John's right, It's about the 1st law of thermodynamics,
delta U = Q - W ;
(change internal energy) = heat transfer (positive if comes to the system - work (if work has a negative in front means it is making something move out of the system, i.e., a piston rod).
Let us take a piston in a cylinder. The cylinder is the system and the water and heads are the surroundings. The work extracted from the system is Pressure X Volume final - P X V initial. During expansion, as the piston goes from the top of the cylinder to the bottom, Work is positive.
In the equation, delta U = Q - W
The more work you get out, the better. Solving for W,
W = delta U + Q
If heat is lost to the surroundings, Q will be - and thus will reduce W (work). That is not good. The less work, the less power (horsepower) which is work per time. If no heat is lost, Q = 0 and thus the equations reduces to :
W = delta U.
Some might ask, why not raise Q, make the engine real real hot? Then you can run into other problems, i.e, overheating block, vaporizing liquid, premature ignition, expensive components, etc, etc, etc.
So, you don't want to much Q gain (overheating) or too little Q (heat loss). As John stated, under adiabatic conditions (no heat transfer), Q becomes 0.
Under adiabatic conditions, as John mentioned, the majority of the initial energy of the piston will be transferred to your drive shaft and wheels more than being transferred to additonal unwanted heating of the engine block. Would you rather have more power going to the wheels or a hotter engine?
In a way think of it as you were running a marathon. Would you rather run a marathon in shorts in Vancouver in January, Miami in July, or New York City in October? In Vancouver, your body won't be able to keep up and will freeze, in Miami, you'll pass out, and it NYC, you'll keep chuggin away..... as long as you trained and ate right
'89 420 SEL
'90 300 SEL
'68 Olds 88 Convertible
'84 300 SD (sold it)