@Robot_Arm: That “always be able to make the runway in a failure” idea was certainly the traditional 1930s way to teach aviation. And given the engine reliability of the day, not bad advice then. IMO it’s rather obsolete today for lightplanes, and fully inapplicable to multi-engine or fast planes. Let’s explore it a bit though.
IMO there’s a lot of old wives’ tales in GA training. Ideas are passed down from generation to generation without really knowing why they were ever invented, and therefore there’s no ability to think through whether they’re still applicable. They may be solving a problem that’s long since been solved by other better means. I think this is (mostly) one of those.
The idea does raise the question of “if I’m supposed to be in gliding range of a runway in the traffic pattern, how come it’s OK to NOT be in gliding range of a runway for all that time in cruise?” Hmm. No good answer there.
Some of the logic is that in cruise you should be high enough to have sufficient glide time to select a tolerable off-field landing spot, remedy the failure to the degree you can (change fuel tanks, fiddle with mags, mixture, etc), and also radio a call for rescue. At low AGL altitudes, there isn’t time for much except aim and flare, so best to make that a soluble problem by having a runway within your engine out glide range. Which really turns the emphasis of the rule on its head to say:
Always have enough AGL altitude so there’s a suitable landing spot within your current glide range. And always keep at least one selected, hopscotching along from takeoff to landing, and especially in cruise where you spend most of your time. When unavoidably at low altitude, that spot should be an airport if possible.
Now let’s talk just in the pattern …
Back in the day of Stearmans and Jennys, airports were square empty fields of grass or dirt ~1/2 mile on a side (hence the term “airfield”), landing rolls were a few hundred feet, and you were the only airplane in the sky. ATC had not yet been invented. Flying a box pattern around the outside of the field where at any moment you could nose down and turn towards the closest edge of the field and still make it was relatively easy. Nowadays in the pattern at a tower controlled GA field where you’re at 800AGL 5 miles from the runway turning downwind to base? Not gonna happen. Engine quits, you’re landing in the surrounding suburb or forest or whatever.
IMO this is why the maxim is obsolete. It simply can’t be implemented in most regimes of most flights. So the emphasis should be elsewhere. Such as how to select the least bad off-airport landing spot, having one of those always in mind, not running out of fuel, etc.
But that’s not addressing your original and completely legit objection:
If I maneuver to always be able to make the runway without an engine, then how do I avoid landing long with an engine?
The answer is simple: drag management.
An typical lightplane piston engine at idle at traffic pattern speeds produces net drag. The same engine failed produces more drag. It’s a bigger increment than you might expect, but it’s not night and day.
So in an old fashioned flapless airplane, the kind the advice was invented for, you …
Fly a tightish steepish pattern where you can turn towards the runway to shorten your path at any point. If the engine quits, turn directly towards the end-ish of the runway, planning to turn again to align with the runwaywhen you get closer. Now you’re probably high, so use a slip to steepen your descent until your non-slipped descent angle takes you to someplace well down the runway. Now stop slipping and glide 'er in. You aim maybe 1/3rd down the runway so it you misjudge that, you have room to come up short. IOW If you land shorter than you’d estimated, you’re still on the runway. And all this is for airplanes that need very little runway length to stop once on the ground.
In the traditional 1940s-1950s airport that had several runways arranged like a asterisk or pickup sticks, when you turn directly towards the field you may be well aligned with a different runway than you’d planned to use while power-on. That makes it even easier because now you don’t need a big turn at the bottom to align. You’re probably real high now. Jump on that slip until things look more normal, then finish the glide to this other runway.
If your engine is still running when you turn final for your original runway, now the problem is different. Your path to the runway is now as short as it can be. So plan to get there high / steep and use a slip to increase your drag enough to descend to impact at your aimpoint. Which starts out maybe 1/3rd down the runway and as you get closer, you move that back towards the threshold. If the engine quits, kill the slip and suddenly your drag is less and your glide is longer and you still make the runway.
Easy peasy. As long as you’re planning all the way around. Which again is IMO a bit silly in 2023. Or at least is often unattainable in 2023.
For one of those modern new-fangled airplanes equipped with flaps (and maybe even a sissy nosewheel 
), the same idea as slipping can be accomplished with flaps. Take a look at the POH for any lightplane and you’ll see that while flaps reduce your stall speed, they reduce your glide ratio even more. If you’re on final with, say, half flaps and the engine quits, poke the nose down to pick up the 5 or maybe 10 knots/mph you need as stall margin, suck up the flaps and suddenly you can still reach the runway. If you’re already at full flaps, reducing flaps to half makes a much larger change in drag than it does in stall speed.
Said another way, the engine power you’re using on a normal flapped approach is at least partly being “wasted” just overcoming the flap drag. If you lose the one, then lose the other and now you’re back in balance and happiness ensues.
[aside]
In jets the problem is different, but a LOT of the reason for large landing flap angles is to create a LOT of drag to be overcome by a LOT of engine thrust. Not for the engine failure scenario, but for the sudden go-around scenario with engines that are inherently slow to accelerate.
With jets, faster the RPM, the faster the engine accelerates; it’s very non-linear, more like exponential. To boot, thrust output is also exponential on RPM, where the last 5% of RPM is 30% of the thrust. Back in the day, idle to full thrust was maybe 10 seconds, while medium power to full thrust was maybe 4 seconds. So approaching at medium power gives a safe quick go-around, and the flaps are sized so one can descend and hold speed with the power way up at “medium”. Newer engines accelerate quicker, but it’s still nothing like an ICE car or plane with negligible spin-up time.
Our go-around procedure is to shove up the power and simultaneously reduce flaps a bunch. The drag reduction happens much faster than the thrust increase does. In either case, that gets us going upward immediately while everything else catches up.
[aside 2]
In the F-16 we practiced engine failure approaches. You’d aim to arrive over the middle of the runway at 10,000 and 300 knots which was a comfortable glide speed. Assuming you got there, you’d start an immediate rather tight turn to a short downwind, which turn would scrub off speed to maybe 250, then hold that speed. Then another 180 turn to final, all the while aiming about 1/3rd of the way down the runway. Somewhere along the way you’d deploy gear which gave you landing flaps at the same time.
As the gear/flap drag came in and you steepened the glide to hold the 250, your impact point would move to about 1/2 mile short of the runway. You’d be jockeying how tightly you turned and how long on downwind to achieve that result. Once you were rolled out on final re-aim for partway down the runway and the speed starts falling off from 250 to the 150ish you want over the threshold. Then assuming your energy management was good, you run out of altitude and over the runway and plop it on. Better to be long than short. The baby wheel brakes aren’t great at stopping you, but the tailhook sure is. Conversely, the “fix” for too short was “Eject! Eject! Eject!”.