Two points.
1. The selection pool is smaller, but the graduate numbers are very small, compared to WWII. In short, the small numbers of graduates we have ensures that even our worst graduate that gets to fighters nowadays is already very near the top of natural flying abilities. He's had to prove an ability to fly and think in 4 dimensions, as well as manage the systems of a modern jet aircraft. The focus in some areas is admittedly different (we provide more training pre-solo than the typical WWII pilot got, for example) but the mishap rate for students is extremely low and we don't train for things that used to be important but are no longer relevant. For another example, we don't spend much time training for needle and ball only instrument flight. The current thinking is that all aircraft now have at least 2 independent attitude reference systems, and if they both fail while in the weather, it's just not your day and you may as well eject, because the occurence rate of dual attitude reference system failure in the weather is vanishingly small and doesn't justify more training than an introduction in the simulator on how it can be done but it sucks so you'll probably never do it for real anyhow.
2. It is a myth that the modern fighter pilot is deficient in close-in dogfighting skills and training. In fact, the reality is that BFM has been picked apart into a science. The development of the E/M diagram and close examination of the physics and geometry involved in a BFM fight means that a modern day fighter pilot is far more likely to pick an optimum tactic against any particular opponent the FIRST TIME he ever meets that opponent in combat. If you read books written by WWI and WWII pilots, you find a lot of word of mouth truisms, but not a lot of specifics on why things work. Huge example, R. Johnson's book "thunderbolt". In one passage, he describes what appears to be a lag roll and offset turn-circle to "surprise" a german pilot into thinking that the P-47 had just turned tighter than he did. Johnson can't describe exactly what he did, and at the time it was word of mouth that a P47 with it's boosted ailerons could roll a heck of a lot faster than some of the german fighters, so that roll rate was used to advantage. But for me, having gone through extensive academic and flying training on BFM, even with me being just an F-15E pilot not nearly as focused on BFM as F-15C or F-16 pilots, I know exactly what he did and why it worked, even if he can't describe it.
Now I can teach stuff in primary flight training that a WWII pilot had to learn through trial and error. I can prove on paper why it's a dumb idea for a P-47 to do a flat turn with a 109, and because of the foundation in 4 dimensional flying provided in pilot training, I can tell a new fighter pilot exactly how to win EVERY TIME he takes a p-47 up against a 109 including the very first time, because the physics and tactics are proven and teachable.
Of course there are exceptions, as Bob Hoover and Chuck Yeager found out. Hoover could do things with a plane that most people thought was impossible, because he instinctively understood the physics and aerodynamics involved and could apply them to any plane he flew, on the first flight, with minimal instruction. So nobody is going to out-fly Hoover. But as he found out, a guy who is a good shot might blow him out of the sky before the merge... So there are always exceptions. But I would guess that on average, the average USAF fighter-bound pilot training graduate has been screened and trained to a higher overall skill level than the majority of new fighter pilots in WWII because of the natural evolution of pilot training methods, and the average operational fighter pilot is at least the match of most WWII fighter pilots in terms of pure BFM skills because we've distilled down the science to give each fighter pilot an advantage of tactic selection before he even gets to the merge. As an F-15E pilot, I know damn well I should never try to outrun a Mig-23. But I can turn with him no problem, so I can take him single or double circle and I get the first shot 99% of the time post-merge. With a mig-21, I might have an on-paper E/M advantage but the mig-21 has good low speed handling due to the delta wing so don't get into a low speed scissors or he just might get lucky and gun me stupid, so I should pick something else to do that takes advantage of my modern avionics and missiles, and the mig's poor rearward visibility.
I know that before I get to the merge, the first combat I ever see. And there is nothing "wrong" with my ability to maneuver in 4 dimensions, and I am physically conditioned to fly an extended period of time at 6 to 7 Gs without a G-suit, and up to 9 Gs with a G-suit, so without seeing one second of actual air to air combat, I have an edge that almost no WWII pilots with my zero hours of air combat time could match. Not only that, I have an education on aerospace physiology that was simply unknown except as vague rules of thumb back in the 1940s, which can make me more effective and less likely to suffer injuries over the long run.
Imagine an old-school barefist boxer going up against a modern MMA fighter... That bare knuckle fighter has probably been fighting his whole life without the benefit of modern training and medicine, so he's a total hard case and undoubtedly a good fighter. And it's very likely that one of those skinny Gracie brothers would bust him up in minutes, breaking every joint and bone in his body if he didn't give up. It's an evolution of the sport, and modern fighter pilots have the benefit of almost 100 years of fighter aviation history guiding their training. It's an unfair advantage no matter how well a WWII pilot might have flown his needle and ball instrument approaches, listening to the "dah" signal in his left ear and the "dit" signals in the right ear...
I will say one thing however... Aviation pioneers did have a significant advantage in their ability to think outside the box, using their knowledge to get things done in tight situations. Nowadays, the drawback to everything being a science is that the art of making things up as you go is nearly lost. As an example, I had a chance to talk to one of the P-40 "Burma Banshee" pilots a couple of years ago, and he explained his "instrument approach" procedure, that worked for up to a 4-ship formation. When the weather was as low as 100' ceilings but with tops no more than 14,000 ft, he would follow this procedure. Use basic navigation and radio homing to fly to a certain mountain peak. Cross the peak at a specific altitude and wind-corrected heading and airspeed. Count 30 seconds, then start a 2000 ft per minute descent straight ahead. Watch the altimeter (calibrated by flying past mountain peaks with known altituded). No lower than 500', you should break out over a wide river valley (with 10,000 ft mtns on both sides) near enough the river to fly up the river just below the clouds. After a sharp right turn where your wingman can shift from fingertip into trail formation, there is a sandbar in the middle of the river with a C-47 parked on it because the guy ran out of fuel and landed on a sand bar that was long enough to stop but not long enough to take off. After the C-47 in the river, you pass one narrow stream on the left and then you see a tree. Put gear and flaps down, climb up and turn hard left over the tree, and as you pass over the tree, chop the throttle to idle and flare because you're already over the short 3000' runway.
That sort of stuff isn't taught anymore because half of the WWII guys got smart doing that sort of thing, and the other half died trying. Nowadays, you do it the standard way and if that doesn't work we have very good ejection seats instead of hoping we don't kill ourselves inventing new stuff on the fly.
