I am trying to understand the efficiency of the following permanent magnet synchronous motor Motor Datasheet (note we are using the medium voltage model), and its relationship with the input voltages and currents. The application is an electric racecar.

I made the following graph from the data given in the sheet. The blue line is directly taken from the datasheet graph, and the red line is generated from the torque constant on the first page of the datasheet. I understand that the difference between these lines is due to magnetic saturation of the motor.   The graph on the right is what I understood to be the efficiency (the graph data / the ideal torque). At 120Nm of torque (~300A) the efficiency appears to be 75%. However, when looking at the datasheet's efficiency map, the efficiency at 120Nm could be as high as 94%. Given the following equations, that does not make sense, unless the voltage vs rotational frequency graph curves in the opposite direction, enough to compensate.

Pin = Vin * Iin Pout = Wm(Vin) * Tq(Iin) Efficiency = Pout / Pin = torqueEfficiency * speedEfficiency

Additionally, I was wondering how the efficiency could be mathematically defined as a function of the input harmonics (we are wondering how to determine the ideal switching frequency of our inverter).

Am I fundamentally misunderstanding electric machines?

I am trying to understand the efficiency of the following permanent magnet synchronous motor Motor Datasheet (note we are using the medium voltage model), and its relationship with the input voltages and currents. The application is an electric racecar.

I made the following graph from the data given in the sheet. The blue line is directly taken from the datasheet graph, and the red line is generated from the torque constant on the first page of the datasheet. I understand that the difference between these lines is due to magnetic saturation of the motor. 

The graph on the right is what I understood to be the efficiency (the graph data / the ideal torque). At 120Nm of torque (~300A) the efficiency appears to be 75%. However, when looking at the datasheet's efficiency map, the efficiency at 120Nm could be as high as 94%.

Given the following equations, that does not make sense, unless the voltage vs rotational frequency graph curves in the opposite direction, enough to compensate.

Pin = Vin * Iin 
Pout = Wm(Vin) * Tq(Iin) 
Efficiency = Pout / Pin = torqueEfficiency * speedEfficiency

Additionally, I was wondering how the efficiency could be mathematically defined as a function of the input harmonics (we are wondering how to determine the ideal switching frequency of our inverter).

Am I fundamentally misunderstanding electric machines?