Napoleon was the Best General Ever, and the Math Proves it.

[Razgovory]

Due to the peculiarity of the system used, the Washington Generals scored remarkably high.

[DGuller]

General (gettit?) comment: I know this one is just for fun, and that the metric has... issues. I do however fully support doing the numbers on history, I think it's very important if we actually want to understand and learn from history. As long as you keep in mind exactly what you're doing and the uncertainties you have any piece of the puzzle is valuable.

I'm not sure I agree.  I think it's dangerous to put numbers on things that are in significant part non-quantifiable.  Numbers are a powerful tool, but powerful tools tend to be dangerous when used for unsuitable purpose.

[alfred russel]

Wikipedia may not be the most reliable source, but it's presumably not a biased source in connection with this kind of information (i.e. its not more likely to get troop numbers wrong for Lee than for Winfield Scott). 

It is deeply biased. Some of these biases are cultural - I have a hunch the battles of european and english speaking cultures are better documented on english wikipedia than others. Other biases simply reflect the source data: troop numbers in the non modern world are notoriously unreliable and biased in favor of the party writing the history. Also, there are only 9 documented battles for Alexander the Great, and not so many for other leaders such as Julius Caesar. Is that because they really fought so many less battles than Napoleon, or because less significant battles were not separately recorded in the narrative style of ancient histories?

[Valmy]

I don't think being limited by language and evidence is the same thing as a bias.

[Jacob]

I don't think being limited by language and evidence is the same thing as a bias.

It is almost the definition of bias: "a systematic distortion of a statistical result due to a factor not allowed for in its derivation."

[garbon]

Yeah bias in data doesn't mean you intentionally did so.

[DGuller]

Bias is one of those useful and well-defined terms that has been hijacked by social justice warriors.  What makes it even worse is that sometimes what is called bias is actually absence of bias, statistically speaking.

[Jacob]

Bias is one of those useful and well-defined terms that has been hijacked by social justice warriors.  What makes it even worse is that sometimes what is called bias is actually absence of bias, statistically speaking.

Not really... bias is one of those terms that have multiple well-defined meanings depending on the field in which it is used, as well has having colloquial meanings. Just because you regularly use the term as defined in statistical jargon doesn't mean that definition is the only one or that it was "hijacked".

BIAS DEFINITION
1.
prejudice in favor of or against one thing, person, or group compared with another, usually in a way considered to be unfair.
"there was evidence of bias against foreign applicants"
synonyms: prejudice, partiality, partisanship, favoritism, unfairness, one-sidedness;
antonyms: impartiality

a concentration on or interest in one particular area or subject.
"he worked on a variety of Greek topics, with a discernible bias toward philosophy"

Statistics
a systematic distortion of a statistical result due to a factor not allowed for in its derivation.

2.
in some sports, such as lawn bowling, the irregular shape given to a ball.
the oblique course taken by a ball as a result of its irregular shape.

3.
electronics
a steady voltage, magnetic field, or other factor applied to an electronic system or device to cause it to operate over a predetermined range.

[DGuller]

The first and thirds definitions are not inconsistent, although the first one is incomplete.  Unfair prejudice is bias in the statistical sense as well.  Fair prejudice is not, though, and that's where the use gets corrupted.

I understand that a word can be used in different ways, but it gets corrupted when it gets used in opposite ways.

[The Brain]

General (gettit?) comment: I know this one is just for fun, and that the metric has... issues. I do however fully support doing the numbers on history, I think it's very important if we actually want to understand and learn from history. As long as you keep in mind exactly what you're doing and the uncertainties you have any piece of the puzzle is valuable.

I'm not sure I agree.  I think it's dangerous to put numbers on things that are in significant part non-quantifiable.  Numbers are a powerful tool, but powerful tools tend to be dangerous when used for unsuitable purpose.

You think that if you keep in mind exactly what you're doing and the uncertainties involved you use stuff for unsuitable purposes?

[DGuller]

General (gettit?) comment: I know this one is just for fun, and that the metric has... issues. I do however fully support doing the numbers on history, I think it's very important if we actually want to understand and learn from history. As long as you keep in mind exactly what you're doing and the uncertainties you have any piece of the puzzle is valuable.

I'm not sure I agree.  I think it's dangerous to put numbers on things that are in significant part non-quantifiable.  Numbers are a powerful tool, but powerful tools tend to be dangerous when used for unsuitable purpose.

You think that if you keep in mind exactly what you're doing and the uncertainties involved you use stuff for unsuitable purposes?

If you're a robot who does everything optimally, more information never hurts.  If you're a human, you're attracted to numbers much more than you're attracted to intangibles.  Quantifying quantifiable things improves efficiency dramatically;  quantifying unquantifiable things can lead you very effectively into a disaster.  Humans are very bad about correctly dealing with uncertainties.

[The Brain]

General (gettit?) comment: I know this one is just for fun, and that the metric has... issues. I do however fully support doing the numbers on history, I think it's very important if we actually want to understand and learn from history. As long as you keep in mind exactly what you're doing and the uncertainties you have any piece of the puzzle is valuable.

I'm not sure I agree.  I think it's dangerous to put numbers on things that are in significant part non-quantifiable.  Numbers are a powerful tool, but powerful tools tend to be dangerous when used for unsuitable purpose.

You think that if you keep in mind exactly what you're doing and the uncertainties involved you use stuff for unsuitable purposes?

If you're a robot who does everything optimally, more information never hurts.  If you're a human, you're attracted to numbers much more than you're attracted to intangibles.  Quantifying quantifiable things improves efficiency dramatically;  quantifying unquantifiable things can lead you very effectively into a disaster.  Humans are very bad about correctly dealing with uncertainties.

Why would you quantify unquantifiables?

[DGuller]

Why would you quantify unquantifiables?

Because some people prefer the comfort of misleading numbers to the discomfort of subjective intangibles.

[garbon]

"And then he showed me his subjective tangibles!"

[mongers]

If you're a robot who does everything optimally, more information never hurts.  If you're a human, you're attracted to numbers much more than you're attracted to intangibles.  Quantifying quantifiable things improves efficiency dramatically;  quantifying unquantifiable things can lead you very effectively into a disaster.  Humans are very bad about correctly dealing with uncertainties.

You've not had much experience with ordinary people have you? 

Some people will happily tell you numbers aren't their thing or that two digit or more numbers make their head hurt.