They have a really great website. I wonder where they got their documents? Wherever it is, I hope there are more.

I believe I could come up with a system for rating the aircraft and have actually put in some time to taht effect. The problem comes in when you throw in the pilots. I'd rate the rural guys higher than the urban pilots, but you nefver know who will be sitting in the cockpit, do you? A great test pilot is not necessarily a great combat pilot and vice versa. But they COULD be both.

The problem comes in getting the data for all the aircraft, not in evaluating it. I once came up witha GREAT system for rating fighter aircraft. The problem came in when I discovered I couldn't get the data for all the main fighter types much less the lesser-known ones. The best system in the world without any data is ... useless. So I let that one go.

For a real comparison, you'd need to know the top speed at maybe three different altitudes, the absolute top speed and its altitude, the rate of climb at maybe three different altitudes, the rate of roll at maybe three different speeds, the armament factor would have to be worked out along with ammunition capacity, and the turn rate woudl ahve to be known at maybe three differetn speeds. You could then work out a rating system to generate a number taking all factors into account, and scale it to whatever range you want, from 0 - 10, 0 - 100, or whatever. It would ne nice to know acceleration, but those are data that are HARD to find, at least quantitatively.

So my problem came in getting the data for the required points. I could get it for a small handfull of US types, but not for a decent cross section of anyone else's fighter types. Ergo, not much of a system.

I'm thinking that taking your data and looking at it might help to develop a rating system, but I don't see any roll, turn, or ammament / ammunition data. The Soviet Union did a good jop of showing the time to make a 360° turn in seconds for some of their fighters, and from that you could get a sustained turn rate. Almost nobody esle shows those data normally, so it makes it tough to include roll and turn, which are vital to a fighter's combat worth ... at least in my opinion.

If we did our data in metric units, it might look somethign like a number related to

1) a factor times top speed divided by our base top speed. I'd choose a base top speed of 675 kph or so. More is better. I'd like to incorporate top speed at 3 different altitudes.

2) a factor related to the time in secoind required to complete a 360° circle. The factor would be in the denominator so a lower tunr time is better. You'd multiply it times the direct factors so more is better. We'd need to know the altitude.

3) a factor related to the figure or merit of the aggregate armament multiplied by the rate of fire factor multiplied by the ammo capacity factor. In all cases, more is better.

4) a factor related to the roll rate and I'd like to incorporate rool rate at three different speeds. More would be better.

5) a factor related to ceiling.

6) a factor related to range.

7) a factor related to rate of climb. I'd like to include rate of climb at three different altitudes. Maybe 2 altitudes at lilitary power and one at WER.

Mix the factors so more is better and use a weighting multiplier for each. Come up with the numbers and normalize them to a convenient scale, say 0 to 100, and look at ALL the fighters. Rank in order of score.

Of course, I would throw in another three factors and weight them heavily. I'd throw in a factor for 1) number of enemy aircraft shot down, 2) number of own losses due to comabt, and 3) number of own losses due to operations or accidents. These three factors would ... in MY system ... count about a third to a half, making it very difficult for great planes that never fought anything and weren't produced in any numbers to get ranked very high. Might have to throw in a production factor for number built. After all, 20,000 mediocre fighters would do a LOT more good than, say, 43 top-rate fighters could ever achieve, regardless of employment.

So, for instance, the Ki-83 might be a great twin, but they only made 4 and they never got into combat. So it's score wouldn't rival; that of, say. the A6M5 Model 52 Zero by any stretch of the imagination, even though it technically might be better since 20,000 Zeros did a LOT more for Japan than 4 Ki-83s ever did.

I think we could work out a system. The problem then becomes finding all the data and getting into a computer-readable form.

Of course, when we got finished, we might never be able to convince anyone of the merits of the system, so it all might come to naught, except our own satisfaction.

If we come up with a PERFECT system that nodoby much uses, then the system is a failure even if it gives relatively "correct" ratings.