Reply To: Sources for Historical Movement Rates

[Jonathan Gingerich]

Bandit wrote:

<div class=”d4p-bbt-quote-title”>Jonathan Gingerich wrote:</div>
It is intellectually illegitimate to take the maximum turn move, divide it by the nominal turn duration and claim that it is what the rules writer believe is the average maneuver rate.

Just to keep the debate honest, that isn’t what he [McLaddie] did.

It is what he did.

Actually, looking at it more closely, I have a second objection.  Repiqueone clearly stated that he believes 3-4  mph represents the top average of marching men, and all the various “frictions” reduce this.  So when you roll the D12 and get a 6, that is not an average drill rate in good conditions.  That is a drill rate beset by various obstactles, alarums,  delays etc.  So when McLaddie compares it against his 75 yd/min “average” he is no longer presenting 75 as the expected rate in good going, but the expected average whatever crops up.  That’s a change.

Bandit wrote:

<div class=”d4p-bbt-quote-title”>Jonathan Gingerich wrote:</div>
Part of the turn the unit is moving and part it is not. The problem is not the movement, but simulating the decision cycle where things simply happen too quickly on the table top.

And the response each time has been: what about the units that are spending the whole time moving? and questioning of whether the “overhead” of the decision making cycle is even remotely representative of what it was historically.

There aren’t any units spending the whole time moving.   The rules don’t allow that.  You don’t have to like it, but you are just kicking the can down the street and making the same argument in a new guise.

Bandit wrote:

<div class=”d4p-bbt-quote-title”>Jonathan Gingerich wrote:</div>
This is anathema to any statistical analysis. If you corrupt the data with systematic bias, your averages will extract the bias.

That presumes that all the data has been systematically biased in a given direction. McLaddie’s statement was that he believed all of the data was biased in that it was all somewhat incorrect, not that it was all incorrect skewed one way. * One thing I’ve noticed is that all objections to McLaddie’s calculations seem to be targeted at a single example: Soult at Austerlitz climbing Pratzen Heights. Perhaps four examples have been given so far. None of the other three have been objected to, discussed or examined at length. But if I were going to evaluate McLaddie’s claims based on his evidence, the first thing I’d ask for is the other 40 some examples he’s said he has. That is either going to confirm that ‘Soult at the Pratzen’ is an outlier, or that it isn’t, or that the data appears skewed, or that it doesn’t…

You are in deep math.  You are going to need a bigger boat.  Conceivably randomization matching the underlying distribution might only require a larger sample size, but, obviously, an author “fitting” timings to make sense is operating with preconceived notions and should go to Vegas if they aren’t skewing the data.

It’s not Soult at Austerlitz.  Soult went some distance in some time.  He might have made a verbal estimate beforehand, which de Segur recorded from memory, or it might have been an anecdote floating around about what a cool hand Napoleon was, which de Segur thought would make a colorful addition to his memoires and pulled the numbers out of his bicorne.  The number, however, yields 9 mph so of course it’s an outlier.  We aren’t objecting to the caculations; We are skeptical of the quality of the datam.