Suggestion to Make Stats More Meaningful

[BaldEagl]

I don't see people landing unusually high numbers of kills.  The guys who land 5-10 kills usually don't need to re-arm to do so and I can't remember seeing many more than that (except maybe in vehicles).  Heck, even I landed 7 in a Spit XVI one night followed by 7 in a F6F-5 the next night without re-arming either time.

I only use the re-arm pad if I land without any kills.  That just won't do.    As long as I have one I'll tower out for a new plane.

[WarTooth]

BaldEagl,

Very good points sir.

1. Do you think most people have your level of skill in the game?

2. Do you think those that rearm to boost their kills\sortie have your level of skill?

I guess what I am trying to determine is that for superior pilots as yourself there seems to be less of a need to rearm for boosting stats.  I am proposing that it will make the game better to rearm because you have a need to rearm and not because you are trying to boost your kills\sortie stat.

Thanks for your comments,

WT

[Zazen13]

BaldEagl,

Very good points sir.

1. Do you think most people have your level of skill in the game?

2. Do you think those that rearm to boost their kills\sortie have your level of skill?

I guess what I am trying to determine is that for superior pilots as yourself there seems to be less of a need to rearm for boosting stats.  I am proposing that it will make the game better to rearm because you have a need to rearm and not because you are trying to boost your kills\sortie stat.

Thanks for your comments,

WT

Most people I ever see re-arm  do so for an entirely different reason. They're at a base with a good fight then a 30k flotilla of fun nazi's drop FHs. So, when they return to base they re-arm so they can re-up in their fighter right there and not have to fly from a sector away to get back into the scrap.

[Murdr]

The guys who land 5-10 kills usually don't need to re-arm to do so and I can't remember seeing many more than that (except maybe in vehicles).

There ya go.  I've landed 14 and 15 back to back no re-arms or vulches (in a dora no less).  19 with no rearms seems to be my limit.  I'd say well over 90% of the 7+ landings I have do not involve a rearm.  The main times I do rearm it has to do with either a lack of resources available at that airfield, or the occasional poor flight that isn't worth landing yet.

I'm of the opinion that even with rearms, if someone can manage to bag 10 kills in a sortie without sustaining any damage, and can do it at a pace that gives them a competitive kills/per/time then there is no "gaming the game" involved.

[chris3]

moin

i dont understand why people have problems with the rearm pads. i always try to rearm at filds with destroyed fhs hangars. or if my plane is still ok and i want to do a new run. for me its a nice feeling fliing the same bird for a few sorties and bring it home save.

why is the rearm pad a problem for some people?????????????????????

cu chris3

[BaldEagl]

The last few posts did allow me to remember a couple of other situations in which I re-arm besides landing without any kills.

If I'm in attack mode and ord is down at the field I land at I'll re-arm for ord and if I'm in a low ENY plane and the ENY limiter kicks in I might re-arm so as not to have to switch rides.

As to your questions:

"1. Do you think most people have your level of skill in the game?

2. Do you think those that rearm to boost their kills\sortie have your level of skill?"

I think it's best to land any kills you have, especially if your less experienced.  The risk of getting shot down after re-arming is exponentially higher the less experienced/skilled you are.

A guy who manages say 3 K/D and 2 K/S is much less likely to get shot down than a guy who manages 0.5 K/D and 0.33 K/S.  Beyond that, adding even one landed kill has a much bigger positive impact for the less skilled pilot.

If you already have kills under your belt then re-arming to boost K/S is like placing a bet in Vegas; the odds are stacked against you.  Those with the best chance of beating the odds are the sharks who, coincidently, play well enough that they don't need to take that extra risk.

[Anaxogoras]

If you already have kills under your belt then re-arming to boost K/S is like placing a bet in Vegas; the odds are stacked against you.  Those with the best chance of beating the odds are the sharks who, coincidently, play well enough that they don't need to take that extra risk.

The odds of getting shot down do not change depending on how many kills you're carrying, nor do they change if you rearm.  True, if you rearm you wager the kills you already have, but you also stand to gain more.  So what changes is the negative value of getting shot down; the more kills you have the worse it is to get shot down before you landed them...But, again, the positive value of landing after you rearm and get more kills goes up.  It seems to be a very balanced 50/50 wager to me, far better odds than any game in Vegas where you play against the house.

Perhaps an even more apt comparison is to poker, where the majority of players are so full of leaks the good player's actions actually have positive expectation.

[BaldEagl]

The odds of getting shot down do not change depending on how many kills you're carrying, nor do they change if you rearm.  True, if you rearm you wager the kills you already have, but you also stand to gain more.  So what changes is the negative value of getting shot down; the more kills you have the worse it is to get shot down before you landed them...But, again, the positive value of landing after you rearm and get more kills goes up.  It seems to be a very balanced 50/50 wager to me, far better odds than any game in Vegas where you play against the house.

Perhaps an even more apt comparison is to poker, where the majority of players are so full of leaks the good player's actions actually have positive expectation.

If you only manage 0.33 K/S and 0.50 K/D, land, re-arm and re-up for what amounts to another sortie what makes you think those numbers will suddenly change?  In that case you are playing against the house.  You beat the odds to land in the first place.  The odds are that for every kill you get you're going to be killed twice and that you'll only get a kill every three sorties.  That means you'll go up, get killed twice then finally land a kill on the third sortie.  Those don't seem like 50/50 odds to me, at least not when I learned math.

At 2 K/D and 1 K/S you do reach even odds.  You're as likely to get killed after the re-arm as you are to get another kill and land with two.  That being the case, if it alternated, then you've gained nor lost anything in re-arming.

Anything over those numbers and you're likely to win by re-arming but at that point, as I've stated, you really don't need to.

[LLogann]

Hey WarTooth, <S>, all respect due but... In essence you want to rewrite what the definition of sortie is.    So.... NO

Should a sortie end when a plane lands? yes or no.

And besides, how many of us are good enough to resup 6 times to get 23 kills?  I know I'm not.

[Anaxogoras]

If you only manage 0.33 K/S and 0.50 K/D, land, re-arm and re-up for what amounts to another sortie what makes you think those numbers will suddenly change?  In that case you are playing against the house.  You beat the odds to land in the first place.  The odds are that for every kill you get you're going to be killed twice and that you'll only get a kill every three sorties.  That means you'll go up, get killed twice then finally land a kill on the third sortie.  Those don't seem like 50/50 odds to me, at least not when I learned math.

You didn't make it clear that you were stipulating a particular stat profile for making a general decision.  I wasn't assuming pilot ability when I made the statement above.

To use your numbers: if player x lands with 2 kills (and no damage), and there's a 33% chance he can rearm and bring back 3 kills, and a 50% chance that he will die trying, he should give it a try.  Why?  Because if we're trying to maximize K/S (K/D takes care of itself), that's the only way to do it.  Even though he beat his normal %'s, landing those 2 kills won't increase K/S as much as trying for more.  The only thing it's good for is getting your name in the text buffer.

Hey WarTooth, <S>, all respect due but... In essence you want to rewrite what the definition of sortie is.    So.... NO

Actually, I think "sortie" is rather ambiguous as to whether an aircraft rearming and going out again is one sortie or two.  We should ask someone who is in the airforce.

[Zazen13]

If you only manage 0.33 K/S and 0.50 K/D, land, re-arm and re-up for what amounts to another sortie what makes you think those numbers will suddenly change?  In that case you are playing against the house.  You beat the odds to land in the first place.  The odds are that for every kill you get you're going to be killed twice and that you'll only get a kill every three sorties.  That means you'll go up, get killed twice then finally land a kill on the third sortie.  Those don't seem like 50/50 odds to me, at least not when I learned math.

At 2 K/D and 1 K/S you do reach even odds.  You're as likely to get killed after the re-arm as you are to get another kill and land with two.  That being the case, if it alternated, then you've gained nor lost anything in re-arming.

Anything over those numbers and you're likely to win by re-arming but at that point, as I've stated, you really don't need to.

That's a beautiful analysis BaldEagle..Very nicely explained and illustrated. 

[Anaxogoras]

At 2 K/D and 1 K/S you do reach even odds.  You're as likely to get killed after the re-arm as you are to get another kill and land with two.  That being the case, if it alternated, then you've gained nor lost anything in re-arming.

That's a beautiful analysis BaldEagle..Very nicely explained and illustrated. 

No, the analysis is flawed because it commits the gambler's fallacy (http://en.wikipedia.org/wiki/Gambler%27s_fallacy), that past events change future probabilities.  In this case, that before getting any kills the pilot is a 2:1 favorite, and after getting a kill he is a 1:1 coin toss.  If the pilot's stipulated k/d expectation is 2:1, then it remains 2:1 even after he lands a kill.  Even if he gets 10 kills in a row without dying, his k/d expectation is still 2:1 for all future sorties.

A good analogy would be if you were playing cards and you made 10 flush draws in a row, even though you were a ~25% dog every time.  Then, you get an 11th flush draw, the pot is offering you 2:1, but you don't put the money in because you just made 10 flush draws in a row and you couldn't possibly make 11.

[Zazen13]

No, the analysis is flawed because it commits the gambler's fallacy (http://en.wikipedia.org/wiki/Gambler%27s_fallacy), that past events change future probabilities.  In this case, that before getting any kills the pilot is a 2:1 favorite, and after getting a kill he is a 1:1 coin toss.  If the pilot's stipulated k/d expectation is 2:1, then it remains 2:1 even after he lands a kill.  Even if he gets 10 kills in a row without dying, his k/d expectation is still 2:1 for all future sorties.

A good analogy would be if you were playing cards and you made 10 flush draws in a row, even though you were a ~25% dog every time.  Then, you get an 11th flush draw, the pot is offering you 2:1, but you don't put the money in because you just made 10 flush draws in a row and you couldn't possibly make 11.

 You're logic is based upon the assumption of random chance and the supposition that over time there will be a wide fluctuation of results eventually leading to a random distribution of all possible outcomes. AH is not a game of chance like drawing cards from a deck, it is a game of skill for the most part. When getting airborne again for the 1,000Th time it is actually very likely they will repeat past performance as chance has very little to do with it. It would be illogical to assume that a person could suddenly deviate wildly from their proven poor skill level based performance to the point it would be to their benefit statistically to rearm. An AH sortie is not a factor of random distribution making recurrence of a pattern less likely upon repetition. Random distribution making all possibilities equally probable over time does not exist in contests of skill...So, if a person sucked the previous 1,000 sorties it is extremely likely they will also suck the next 1,000, not less likely.

Take it from someone who plays pool for money, if random chance noticably influenced my results and I could not anticipate future probable outcomes based on past performance I'd be living in a cardboard box under a highway overpass.

[Masherbrum]

Stats are useless to me.   Stats are manipulated.   

In a furball or 1 vs 1 in the MA, I often know who I'm fighting.   This is about 75% of the time too.   THAT is more meaningful to me than anything else.

[Anaxogoras]

Take it from someone who plays pool for money, if random chance noticably influenced my results and I could not anticipate future probable outcomes based on past performance I'd be living in a cardboard box under a highway overpass.

That's a nice appeal to authority.   On the flip side, I used to gamble to supplement my income, and I promise you that random chance does not mean probable future outcomes cannot be anticipated.

You're logic is based upon the assumption of random chance and the supposition that over time there will be a wide fluctuation of results eventually leading to a random distribution of all possible outcomes. AH is not a game of chance like drawing cards from a deck, it is a game of skill for the most part. When getting airborne again for the 1,000Th time it is actually very likely they will repeat past performance as chance has very little to do with it. It would be illogical to assume that a person could suddenly deviate wildly from their proven poor skill level based performance to the point it would be to their benefit statistically to rearm.

Yes, it is very likely they will repeat their past performance.  But if we are assuming a probabilistic approach in our decision over whether or not to rearm, then we don't change k/d probabilities simply because we landed one kill; we are not talking about large deviations.  Moreover, we both know that games of skill are not immune to probabilistic analysis, e.g. Baseball, so I don't quite understand the relevance of AH being a game of skill.  I'm not saying that whether or not a pilot lands a kill or dies is a matter of random chance.  I'm saying that just because you landed a kill and you average 1:1 K/D, that doesn't mean you're worse than 50/50 to kill or die when you take off.  Small deviations in performance happen all of the time in this game, and sometimes in ways that are totally out of our control (like the tail-less Niki that fell into my windscreen last night ).

It is only if landing one and only one kill takes priority over K/S that there is no benefit to rearming when your aircraft is undamaged and you landed a kill.