AAMOF, this does get covered, at least by implication, early on, but is then ignored to the point where it's not terribly surprising that it seems to be an inconsistency [in fact, it probably has become one]
To whit: that data was being collected by passive systems, and all the numbers in dispute are directly dependent on your estimate of range, and the only thing a passive sensor can measure directly - the bearing to the target. The accuracy and precision of the bearing are under your control, and the better the sensor suite the better both are going to be. And if +/-10% is all you need to get you to within the ranged of traffic control's active systems so they can tell you where to go, that's probably the level your owners are going to pay to maintain [and if the max practical is +/-2%, then that's the best even the most zealous navy is going to do]. While the calibration of your receivers is your responsibility, that's the _only_ part of the distance estimate you influence unless you have really, really good intelligence and very accurate system maps. To get the distance, you measure the signal strength at your receiver, and then plug in assumptions about the strength of the source and the attenuation along the path [such as the famous 1/r^2 assumption for an isotropic radiator]. And therein lies the rub. Fail to keep your calibration up and you can't get an accurate distance even if the source is telling you how strong it is. Plug in inaccurate guesses about source strength, because you don't know and they aren't going to tell you, and it doesn't matter how good your receiver is. And, of course, anybody reviewing your conclusions can insert his own guess if he decides that yours is giving unreasonable results. Add all that up [in quadrature, of course] and you could well be looking at error bars of 15-20% on civilian sensors and 5% or worse even with the best military equipment.
All of which Himself is quite well aware of, since he's discussed back in the day while explaining why fire-control requires active sensors, but a big deal tends not to be made of it.
As an aside, you can collect accurate range data using passive sensors, but only if you have enough stations arranged correctly with precisely-known baselines between them. And the accuracy is a function of the length of the baselines and the orientation of the formation.
ericth wrote:How about this one, then, which I dont believe has been covered before:
One thing that has bothered me is I dont think the books ever adequately explain how even civilian grade sensor suites can have the sort of inaccuracies that become plot points.
One of the most prominent is Crandall & Cronies dissing the reports of the dispatch boat who recorded Byng's demise.
They diss the ship acceleration rates, the missile ranges and accel, and just about everything. I know the plot point is to showcase SLN arrogance, but HTH can even a weak sensor suite be wrong about something like the accel a ship or missile, or the position/range of a launch? Nowhere else in the books is this sort of inaccuracy demonstrated.
Very true. I think the explanation comes not from any inaccuracies or insufficiencies in the sensor suites but from the grey matter accepting or rejecting the data presented due to preexisting prejudices. And as you correctly state "arrogance."
It's hard to reject or "unlearn" those things you spent a lifetime "knowing" are true in favor of new ideas or data. Human psych 101 - Humans can teach mules stubbornness.