I guess for wheeler with power one just coordinate power & attitude to make it work without increasing ASI, I don’t recall speed changing that much when you gun it with power to flatten it? actually on level flight speed will keeps decreasing unless you add more than 40% power (maybe way less with ground effect), that is assuming you are flying min drag speed…even more pronounced on backside of the drag curve where adding power will reduce speed to maintain level attitude & flight, when you are flying level (say during flare), the key thing for wheelers is to come fast above min drag speed and add enough to keep in fast level attitude
For wheeler without power, I can’t see how it works unless you start flaring at higher speed and touchdown very fast at the right moment of time when attitude is flat
Most vintage tailwheels don’t cruise fast and are draggy without power, I would be impressed if one can pull that many wheelers without power consistently every time, one has to convert the slow or botched ones into 3 points or push the throttle a bit 
I don’t think “arrive too fast and you balloon” is the right way if analysing it.
I believe the flare should be approached as a 2 step process. First part is flare to fly level along the runway. Then “detect the sink” and at this point lift nose to three point attitude. Oh, and this works great on nosewheel too.
This way a 100 knot approach works just fine…. You just end up flying level along the runway as the speed bleeds off, and you still touch down with high AoA (three point)
So this really means that unless you are notably too slow, the approach methods and techniques are almost entirely irrelevant to the round out, flare, touchdown, whatever you want to call it.
Tailwheels are often very light so you have very little inertia with no power, but if you are carrying too much speed into the flare and you put the stick to the neutral position you most certainly will balloon in most of the taildragger (in fact most aircraft) I have flown. You will then have no inertia and will come down like a brick where the bungie suspension will launch you back into air again, and unless you.do something about it such as adding a little power and letting yourself down gently, you can find yourself bouncing along the runway. This is particularly the case on aircraft such as the Stampe. A Jodel is much more forgiving in that respect.
I will own up that I have never flown a taildragger with a 100 knot speed over the hedge. Most of the ones I fly are somewhere around 110 to 140 kilometres per hour. So about half the speed Carl is talking about.
My experience is similar to Carlmeek above. When practicing float flying, I did hundreds (or thousands?) of landings (one every 45 seconds to a minute) along lakes. After a period of practice, you can land at any pitch attitude you want :-). As I practiced, especially when doing deadstick landings, my flare evolved into the two stage process, very much as Carl points out.
First you are descending at whatever rate of descent that you get for your approach speed (usually with a small amount of excess speed to give you time to judge how things are going). This could be a significant ROD, so as you near the ground I would round out, usually trying to fly a smidge below level (no ballooning). Thus you are still closing on the ground (or water) with a low sink rate, but you have time to evaluate the closure rate, your airspeed and the height above the surface. Then you can make small adjustments to fine tune the contact point and pitch angle at contact (e.g. how long you hold off). Floatplane landings are mostly flat, e.g. very much a wheel landing, or a wheel landing with the tail slightly low, and you must control the pitch angle as you contact the water (usually a slight tug back on the stick to counter the drag of the floats – the opposite would be required for a wheeler on land). After a few of these, you can fine tune the approach speed with the knowledge of how much excess is required to let you nail the touchdown. Goldilocks! Not to much, and not too little. Of course you can recover with smidge of throttle if you are a bit low on energy…
When I went back to tailwheel flying, my wheel landings and three point landings were much, much, much, better :-). I think the above technique applies more to wheel landings, which will necessarily be a bit faster, but it works well for both types.
If was doing three pointers only, my technique might be more for stabilised approach (with some throttle, so not a massive rate of descent) at a slightly slower approach speed and then a smooth round out. If you get too slow you are going to need some throttle in the flare.
All good fun, and to for me it is a joy to fly the lightly loaded, low power ragwings…
I don’t think “arrive too fast and you balloon” is the right way if analysing it
I completely agree as long as you can “detect sink” your approach and initial flare speed is irrelevant but then your touchdown speed is further down the line, so you need lot of runway distance
This is very interesting when teaching landing in glider versus power, if you don’t care much about landing on the numbers a fully held off landing with kiss touchdown works beautifully and the resulting skill carry over when flying anything else as long as the runway is long enough
The “don’t worry, I can fly as fast as I can as long as I can detect sink” does not fly very high when you jump into aircraft where people claim and swear they have to be flown by the numbers (Mooneys, Bonanza, Cirrus…) or they will balloon or hit heard, still they do get impressed by your first touchdown on type, the secret is you are floating on 2km of pavement with some residual power, that is cheating
The only time when it does not work is when you are very slow on high rate of descent without energy to flare and then got “hit by a gust” like the majority of pilots say on their landing accidents reports…
The Pitts is not quite 100 KIAS over the fence, but still slightly below around 80 KIAS.
Because it is short coupled and has a strong tailwheel leaf spring, in terms of springiness not material strength, you round out and gently flare while closing the throttle, but not holding the stick full back as in a conventional conventional tail dragger, as this will result in crow hopping. In effect keeping forces on the tailwheel gentle. Have never attempted a wheel landing in a Pitts, although some people use this technique.
The best technique for conventional aircraft is a tail low wheel landing. You don’t carry more speed, just the same 1.2 x Vs you would normally use on final, but on round out (the phase where you change the aiming point and hold the aircraft in a parallel attitude to the surface) you maintain a little power to manage the sink and gently flare. You should touch down at only a few mph faster than a three point, and then you would relax, or slightly move the column forward to reduce AoA and remain landed. As speed decelerates you can move the column forward again, until you then gently pull back to ‘land’ or ‘fly on’ the tailwheel.
Some aircraft can only be landed using the wheel technique, for example the Globe Swift.
The technique works on slow aircraft too. I test flew a weird tail wheel type recently and first landing had no idea how it would behave, and I didn’t trust ASI. As such I added 10 knots, rounded out to level above runway and then tweaked it up as it sank and did a perfect landing.
Also worked when I came in a bit too fast following a total oil pressure loss landing on a eurofox.
To do a short field landing at back-of-drag curve type speeds, it all blurs into one, but I believe if you have the “fly level before flare” in your mind you should never balloon. I’ve taught it to a few students and watched them do some very versatile landings quite quickly once they get it.
The obvious cause of balloon and/or hard landing is primarily from “let’s aim to land on painted numbers”
Once the pilot masters his “fully held off landings” with free boundary condition and touchdown points (no extra other than depth perception, getting used to attitude and very long runway, you don’t even need an ASI), he can go back to “practice spot landings” with glide path planted on the painted numbers or aiming point (requires tight speed & energy control and most importantly a very reliable ASI and some well calculated Vref for config, weight and wind)
Starting “spot short landing” with 0h on new type is not easy 
RobertL18C wrote:
The best technique for conventional aircraft is a tail low wheel landing. You don’t carry more speed, just the same 1.2 x Vs you would normally use on final, but on round out (the phase where you change the aiming point and hold the aircraft in a parallel attitude to the surface) you maintain a little power to manage the sink and gently flare. You should touch down at only a few mph faster than a three point, and then you would relax, or slightly move the column forward to reduce AoA and remain landed. As speed decelerates you can move the column forward again, until you then gently pull back to ‘land’ or ‘fly on’ the tailwheel.
This describes what I do and teach. I only minimally rely on power for glidepath control, as engines have been known to fail. I prefer more steep approaches, modulated by slipping if needed. If you’ve crossed the fence at the correct speed and height, and close the throttle, pausing over the runway (as opposed to forcing it on in a rush) is helpful. I pause, and allow the plane to slow in the wheel landing attitude I choose, it will settle on, as it slows with no further attitude change. Once the mains are on, I gently move the stick forward to lift the tail. Occasionally, the tail wheel might slightly touch, so I will lift it off. I continue to move the stick forward to the stop, until the tailwheel settles on gently. I do not apply the brakes until the tailwheel is on, and I can apply nose up elevator to oppose the braking forces.
I choose to wheel land all taildraggers I fly. I reserve a three point landing for an emergency procedure into a very small landing area if ever needed.
