If we assume the editor instance ran at 60 FPS, the tick would have applied 60 times per second. So, the pipes moved -15 * 60.
Say your game outside the editor runs at roughly 180 FPS, which means the pipes would have moved at -15 * 180.
To counteract this effect, you can take into account delta time. Delta time tells you how much time has passed since last tick. So, if a game runs at steady 60 FPS, the delta time value, in seconds, would be 1/60, or 0.0166666.
If you only multiply the -15 with Delta Time, this will make the movement of the pipes framerate independent (for the most part anyway), but it will also slow the movement of the pipes down by a factor of 60. So, if you playtested the game at around 60 FPS, the approach is to take the -15 and multiply it by 60. In the end you would move the pipes by -900 * Delta Time for every tick.
So now, if the game runs at 60 FPS, it will move the pipes by (-900 * 0.01666) or -15 every tick, for a total movement of -900 per second, as ticks occur 60 times per second.
If your game runs at 180 FPS, it will move the pipes by (-900 * 0.00555) or -5 every tick, for a total movement of -900 per second, as ticks occur 180 times per second.
Don’t try to plug wires into each other randomly, your first graph makes way more sense except it doesn’t take framerate into account, in the second, you’d have to ask yourself what you are even multiplying.
Edit: Also don’t use delays in tick to destroy the actor. Either check the Y position and see if the value is too low, then destroy them, or use some sort of a trigger volume to destroy the pipes once the pipes touch it.