Looking at that graph makes me wonder about the stories about running engines too lean resulting in engine failures after time. On the graph running it lean of peak reduces egt, cht , and pressure, in saying that, it must be detonation that causes the p's and t's resulting in engine failures
Detonation occurs at peak cylinder pressures (its the excess pressure in the cylinder which is causing the fuel to burn too fast). So detonation is going to occur at peak TCO in the diagram above, which is around 50 degrees rich of peak EGT. That is why its the worst place to run your engine. Detonation will also only occur when cylinder temps are also high, or obviously your engine would kill itself all too often as you reduce power/mixture through this region. Looking at the above diagram.. leaner than peak EGT means a
cooler engine, and a cooler engine is a happy engine.
BUT
With stock standard fuel injectors, the fuel distribution across cylinders is not usual uniform. Running lean of peak EGT will result in an engine that does not run smoothly, as each cylinder is not producing equal power. The diagram above shows that power (HP) drops sharply lean of peak, so this unequal distribution leads to rougher running, than running rich of peak (ROP). With newer fuel injectors (like GAMI's) the fuel distribution is almost exactly right, so your engine will run smoothly no matter what the mixture is. With this uneven fuel distribution, you may have set 100 degrees ROP, but without an EGT per cylinder, it is hard to know what all is doing, so its wise at high power settings to set 100-150 rich of peak rather than 50.
Now, splatgothebugs said
As far as reducing the power as soon as you are airbourne, remember the most common time for an engine failure to occur is on the first power reduction. I would rather be at 500' plus prior to that happening.
This is also another myth. When is the most likely time for an engine failure? Its not takeoff, its in cruise. This is becuase most time is spent in cruising flight. Looking at real NTSB data on engine failures it has been shown that an engine can fail at any time, not on the first power reduction. Also the most likely cause of an engine failure is the absence of fuel (due to selection of empty tank or fuel exhaustion). So make sure you have juice in the correct tank and you cut out around 75% of the chance of an engine failure.
Bevan..