturn) lacked the power to do this at all--both of them "cheated" by scaling texture bitmaps in 2-dimensions, producing a variety of graphical artifacts including textures that shifted bizarrely as the perspective changed.
SNES and N64
It's not like calculating the probability of a dice roll. There are a lot of empirical assumptions in Silver's model. He adjusts for economic conditions, he normalizes polls for historical bias, etc. He has to decide what distributions to use for his Monte-Carlo simulations. So there's plenty of room for him to refine his methodology based on performance. Sam Wang using a simpler median-based method achieved similar accuracy and he argues that Silver's confidence limits are too wide.