where? sorry i don't see your demostration. i see only your opinion.
Underappreciated Aircraft of WWII
- Thread starter kration
- Start date
Ad: This forum contains affiliate links to products on Amazon and eBay. More information in Terms and rules
where? sorry i don't see your demostration. i see only your opinion.
From AHT (Dec 3, 1941):
"VF-2 reported: that "it has ceased all operations until enemy contact became imminent. Progressive landing gear failures had started in 12 of the 17 aircraft."
(The landing gear problem was the most serious , but the arresting hook problem was severe also (possibly due to sabotage)).
I don't interpret this as "liked OK".
I guess I don't understand what you are asking. Sorry.
claidemore
Senior Airman
The question isn't about what advantages the Mohawk had over the Hurricane, it's what advantages either plane had over their opposition.
The Mohawk had a definite advantage in roll rate over the Hurricane, as confirmed by the RAE comparison and pilot anecdotes indicate that the Mohawk could outmanuever the Oscar, while the Hurricane could not.
On two occasions Mohawks survived collissions with Oscars, (the Oscars did not), and the Wright radial engine of the Mohawk would have given it an edge in surviving battle damage over the liquid cooled Merlin in the Hurricane.
Some stats on the Hawk75A/Mohawk IV from the manual (sales brochure?):
max range, 1230 miles
max speed 323 mph
stalling speed 70mph
wing loading, 25.1 lbs/sq ft (29lbs for Hurricane II, 24.1 for Oscar)[edit: oops, should be 24.8 lbs / sq ft for Oscar]
power loading 6.23 lbs/hp
gross weight, 5922 (7600 lbs for Hurricane)
max dive speed 455 mph
The Mohawk had a definite advantage in roll rate over the Hurricane, as confirmed by the RAE comparison and pilot anecdotes indicate that the Mohawk could outmanuever the Oscar, while the Hurricane could not.
On two occasions Mohawks survived collissions with Oscars, (the Oscars did not), and the Wright radial engine of the Mohawk would have given it an edge in surviving battle damage over the liquid cooled Merlin in the Hurricane.
Some stats on the Hawk75A/Mohawk IV from the manual (sales brochure?):
max range, 1230 miles
max speed 323 mph
stalling speed 70mph
wing loading, 25.1 lbs/sq ft (29lbs for Hurricane II, 24.1 for Oscar)[edit: oops, should be 24.8 lbs / sq ft for Oscar]
power loading 6.23 lbs/hp
gross weight, 5922 (7600 lbs for Hurricane)
max dive speed 455 mph
Last edited:
buffnut453
Captain
So, according to your figures (rather than anecdotes) the Oscar ought to have been capable of outmanoeuvering the Mohawk due to its lower wing loading???
Does the greater relative manoeuverability of the Mohawk and its radial engine account for the huge disparity in kill/loss ratio between it and the Hurricane? If the consensus is that it does adequately explain it then I'm happy. My gut tells me it's not a sufficient explanation but I'm happy if others disagree.
That is a really good question.
The problem with the Mohawk sample size is it is so small that one good, onesided battle could skew the results decidedly in the Mohawks favor while the Hurricane had, generally, better results throughout.
For instance, if a squadron of Mohawks bounced a Squadron of Oscars, knocking down 5, without loss. Then lose 8 over the succeeding battles while only getting 3, the numbers won't show it (going on an 88 spread for Oscars V Mohawks).
I guess my bottom line is it is dangerous to draw conclusions from so small a sample without a lot more data about how they got there.
I explained the stats math in post above. By my count 8:7 (as I said includes 'crashland', 'belly land' collisions in air combat and the specific cases I noted) in favor of Mohawk is statistically significantly different than Hurricane's 12:55 at a very high level of confidence. Even if we go with your count, if the true underlying ratio of kills/losses is 12/55=.22, it's still only 9% likely you'd get a 4 or more kills v 6 losses, not so far away from statistical significance at 95% (but I don't know how you get that smaller count).
On Hurricane and Hawk v 109, that's a good additional set of info from Fledgling Eagles but you seem to be discussing it as if it's the only data set. See my previous post, Hurricane v 109 in the much larger sample of Battle of France from May 1940, as counted in the "The Battle of France Then and Now", was similar to the sitzkreig period, 74 Bf109's downed for loss of 151 Hurricanes in air combat. In that period Hawk went 23:38 v Bf109, all opponents 'E' type. Actually that difference is *not* statitistically significant at 95% confidence, there's about 25% chance of getting 23:38 randomly if the true probability is reflected by 74:151. Adding in the sitzkrieg numbers, it would be Hurricane 81:164, Hawk, 36:53, .50 v .68:1, with then only about 8% chance of achieving the Hawk's ratio by random if the true ratio were equal to the Hurricane's, IOW that's around as significant as the Burma result to reject the hypothesis that the two had equal chance of success (even per your count with which I respectfully disagree).
As far as Hurricane's other episodes v Bf109E seems "Air War for Yugoslavia Greece and Crete" is the outlier, and I don't understand why we'd discount the Malta 109E results* for elite Germans when smaller Greece result included a number of kills by perhaps the best British fighter pilot of WWII, Pattle. I would just keep aggregating to a bigger sample, and if so the ratio of Hurricane to 109E is not going to improve from 1:2 in well documented episodes, probably less including all of France 1940, Malta, North Africa and Greece. Others sometimes refer to better Hurricane results in other North Europe periods (BoB, sweeps over France later on, etc) but I don't know of books which allow you to count air combat results granularly in those periods, as opposed to reading tea leaves of total periodic losses.
Re: Buffnut that 'we'd accept the Hurricane's lack of success if you gave a reason' I don't really see the logic in that, with all due respect. The results are what they are*. The tendency seems to be to just say differences from whatever it is that people expect *must* be due to the situations of combats. But how really likely is it that Hurricanes were in unfavorable situations continously in FE throughout 1942 and 43 campaigns? Firstly, I just don't see that in aggregate in the books mentioned. And even if so, isn't *always* being in disadvantageous position at *some point* a reflection on the a/c? I just don't accept the logic that somebody presenting results has to have a complete explanation or else there's something wrong or irrelevant about the stats.
I doubt it's generally possible to *prove* why one a/c did better than another in combat, but it has to be shown IMO that there was a some 'unfair' difference causing that, not just assuming it.
*I get 30:0 for Hurricane v Bf109E over Malta in "Hurricanes over Malta", you said 35. I took notes on "Air War for Yugoslavia Greece and Crete" a long time ago but never added it up...I suppose in every case there will be differences due to error or interpretation, but they don't seem that dramatic in general.
Joe
Last edited:
buffnut453
Captain
Joe,
As always your arguments are well presented. However, the sample set is absolutely relevant. A single combat where a formation of K-43s was bounced by Mohawks, the latter having a tactical advantage, which resulted in the downing of 4 Japanese fighters would hugely skew the stats. I'm not saying such an event happened, merely pointing out that sample size IS relevant to the discussion.
I'm interested in the reason because that's how we discover more about the topic under discussion. Causal factors are pivotal to a deeper understanding of issues. Good/Bad, Black/White, Right/Wrong simply isn't good enough for me - there are degrees and shades to all issues and it's the exploration of those shades that I find fascinating. I'm not being revisiosnist, I'm simply trying to understand things in ways that are often under-represented in published works.
KR
Mark
As always your arguments are well presented. However, the sample set is absolutely relevant. A single combat where a formation of K-43s was bounced by Mohawks, the latter having a tactical advantage, which resulted in the downing of 4 Japanese fighters would hugely skew the stats. I'm not saying such an event happened, merely pointing out that sample size IS relevant to the discussion.
I'm interested in the reason because that's how we discover more about the topic under discussion. Causal factors are pivotal to a deeper understanding of issues. Good/Bad, Black/White, Right/Wrong simply isn't good enough for me - there are degrees and shades to all issues and it's the exploration of those shades that I find fascinating. I'm not being revisiosnist, I'm simply trying to understand things in ways that are often under-represented in published works.
KR
Mark
Last edited:
Again, please see the post with statistics analysis. If there wasn't a systematic difference in conditions, then 8:7 and 12:55 are statistically significantly different results at a very high level of confidence. So the issue would be systematic difference, not randomness per se.
The actual cases as I counted were, only combats with one or more real kills, there were a few inconclusive ones, are:
Nov 10 '42: 2 Mohawks, 3 Type 1 (1 Type 1 to collision and 1 Type 1 pilot returned but DOW, I counted)
Dec 5: 1 Mohawk, 1 Type 1
Jan 19: 1 Type 1
Jan 22 '43: 1 Type 1
Feb 12: 1 Type 1
Mar 29: 1 Mohawk (might be operational, but I counted)
Mar 30: 1 Mohawk
Apr 20: 2 Mohawks, 1 Type 1
So the individual outcomes were all pretty close too, and of 8 decisive combats 4 came out in Mohawk's favor, 3 in Type 1's and 1 'tied'. The Hurricane results are a lot to type in case by case, but likewise the variance in results wasn't that great from combat to combat: 4 combats in Hurricanes's favor, 22 in Type 1's favor, 9 tied.
Both stats analysis and a general feel when you look at the details tells you aren't looking at two similar sets of results there. The small sample limits how much you can *quantify* the difference. You can't say the Mohawk was 4 times more effective, the sample *is* too small for *that*. But it's not too small a sample to reject the hypothesis that Hurricane and Mohawk results were really equal but just appear different due to statistical noise.
Joe
As I've said, IMO, the limited engagement of Mohawks in Burma is too small to draw any definitive comparison betweenst it and fighter types based on that Theater. Only 14 Mohawks were lost in total in my calcuations (of which 6 were lost to Ki-43's) A far smaller # than the total losses of Hurricanes and Spitfires.
Don't know why you'd think that.
I don't recall discounting Malta's results. I find it odd though that you'd mention that some of the 109 kills scored by the RAF iin Greece were done so at the hands of one of Britian's better fighter pilots as if this somehow takes away from the fact that it was done from the cockpit of a Hurricane. I mentioned the 7/JG-26 period specifically, along with the results from Greece to emphasis that Hurricane ratios could vary substantially from Theater to Theater. In the former case, the high level of experience and cohesiveness of the German staffel played a major part in their dominance over the ad-hoc Hurricane units stationed at Malta. The Greece fighting, along with the BoB show that the Hurricane could and did achieve competetive and/or a positive kill ratio against enemy fighters along with other forms of success not touched on in this thread. (which seems to focus exclusively on fighter vs fighter)
claidemore
Senior Airman
The wingloading figures of the Oscar and Mohawk are very close, and wingloading is an 'indicator' of turning ability and agility, but not a' measurement' of turning ability. The high speed roll ability of the Mohawk would IMO be a more important factor.
I for one do not discount the anecdotal evidence, particularly when it is supported by data such as the 25 lb wing loading figure and the RAE report comparing roll rates vs Hurricane and Spitfire. If the Mohawk pilots say they could outturn the Oscar, I believe them.
Those pilots also said that visibility was better from the Mohwak than the Hurricane (or Spitfire).
It would be silly to say that one or two factors are responsible for the results of the Mohawk in combat, there are too many factors, some measureable, some not.
The ones we can measure are: 455 mph dive speed, 25.1 lbs/sq ft wing loading (24.8 for Ki43, the figure I gave in previous post was incorrect), max weight under 6000 lbs, roll rate at higher speeds (sorry don't have the RAE report in front of me).
Basically the Mohawk was about the same weight as an Oscar, the same wing area, same horsepower engine, similar armament (slightly better with wing guns added). It should be no surprise that they achieved basic parity in combat vs each other.
The Hurricane II was bigger, heavier(1500 lbs with similar horspower engine), less agile, and had a lower climb rate (+2500 for IIc, 2750 for IIb compared to 3280ft/min for Mohawk). Should be no surprise that RAF pilots who had a long standing paradigm of relying on turning ability would have a tough time flying the Hurricane vs more agile planes.
Statements to the effect that the Hurricane and Mohawk were basically similar in performance are erroneous and misleading. They were in fact quite different planes.
buffnut453
Captain
Joe,
I may have missed it but I couldn't find the Ki-43 loss on 22 Jan 43 so I ended up with a total of 6 Ki-43 "kills" by Mohawks (I didn't count losses where the aircraft return to its operating base). However, 2 of those were due to mid-air collisions with Mohawks while another resulted from the Ki-43 crashing into the ground while chasing the Mohawk at low level. These accidental losses had nothing to do with the Mohawk's ability as a fighter and hence must be discounted - accidents happen but they cannot be said to contribute to the combat performance assessment of the Mohawk (or any other aircraft). So my earlier comment about data potentially being skewed was correct - Mohawks only shot down 3 Ki-43s. Of interest, on half of the engagements you listed, the Mohawks engaged Ki-43s with the former having greater numbers and operating at higher altitude; the Mohawk only achieved kills when it had these tactical advantages.
Now, that said, I haven't been through the Hurricane data (just haven't the time) to do a similar analysis of that aircraft's combats with Ki-43s. I want to compare apples to apples - work out kills directly attributed to gunfire - and determine the combat conditions at the time. It may take me some considerable time to do that for the Hurricane but I'm interested in finding out how it compares.
Kind regards,
Mark
Last edited:
claidemore
Senior Airman
After checking the pilots notes for the Hurricane II models, I found the dive speed limitation to be 390 mph. Don't know how I never found that before.
They also mention that speed builds up slowly in a dive. It would appear that diving away from Japanese planes would not be to the Hurricanes advantage.
They also mention that speed builds up slowly in a dive. It would appear that diving away from Japanese planes would not be to the Hurricanes advantage.
the point it that too small can't be a opinion the statistics give the instruments...
buffnut453
Captain
Vicenzo,
The problem is sample size. When using statistics, small sample sizes mean the results are prone to wide variation depending on how the data was collected. For example, in a political election, a survey may ask 10 people how they will vote. If they ask 10 rich suburbanites, they will likely get a different answer than if they ask 10 poor inner-city dwellers, and a change of one person either way results in a 10% swing in the results. To be reliable, surveys should be taken over as wide a sample set as possible to ensure the results represent the overall view and are not subject to wide swings due to a small number of changes.
The problem with the Mohawk, as I believe I've pointed out, is that it achieved relatively few "kills". Definition of what constitutes a "kill" therefore becomes important, per my previous post - of the 6 known losses suffered by the Ki-43s in combat with the Mohawk, 50% were due to accidents not RAF fire.
I'm in the process of evaluating Hurricane data for the same time period and will post those results when I have them.
Cheers,
Mark
The problem is sample size. When using statistics, small sample sizes mean the results are prone to wide variation depending on how the data was collected. For example, in a political election, a survey may ask 10 people how they will vote. If they ask 10 rich suburbanites, they will likely get a different answer than if they ask 10 poor inner-city dwellers, and a change of one person either way results in a 10% swing in the results. To be reliable, surveys should be taken over as wide a sample set as possible to ensure the results represent the overall view and are not subject to wide swings due to a small number of changes.
The problem with the Mohawk, as I believe I've pointed out, is that it achieved relatively few "kills". Definition of what constitutes a "kill" therefore becomes important, per my previous post - of the 6 known losses suffered by the Ki-43s in combat with the Mohawk, 50% were due to accidents not RAF fire.
I'm in the process of evaluating Hurricane data for the same time period and will post those results when I have them.
Cheers,
Mark
Sorry is not a problem of sample size.
It seemed you were discounting the Malta results, still does actually in this more recent post. I was simply pointing out that the 'but' you are mentioning for Malta (the qualities of particular German unit) could be answered by a 'but' for Greece (fairly small sample in which a number of the 109 kills were by a single excellent pilot). Of course there's a human factors 'but' in every case of combat between air arms, and when comparing different air arms' results v common enemy . The general advantage of German fighters over British (and almost everybody else's) in 1939-42 obviously had to do with factors besides just the a/c, though also was affected in degree by the a/c (eg the Spitfire was generally more successful than the Hurricane v the Bf109). Again, I would just aggregate all the well two-side documented Hurrican v 109E examples, with no special 'asterisks' or weighting for any particular case besides their relative numbers in the average.
As far as BoB, like I said, I don't know of a 'granualar level' analysis that allows counting of Hurricane v 109 kills/losses comparable to the books we've been discussing. I've read general statements that it was something like 1:1.7 (IIRC) in favor of 109 but again I don't know exactly what's behind that (daily totals, weighting them, all causes...it can be pretty murky). Same was true of BoF until "BoF Then and Now" book appeared recently. Anyway, I'm not sure there are well documented examples of Hurricane v 109E a lot better than 1:2 except Greece. Though usual disclaimer, 1:2 kill ratio doesn't mean abject failure depending on a lot of other factors (what else is being accomplished, ability to replace losses v that of enemy, etc).
Re: Buffnut, I always count the same way*. The Hurricane results are counted the same way. In many or most cases stuff like flying into ground 'accidentally' isn't agreed by both sides; and in this particular case the Mohawks 'engaged' the Type 1's first, who is to say they didn't hit and wound Major Yagi? And as you see the British in second case of Type 1 collision believed that a/c turned violently and hit a Mohawk after being hit by gunfire from another Mohawk. Rather than go through this every time with dozens of 'asterisks' in bigger samples, I just ask: was the a/c lost as a direct result of an air combat? The answer in all three of those cases is clearly 'yes'. As far as giving benefit of doubt to reduce Japanese losses (as in not counting as a loss 'managed to force land at Akyab but died in hospital') I could imagine the howling from some if Hurricane/Spitfire claims were ruled 'overclaims' in cases like that.
As far as 'asterisking' altitude, I see that as counter logical: which plane had altitude advantage is clearly at least partly a function of the plane, and/or the operating practice of the air arm.
*subject to vagueness of info in books, and my own mistakes, like counting Jan 22 which was a bomber kill, scribbled notes, sorry. 7:7 per my method.
Joe
Last edited:
buffnut453
Captain
Need to double-check, but IIRC it was the authors' who suggested that the Ki-43 might have been hit by gunfire. Agree we can't asterisk every time we come up against something unusual, hence why I'm sticking to confirmed casualties on both sides that were, without question, the result of gunfire (it ensures we're measuring apples and apples when there's such a small sample set for the Mohawk).
Partly a function of the plane, but also a function of the amount of warning available which, depending on the defence network, is usually less for forward airfields than for those in the read (eg 11 Gp during BoB). I need to complete my analysis of the Hurricane figures but one thing that immediately leaps out at me for the early engagements thru the end of Feb 43 is that many Hurricane sorties were scrambles against IJAAF combined bomber and fighter attacks against the airfields they were operating from. Agartala never had to endure these attacks. These IJAAF raids were also not small affairs - bombers escorted by 30 Ki-43s from both 64th and 50th Sentais were not uncommon. So the Hurricane's tactical operating environment WAS different to that of the Mohawk.
I'm not suggesting one causal factor for the poor performance of the Hurricane - I want to go through the entire period thru end Dec 43 to ensure I've been even-handed in my treatment of both aircraft types.
Cheers,
Mark
